JP2003014433A - Shape measuring apparatus, control device for shape measuring apparatus, and control program for shape measuring apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To surely measure the shape of a measuring surface by use of an interferometer even if the measuring surface has an optical characteristic disper sion. SOLUTION: This measuring apparatus comprises the interferometer for irradiating light to the measuring surface and a prescribed reference surface and detecting the interference fringes formed by the light from these surfaces by an image pickup element; and a control device for drive-controlling the interferometer. The control device comprises a measurement part for setting the luminous quantity of the light irradiated to the measuring surface to each value of a predetermined value group, and driving the interferometer when the luminous quantity is set to each value to perform an interference measurement two or more times; and a calculation part for acquiring the shape information of the measuring surface based on a one showing a proper value among the pixel outputs of the image pickup element in each interference measurement. Accordingly, even if the measuring surface has a dispersion of optical properties, the shape information of the measuring surface can be surely obtained.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a shape measuring apparatus for measuring the shape (height distribution) of a surface to be measured such as the surface of a semiconductor wafer, a controller for the shape measuring apparatus, and a control program for the shape measuring apparatus.

[0002]

2. Description of the Related Art Generally, an LSI device is manufactured by going through a process including film formation, lithography, etching and the like on a semiconductor wafer many times. In the lithographic process, if the flatness of the semiconductor wafer is poor, the focusing accuracy of the optical exposure device used for transferring the LSI pattern becomes difficult, so that the transfer accuracy or transfer accuracy decreases. Even if it does not, the time required for focusing may be prolonged.

Therefore, semiconductor wafers are inspected for flatness before shipment, that is, in the state of bare wafers before the start of all processes, and only products whose flatness is within the standard are shipped. Further, at the time of shipment, some evaluation value indicating the flatness may be added as product information. Incidentally, at present, the capacitance sensor is mainly used for the inspection of the flatness.

[0004]

By the way, in recent years, in order to cope with the miniaturization of LSI patterns, the exposure wavelength of the light exposure apparatus is shortened and the projection lens of the light exposure apparatus is increased in NA. There is a tendency. However, this shortening of the wavelength and the increase in the NA are accompanied by a decrease in the depth of focus of the projection lens, so that the precision required for the focusing of the optical exposure apparatus becomes more severe.

Therefore, in recent years, there has been a demand for finer and more precise measurement of the flatness of a semiconductor wafer, and the demand for the measurement tends to increase. Here, the demand for measurement increases because the surface shape of the semiconductor wafer changes with each process, so the need for measuring the flatness is not limited to the state of the bare wafer before the process. , Is also generated in each state of the process wafer during the process.

However, since the process wafer has a pattern formed on its surface, it cannot be measured by the above electrostatic sensor. Even if it is possible to measure, the electrostatic sensor, in principle, can only measure the surface to be measured roughly in spite of the long time required for the measurement. I can't. It is possible to use an AFM (atomic force microscope) for measurement of the process wafer. However, although the AFM can measure finely and highly accurately in principle, the time required for measurement is very long, so It is not so preferable because it prolongs the time.

Therefore, in recent years, an interferometer capable of measuring the surface shape of the surface to be measured minutely and highly accurately and in a short time has attracted attention. The interferometer uses a two-dimensional detector (imaging element) to detect the interference fringes formed by the reflected light from the entire surface to be measured.
Such a measurement is possible because the detection is performed at once by. In fact, in measuring the flatness of a bare wafer, it has already been considered to apply an interferometer instead of the capacitance sensor.

However, since the optical characteristics of the surface of the process wafer on which the pattern is formed are not uniform (there are variations), it is difficult to directly apply the interferometer to the flatness measurement of the process wafer. Is. That is, since various materials such as metals, semiconductors, and dielectrics can be used for each pattern, there are large variations in optical characteristics (especially reflectance), and the difference between the lowest region and the highest region is several times greater. It may also extend. At this time, if the internal setting of the interferometer is adjusted to the metal (high reflectance region), the signal indicating the height of the semiconductor or the dielectric (low reflectance region) effectively uses the dynamic range of the detector. S / N becomes extremely bad without being able to do it. Conversely, if you set the internal settings of the interferometer to semiconductors or dielectrics (low reflectivity regions), the signal indicating the height of the metal (high reflectivity region) will saturate the detector and invalidate it. It only shows the value.

In view of this, the present invention uses an interferometer and is capable of reliably measuring the surface shape even if there are variations in the optical characteristics within the surface to be measured.
An object of the present invention is to provide a control device for a shape measuring device and a control program for the shape measuring device.

[0010]

A shape measuring apparatus according to any one of claims 1 to 3 irradiates light on a surface to be measured and a predetermined reference surface, and at the same time, the light from those surfaces is emitted. A shape measuring apparatus comprising an interferometer for detecting interference fringes formed by an imaging device, and a control device for driving and controlling the interferometer, wherein the control device controls a light amount of light irradiated on the surface to be measured. , A measuring unit that sets each of the predetermined value groups, and drives the interferometer to perform a plurality of interferometric measurements when the light amount is set to each value, and the measuring unit at the time of each interferometric measurement. It has a calculating part which acquires the shape information of the said to-be-measured surface based on what shows an appropriate value among the pixel outputs of an image sensor.

Therefore, even if there are variations in the optical properties of the surface to be measured, it is possible to reliably obtain the shape information of the surface to be measured. Further, the control device in the shape measuring apparatus according to claim 2 or 3, prior to the operation of the measuring section, sets the light amount of the light with which the surface to be measured is irradiated to a predetermined value, and at that time. A test measurement unit that refers to the pixel output of the image sensor, and a value determination that determines the value group to be set by the measurement unit according to the variation in the optical characteristics in the measured surface indicated by the pixel output. And a department.

Further, the control device in the shape measuring apparatus according to claim 3 has a storage section for storing a plurality of kinds of value groups in association with the kind of the surface to be measured, and the operation of the measuring section. Prior to the above, there is a value deciding unit that decides a value group to be set by the measuring unit into a value group stored in the storage unit in association with the type, according to the type of the surface to be measured. It is characterized by

The control device of the shape measuring apparatus according to any one of claims 4 to 6 irradiates the surface to be measured and a predetermined reference surface with light, and the interference fringes formed by the light from those surfaces. Is a control device of a shape measuring device applied to a shape measuring device provided with an interferometer for detecting by means of an image pickup device, wherein the amount of light applied to the surface to be measured is set to each of a predetermined value group. Along with the setting, when the light amount is set to each value, the measuring unit that drives the interferometer to perform the interferometric measurement a plurality of times, and the pixel output of the image sensor during each of the interferometric measurements And a calculation unit that acquires the shape information of the surface to be measured based on what indicates a value.

Therefore, even if there are variations in the optical properties of the surface to be measured, the shape information of the surface to be measured can be reliably obtained. Further, the control device of the shape measuring apparatus according to claim 5 or 6, sets the light amount of the light with which the surface to be measured is irradiated to a predetermined value prior to the operation of the measuring unit, and A test measurement unit that refers to the pixel output of the image sensor, and a value determination unit that determines the value group to be set by the measurement unit according to the variation in the optical characteristics in the measured surface indicated by the pixel output. Have and.

Further, the control device of the shape measuring apparatus according to claim 6 has a storage unit for storing a plurality of kinds of value groups in association with the kind of the surface to be measured, and the operation of the measuring unit. Prior to this, there is provided a value deciding unit for deciding a value group to be set by the measuring unit into a value group stored in the storage unit in association with the type according to the type of the surface to be measured. The control program of the shape measuring apparatus according to any one of claims 7 to 9, which irradiates a surface to be measured and a predetermined reference surface with light and forms an interference fringe formed by the light from those surfaces. Is a control program of a shape measuring device for operating a computer as a control device of a shape measuring device having an interferometer for detecting, the light quantity of light radiated to the surface to be measured, a predetermined value group Of the pixel output of the image sensor at the time of each interferometric measurement, and the measurement procedure of performing the interferometric measurement a plurality of times by driving the interferometer when the light amount is set to each value. , And a calculation procedure for acquiring the shape information of the surface to be measured based on a value indicating an appropriate value.

Therefore, even if there are variations in the optical properties of the surface to be measured, it is possible to reliably obtain the shape information of the surface to be measured. Further, the control program of the shape measuring apparatus according to claim 8 or 9, in the control program of the shape measuring apparatus according to claim 7, irradiates the surface to be measured before performing the measurement procedure. While setting the light amount of light to a predetermined value, the test measurement procedure referring to the pixel output of the image sensor at that time, and the measurement according to the variation in the optical characteristics in the measured surface indicated by the pixel output. A value determination procedure for determining the value group to be set in the procedure.

Further, the control program of the shape measuring apparatus according to claim 9 is the control program of the shape measuring apparatus according to claim 7 or 8, wherein the measurement surface of the measured surface is prior to the execution of the measuring procedure. A value determination procedure for determining a value group to be set in the measurement procedure into a value group associated in advance with the type according to the type.

[0018]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings.

[First Embodiment] The first embodiment of the present invention will be described with reference to FIGS. 1, 2, 3, 4, 5, 6, 7 and 8. (Configuration) FIG. 1 is a configuration diagram of a flatness measuring apparatus of the present embodiment (and second and third embodiments described later).

The flatness measuring apparatus of this embodiment comprises an interferometer apparatus 10 and a computer 30 for driving and controlling the interferometer apparatus 10. The surface to be measured 20a in this flatness measuring device
Is the surface of the semiconductor wafer 20. The semiconductor wafer 20 is a process wafer having a pattern formed on its surface. Although the Fizeau interferometer is applied to the interferometer device 10 described below, other interferometers such as the Twyman Green type can also be applied.

In the interferometer device 10, the light emitted from the light source 11 (hereinafter referred to as "measurement light") is adjusted in its light amount by the light amount setting device 12, and then the beam expander including the lenses 13A and 13B. It is converted into a parallel light flux by 13 and is incident on the Fizeau flat 15A. Part of the light that has entered the Fizeau flat 15A is reflected by the reference surface 15a of the Fizeau flat 15A, and the other part of the light passes through the reference surface 15a and enters the measured surface 20a of the semiconductor wafer 20.

The light reflected by the reference surface 15a (hereinafter referred to as "reference light") and the light reflected by the measured surface 20a (hereinafter referred to as "test light") are both included in the half mirror 14.
After being deflected to the side different from the light source 11 at
It is incident on the imaging unit 18 via 7. The image pickup device 18a in the image pickup unit 18 can pick up an interference fringe formed by the reference light and the test light.

Here, if the surface accuracy of the reference surface 15a of the Fizeau flat 15A is sufficiently high (if it is a flat surface of sufficient accuracy), the interference fringes have a shape (height distribution) of the surface 20a to be measured. (Note that even if the surface accuracy of the reference surface 15a is not sufficient, if the surface accuracy error is known, it is possible to correct it by calculation and extract only the height distribution of the measured surface 20a. is there).

Here, the light quantity setting device 12 includes the measured surface 2
The light quantity of the measurement light can be set to various values in order to obtain a high-contrast interference fringe regardless of the optical characteristics of 0a. The light amount setting device 12 includes a light amount adjusting filter 12A and a filter driving unit 12B that drives the light amount adjusting filter 12A, and sets the light amount of the measurement light under the instruction of the computer 30.
For example, the light amount adjustment filter 12A is an ND filter having a different density value depending on the location, and the filter driving unit 12
B is a mechanism for rotating the ND filter and a motor for driving the mechanism. In this case, the amount of measurement light is determined by the rotation position of the ND filter.

The fringe scan interferometry method is applied to the interferometer device 10 in order to obtain the height distribution of the surface 20a to be measured with higher accuracy. That is, the Fizeau flat 15A can be moved in the optical axis direction by the optical path difference varying device 15 over a distance of ½ or more of the wavelength of the light source 11.

The optical path difference varying device 15 is composed of, for example, a piezo element and a drive section for driving the piezo element, and can move the Fizeau flat 15A from a predetermined position at a predetermined speed with sufficient accuracy. According to this movement, the phase difference between the reference light and the test light gradually changes, so that it is possible to change (fringe scan) the brightness of each point of the interference fringes.

Further, the reproducibility of the movement pattern of the Fizeau flat 15A by the optical path difference varying device 15 and the reproducibility of the timing of the interference fringe imaging (described later) by the imaging section 18 are sufficiently high (that is, fringe scanning Reproducibility is high enough). Next, the computer 30 is, for example, a general-purpose computer in which a dedicated control board is mounted on the interferometer device 10, and based on a program installed in advance, the light source 11 and the light amount setting device 12 in the interferometer device 10. By giving each instruction to the imaging unit 18 and the optical path difference varying device 15, "interference measurement" is performed to obtain information indicating interference fringes (details will be described later).

Further, the computer 30 carries out a "fringe analysis" for analyzing the information obtained by this interferometry based on the program (the fringe analysis allows the measured surface 2 to be measured).
The height distribution of 0a and the flatness are acquired. Details will be described later. ). In the interferometer device 10, the semiconductor wafer 20 is preferably held on the same wafer holder 16 as used in the light exposure device. This is because the semiconductor wafer 20 is placed under the same conditions as when used.

FIG. 2 shows the surface 20 to be measured of the semiconductor wafer 20.
It is a figure explaining a. As shown in FIG. 2A, the surface 20a to be measured has a plurality of drawing areas (in the figure, one drawing area corresponds to one device). , A plurality of types of patterns Pa and P having different optical characteristics from each other as shown in FIG.
It is formed by mixing b, Pc, and Pd. It should be noted that, in each drawing area, even a region in which no pattern is formed can be regarded as a pattern having optical characteristics different from other regions, and therefore, is denoted by “Pe” and hereinafter referred to as “pattern Pe”.

Hereinafter, the patterns Pa, Pb, Pc, Pd,
It is assumed that the magnitude relationships among the reflectances Ra, Rb, Rc, Rd, and Re of Pe are Ra>Rb>Rc>Rd> Re. (Operation) FIG. 3 shows a computer 30 according to this embodiment.
3 is an operation flowchart of FIG. The computer 30 of the present embodiment sets the light amount of the measurement light to L via the light amount setting device 12.
Set to a, Lb, Lc, Ld, and Le,
Interference measurement is performed under each of these light quantities (steps S1 to S4).

Hereinafter, the interference measurement under the light amount La, the light amount L
interference measurement under b, interference measurement under the light quantity Lc, interference measurement under the light quantity Ld, interference measurement under the light quantity Le,
"Measurement of light intensity La", "Measurement of light intensity Lb",
"Measurement of light intensity Lc", "Measurement of light intensity Ld", "Light intensity Le"
Measurement of. In each measurement, the computer 30 gives an instruction to the optical path difference varying device 15 to perform a fringe scan, while the image capturing unit 18 (image capturing unit 18) is used.
The image pickup device 18a) is driven to image the interference fringes a plurality of times.

For example, based on a certain method of fringe scanning, the Fizeau flat 15A is moved by a distance corresponding to 1/2 of the wavelength of the light source 11 and is imaged five times at equal intervals. 4, FIG. 5, FIG. 6, FIG. 7, and FIG.
It is a conceptual diagram explaining measurement of light quantity La, measurement of light quantity Lb, measurement of light quantity Lc, measurement of light quantity Ld, and measurement of light quantity Le.

In each of these figures, the pictures on the left side show various patterns Pa, Pb, Pc, Pd on the semiconductor wafer 20.
(Concept) is shown, and the graph on the right shows the image sensor 18
Of the pixel output of a, each pattern Pa, Pb, Pc, Pd
Each pixel output (concept) corresponding to each of the above is shown.

During the fringe scanning, the brightness and darkness of each point of the interference fringe changes sinusoidally with time, so that the output of each pixel shown on the right side of each drawing also changes with time in a sinusoidal manner (after passing through light and dark). Return to the original value). Further, the phase of each sine wave corresponds to the height of each point (however, each phase in the figure has no special meaning). First, in the measurement of the light amount La, as shown in FIG. 4, the level of the pixel output corresponding to the pattern Pa is maintained at an appropriate level.

Further, in the measurement of the light quantity Lb, as shown in FIG. 5, the level of the pixel output corresponding to the pattern Pb is maintained at an appropriate level. Further, in the measurement of the light amount Lc, as shown in FIG. 6, the level of the pixel output corresponding to the pattern Pc is maintained at an appropriate level. Moreover, in the measurement of the light amount Ld,
As shown in FIG. 7, the pixel output level corresponding to the pattern Pd is maintained at an appropriate level.

In the measurement of the light quantity Le, the pixel output level corresponding to the pattern Pe is maintained at an appropriate level as shown in FIG. Here, the appropriate level is a level that does not reach the saturation level of the image sensor 18a even when the pixel output takes the maximum value, and that the dynamic range of the image sensor 18a is sufficiently effectively used (for example, , Above 80% of saturation level). That is, the pixel output of the appropriate level has a sufficiently high S / N and shows a highly reliable value.

As a result, in the present embodiment, in the measurement of the light amount La (FIG. 4), the pixel corresponding to the pattern Pa among the pixels of the image pickup device 18a is an effective pixel (highly reliable pixel).
Therefore, in the measurement of the light amount Lb (FIG. 5), the image sensor 18a
Of the pixels of the image sensor 18a, the pixel corresponding to the pattern Pb becomes the effective pixel, and in the measurement of the light amount Lc (FIG. 6), the pixel corresponding to the pattern Pc of the pixels of the image sensor 18a becomes the effective pixel and the light amount Ld is measured (see FIG. ), The pixel corresponding to the pattern Pd among the pixels of the image sensor 18a becomes the effective pixel, and the pixel corresponding to the pattern Pe among the pixels of the image sensor 18a becomes the effective pixel in the measurement of the light amount Le (FIG. 8).

The light amounts La, Lb, L
For each value of c, Ld, Le, each pattern Pa, P
b, Pc, Pd, Pe optical characteristics (for example, the reflectances Ra, Rb of the patterns Pa, Pb, Pc, Pd, Pe).
Rc, Rd, Re). Also,
These value groups (La, Lb, Lc, Ld, Le) can be stored in the computer 30 before the operation flowchart of FIG. 3 is started.

Incidentally, as described above, the pattern Pa,
The reflectances Ra, Rb of Pb, Pc, Pd, and Pe, respectively.
The magnitude relationship between Rc, Rd, and Re is Ra>Rb>Rc> Rd.
If> Re, the light amounts La, Lb, Lc, Ld, L
The magnitude relation of e is Le>Ld>Lc>Lb> La. It is also possible for the computer 30 to store which pixel is the effective pixel in each measurement before the operation flowchart of FIG. 3 is started.

Incidentally, the effective pixel of each measurement is the measured surface 2
0a pattern arrangement (FIG. 2) and measured surface 20
It is determined by the correspondence between each point on a and each pixel of the image sensor 18a. Further, even if the computer 30 does not store the data, the computer 30 can also distinguish the effective pixels based on the output of each pixel.

Returning to FIG. 3, in the subsequent step S5, the computer 30 converts the pixel output of the effective pixel (that is, the pixel output corresponding to the pattern Pa) in the measurement of the light amount La into the phase distribution, and then the height of the pattern Pa. It is converted into data indicating the distribution (hereinafter referred to as “height distribution data”). Further, the computer 30 in step S5
Similarly, the pixel output of the effective pixel in the measurement of the light amount Lb,
Pixel output of effective pixel and light amount Ld in measurement of light amount Lc
The pixel output of the effective pixel in the measurement of 1 and the pixel output of the effective pixel in the measurement of the light amount Le are also converted to the phase distribution and then to the height distribution data.

The height distribution data obtained in step S5 shows only a partial area of the surface 20a to be measured, so that the computer 30 synthesizes these height distribution data in step S6. Then, height distribution data for the entire area of the surface 20a to be measured is acquired. Here, the conversion in step S5 will be briefly described.

First, the height distribution of the surface 20a to be measured appears in the phase distribution of the interference fringes. In addition, the brightness of each point of the interference fringe is 1
The distance traveled by the Fizeau flat 15A when the period changes is half the wavelength of the light source 11 (for example, the light source 1
In the case of using a helium neon laser having a wavelength of 633 nm for 1), it becomes 317 nm). The conversion is performed by a conversion formula based on these relationships and the system applied to the fringe scan (that is, the movement pattern of the Fizeau flat 15A and the imaging timing defined by the system).

As described above, since the reproducibility of the fringe scan is sufficiently high, the same conversion formula can be used for each conversion in step S5. Also,
This conversion formula does not depend on the value of the amplitude of light and dark of interference fringes (that is, the amplitude of pixel output). Therefore, even if the amplitude of the pixel output of the effective pixel is slightly deviated between the measurements, the value of the height distribution data obtained by the conversion is not affected.

As a fringe scan method,
Although the one in which the Fizeau flat 15A is imaged five times at equal intervals while moving the Fizeau flat 15A by a distance of ½ of the wavelength of the light source 11 has been mentioned, any known other method may be applied. . In the following step S7, the computer 30
Extracts an evaluation value representing the flatness of the measured surface 20a from the height distribution data acquired in step S5.

As the evaluation value, for example, the measured surface 20a
The maximum displacement width GFLR from the reference surface, the deviation GFLD from the reference surface, and the like for the entire area of the are considered. Further, these evaluation values may be calculated not only for the entire area of the surface 20a to be measured but also individually for each drawing area. That is, the maximum displacement width SFQR from the reference surface, the deviation SFQD from the reference surface, and the like for each drawing area may be obtained.

As described above, in this embodiment, the interferometer device 10 is
This is used for measuring the shape of the surface 20a to be measured (and by extension measuring the flatness) with variations in optical characteristics. At this time, the measurement is performed under various predetermined light amounts, and the results of the respective measurements are combined, so that the measurement of the surface 20a to be measured is possible. Further, the present embodiment can be realized by a known flatness measuring device (the known interferometer device 10 and a general-purpose computer) by only partially changing the software without changing the hardware. That is, this embodiment can be realized at low cost.

In this embodiment, the number of values of the value group (La to Le) set for the quantity of measurement light is the same as the number of types of patterns (Pa to Pe) on the surface 20a to be measured. , Even if the patterns have different optical characteristics, if the difference in optical characteristics between them is small, it is possible to sufficiently measure even with the same light amount. You can reduce the number of values.

Further, in the present embodiment, in steps S5 and S6, the pixel output between the respective patterns is combined after conversion into height distribution data, but it is carried out before conversion into height distribution data. I don't mind. [Second Embodiment] A second embodiment of the present invention will be described with reference to FIGS. 1, 2, 9 and 10. Only the differences from the first embodiment will be described below.

(Structure) As shown in FIG. 1, the flatness measuring apparatus of this embodiment is the same as the flatness measuring apparatus of the first embodiment except that a computer 40 is provided instead of the computer 30. Like the computer 30, the computer 40 is a general-purpose computer in which a dedicated control board is mounted on the interferometer apparatus 10, but the contents of the installed program are partially different from the computer 30.

(Operation) FIG. 9 is an operation flowchart of the computer 40 of this embodiment. Computer 40
Executes the operation flowchart shown in FIG. 3 similarly to the computer 30. However, the computer 40
By executing the operation flowchart shown in FIG. 9 before the execution, the steps S1 to S in FIG.
4 is different from the computer 30 in that the value group of the light quantity to be set for the measurement light in 4 is automatically determined according to the variation in the optical characteristics of the measured surface 20a.

In making this decision, first, the computer 4
0 sets the amount of measurement light to a predetermined value (step S
21). This predetermined value is at least a value that does not saturate the image sensor 18a with reflected light from any region of the surface 20a to be measured. Moreover, this predetermined value is a value that does not saturate the image sensor 18a even when the fringe scan is performed.

In this state, the computer 40 drives the image pickup section 18 to pick up an image of interference fringes at the time of fringe scanning (step S22), and refers to the histogram of the amplitude output of the image pickup element 18a at this time (step S23). . FIG. 10A is a histogram of the amplitude output of the image sensor 18a. Here, the distribution of the amplitude output of the image sensor 18a when the amount of the measurement light is constant reflects the distribution of the optical characteristics of the surface 20a to be measured.

Therefore, if the patterns Pa, Pb, Pc, Pd, and Pe shown in FIG. 2 are formed on the surface 20a to be measured, in this histogram, peaks pa, pb, pc, p
d and pe are generated. Each peak pa, pb, pc, pd,
Each amplitude output Ia, Ib, Ic, Id when pe is taken,
Ie are patterns Pa, Pb, Pc, Pd, and P, respectively.
optical characteristics of e (in many cases, reflectances Ra, Rb, R
c, Rd, Re) are shown.

Therefore, the computer 40 causes the amplitude outputs Ia, Ib, Ic, Id, Ie indicated by the histogram to be obtained.
In accordance with the above, the value groups La, Lb, Lc, L to be set for the light amount of the measurement light in steps S1 to S4 of FIG.
d and Le are determined. Then, the determined value groups La and L
b, Lc, Ld, and Le are stored in a predetermined storage unit (hard disk or the like) in the computer 40 as information for executing the operation flowchart of FIG. 3 (step S24).

As described above, according to this embodiment, the surface to be measured 20 is measured.
Even if the distribution of the optical characteristics of a is unknown, the value group L necessary for measuring its shape (and thus flatness)
a, Lb, Lc, Ld, and Le are automatically determined.
In the present embodiment, the number of values in the value group (La to Le) is
Although the number of types of patterns (Pa to Pe) on the surface to be measured 20a is the same, even if the patterns have different optical characteristics, if the difference in optical characteristics between them is small, the same light amount is obtained. Since it is possible to perform sufficient measurement, the number of values may be reduced when there is a pattern of such a relationship.

That is, the computer 40, for example,
As shown in FIG. 10B, in the histogram, 2
If the two peaks pg and pf are close enough,
Two values Lg depending on the amplitude outputs Ig and If,
Instead of determining Lf, only one value Lgf may be determined.

The above calculation for determining the value group from each amplitude output corresponding to each peak is performed by making a sharp distinction in which range each amplitude output falls.
It can be done relatively easily. [Third Embodiment] A third embodiment of the present invention will be described with reference to FIGS.
An embodiment will be described. Only the differences from the first embodiment will be described below.

(Structure) As shown in FIG. 1, the flatness measuring apparatus of this embodiment is the same as the flatness measuring apparatus of the first embodiment, except that a computer 50 is provided instead of the computer 30. Like the computer 30, the computer 50 is a general-purpose computer in which a dedicated control board is mounted on the interferometer apparatus 10, but the contents of the installed program are partially different from the computer 30.

(Operation) Computer 50 of this embodiment
Assumes various process wafers determined in advance as measurement targets. Since the quality of process wafers is small, if the value group of light quantity to be set is determined when measuring a certain type of process wafer, use the same value group when measuring the same type of process wafer after that. Is possible.

Therefore, the computer 50 of the present embodiment.
Stores a type of process wafer and a value group adapted to the type in association with each other. That is, the computer 50 stores the correspondence table of the types of process wafers and the group of values in the internal storage unit (hard disk or the like). Then, the computer 50 causes the operator to input the type of the process wafer via the display device 50a, the input device 50b, or the like, and performs the measurement using the value group previously associated with the type.

Therefore, even if the process wafer to be measured is exchanged for another type, the operator can immediately start proper measurement just by inputting the type into the computer 50. Note that FIG. 11 is an example of a correspondence table stored in the computer 50 of the present embodiment.

Generally, the type of process wafer is determined by the type of device manufactured therein and the number of processes (process stage). According to this correspondence table, the computer 50 can determine the value group according to the type of device and the number of processes. That is, the operator
The computer 50 can recognize the type of process wafer by simply inputting the number of processes and the type of device.

By combining the above-described third embodiment with the above-mentioned second embodiment, the contents of the correspondence table can be automatically created by the computer. [Others] In the first embodiment, each light amount is set to each value according to the variation in the optical characteristics of the measured surface 20a, but the measurement can be performed regardless of the type of the measured surface 20a. Therefore, it is possible to measure a relatively large number of times while changing the light amount in a relatively fine step.

In this case, the measurement time is unnecessarily long for the measured surface 20a having a small variation, but it is not necessary to perform the processing (processing for determining the value group) of the second or third embodiment. However, in this case, the computer needs to discriminate the effective pixels in each measurement. Although a general-purpose computer is used in each of the above-described embodiments, a part or all of the above-described computer operation may be executed by a control device dedicated to interferometer device 10.

Further, in the above-described first or third embodiment, when the operator inputs the design value (pattern material etc.) of the process wafer, the computer automatically determines a group of values to be set as the light quantity at the time of measurement. It is also possible to configure so as to determine.

[0067]

As described above, according to the present invention,
Using an interferometer, even if there are variations in the optical characteristics of the surface to be measured, the shape measuring device capable of reliably measuring the surface shape, the control device of the shape measuring device, and the shape measuring device A control program is realized.

[Brief description of drawings]

FIG. 1 is a configuration diagram of a flatness measuring apparatus according to a first embodiment, a second embodiment, and a third embodiment.

FIG. 2 is a diagram illustrating a measured surface 20a of a semiconductor wafer 20.

FIG. 3 is an operation flowchart of a computer 30 according to the first embodiment.

FIG. 4 is a conceptual diagram illustrating measurement of a light amount La.

FIG. 5 is a conceptual diagram illustrating measurement of a light amount Lb.

FIG. 6 is a conceptual diagram illustrating measurement of a light amount Lc.

FIG. 7 is a conceptual diagram illustrating measurement of a light amount Ld.

FIG. 8 is a conceptual diagram illustrating measurement of the light amount Le.

FIG. 9 is an operation flowchart of the computer 40 according to the second embodiment.

FIG. 10 is a histogram of the amplitude output of the image sensor 18a.

FIG. 11 is an example of a correspondence table stored in the computer 50 of the third embodiment.

[Explanation of symbols]

10 Interferometer device 11 light source 12 Light intensity setting device 12A Light intensity adjustment filter 12B filter driver 13A, 13B, 17 lenses 15 Optical path difference variable device 15A Fizeau Flat 15a Reference plane 20 Semiconductor wafer 20a surface to be measured 18 Imaging unit 18a image sensor 30, 40, 50 computers 50a display 50b input device Pa, Pb, Pc, Pd, Pe patterns

Claims (9)

[Claims] 1. An interferometer for irradiating light on a surface to be measured and a predetermined reference surface and detecting an interference fringe formed by the light from these surfaces by an image sensor, and a control device for driving and controlling the interferometer. A shape measuring device provided with, wherein the control device sets the light amount of the light irradiated on the surface to be measured to each of a predetermined value group, and the light amount is set to each value. Sometimes, the measurement unit that drives the interferometer to perform the interferometric measurement a plurality of times, based on the pixel output of the image sensor at each of the interferometric measurements, which shows an appropriate value, the shape of the surface to be measured. A shape measuring device, comprising: a calculation unit that acquires information. 2. The shape measuring device according to claim 1, wherein the control device sets the light amount of the light with which the surface to be measured is irradiated to a predetermined value prior to the operation of the measuring section, and at that time. A test measurement unit that refers to the pixel output of the image sensor, and a value determination that determines the value group to be set by the measurement unit according to the variation in the optical characteristics in the measured surface indicated by the pixel output. A shape measuring device comprising: 3. The shape measuring device according to claim 1, wherein the control device includes a storage unit that stores a plurality of types of value groups in association with the types of the surface to be measured, Prior to the operation of the measuring unit, a value group to be set by the measuring unit according to the type of the surface to be measured is determined as a value group stored in the storage unit in association with the type. A shape measuring device comprising: a determining unit. 4. A shape measuring apparatus applied to a shape measuring apparatus equipped with an interferometer for irradiating light on a surface to be measured and a predetermined reference surface and detecting an interference fringe formed by the light from those surfaces by an image sensor. Of the light amount of the light irradiated on the surface to be measured is set to each of a predetermined value group, and the interferometer is driven when the light amount is set to each value. Then, a measurement unit that performs interferometric measurement a plurality of times, and a pixel output of the image sensor at the time of each interferometric measurement, based on the one showing an appropriate value, a calculation unit that acquires the shape information of the surface to be measured. A control device for a shape measuring device, comprising: 5. The control device for the shape measuring apparatus according to claim 4, wherein the light amount of the light with which the surface to be measured is irradiated is set to a predetermined value prior to the operation of the measuring section, and A test measurement unit that refers to the pixel output of the image sensor, and a value determination unit that determines the value group that the measurement unit should set according to the variation in the optical characteristics in the measured surface indicated by the pixel output. A control device for a shape measuring device, comprising: 6. The shape measuring apparatus control device according to claim 4, further comprising a storage unit that stores a plurality of types of value groups in association with the types of the surface to be measured, Prior to the operation of the unit, a value determination unit that determines a value group to be set by the measurement unit according to the type of the surface to be measured, into a value group stored in the storage unit in association with the type. A control device for a shape measuring device, comprising: 7. A computer is operated as a control device of a shape measuring apparatus equipped with an interferometer for irradiating light on a surface to be measured and a predetermined reference surface and detecting an interference fringe formed by the light from those surfaces by an image sensor. A control program of a shape measuring device for causing the light amount of the light irradiated on the surface to be measured to be set to each of a predetermined value group, and when the light amount is set to each value In the measurement procedure of driving the interferometer to perform interferometric measurement a plurality of times, among the pixel outputs of the image sensor at the time of each interferometric measurement, based on the one showing an appropriate value, the shape information of the measured surface. A control program for a shape measuring apparatus, comprising: 8. The control program of the shape measuring apparatus according to claim 7, wherein the light quantity of the light with which the surface to be measured is irradiated is set to a predetermined value prior to the execution of the measurement procedure, and A test measurement procedure that refers to the pixel output of the image sensor, and a value determination procedure that determines the value group to be set in the measurement procedure according to the variation in the optical characteristics in the measured surface indicated by the pixel output. A control program for a shape measuring apparatus, comprising: 9. The control program for the shape measuring apparatus according to claim 7, wherein a group of values to be set in the measuring procedure according to the type of the surface to be measured prior to executing the measuring procedure. And a value determination procedure for determining a value group associated in advance with the type, a control program of the shape measuring apparatus.