JP2009236706A - Shape calculator, shape calculation program, shape calculating method, and shape-measuring device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a shape calculator for highly accurately measuring a shape by performing phase connection processing based on phase data obtained by an optical interferometer even concerning an object to be measured in which a singular part of a dust deposition part, a damaged part or the like partially exists, and to provide its program, its method and a shape-measuring device. <P>SOLUTION: The shape calculator includes: dividing a plurality of phase data obtained to detect the phase of an intensity signal of interference light into processing object data in which amplitude of an interference light intensity signal corresponding to the phase data is within an amplitude allowable range and non-processing object data other than them while changing measuring points with the optical interferometer for irradiating object light on the surface of the object to be measured 1; sequentially executing phase connection processing based on the two processing object data in which the measuring points are adjacent along one scanning line; and executing phase connection processing based on the processing object data and adjacent other processing object data in a direction where the measuring points to the processing object data are intersected to one scanning line when the measuring points of the non-processing object data and the measuring points of the processing object data are adjacent along one scanning line. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a shape calculation device that calculates the shape of an object to be measured such as a semiconductor wafer based on phase data obtained through an optical interferometer, a program and method thereof, and a shape measurement device including the shape calculation device. It is.

In the shape measurement of a thin semiconductor wafer (an example of an object to be measured, hereinafter referred to as a wafer), a non-contact type shape measuring apparatus using an interferometer has become widespread. This is because measurement light that is reflected light that reflects one of the light beams branched into two on the surface of the object to be measured, and reference light that is reflected light that reflects the other light beam on a predetermined reference surface, and The surface shape (surface height distribution) of the object to be measured is obtained from the interference image formed by the interference light.
More specifically, when measuring the surface shape of a wafer with an optical interferometer, it is obtained by irradiating a number of measurement sites on the surface of the wafer with object light using an optical interferometer arranged opposite to the wafer surface. The phase of the intensity signal of the interference light is detected, and phase connection processing (so-called unwrap processing) based on the phase data for each of the plurality of measurement sites obtained thereby is performed. Each phase in the phase distribution information obtained by this phase connection process can be converted into a surface height dimension value based on the wavelength of the object light. Therefore, the phase distribution information obtained by the phase connection process is equivalent to the wafer surface height distribution information, that is, shape information.
As a result, the surface shape of the wafer can be measured in a non-contact manner, so that the surface shape can be measured without causing scratches or the like on the wafer surface as in the case of measuring with a stylus type shape meter. In the measurement of the shape of the wafer, it is necessary to measure the shape over the entire surface thereof, and therefore generally the measurement is performed with the edge portion around the wafer supported (usually, three points are supported).

In Patent Document 1, optical interferometers are arranged on the front and back of the wafer 1, and object light (measurement light) and reference light obtained by one optical interferometer are guided to the other optical interferometer by a prism or the like. There is shown a shape measuring apparatus that causes interference again and measures the thickness (an example of the shape) of the wafer 1 based on the interference light.
According to the shape measuring apparatus shown in Patent Document 1, the displacement of the object to be measured caused by vibration is canceled by irradiating the same object light to the measurement parts opposite to each other, and the vibration of the object to be measured. It is possible to measure the thickness with high accuracy without being affected by the above.

Patent Document 2 discloses details of the phase connection process.
Patent Document 2 discloses a technique for measuring a change in characteristics of a fluid contained in a cell by detecting a change in phase of interference light in which object light passed through the cell and other reference light are superimposed. It is shown. At that time, the phase data is sampled at a predetermined cycle, and the phase data at a certain time point is set so that the phase difference falls within the range of −π to + π with reference to the phase data at the previous time point. Then, a phase connection process for shifting the phase by an integral multiple of 2π is performed.
Similarly, in phase connection processing in shape measurement, one phase of two phase data obtained at two adjacent measurement points is set to a phase difference b in the range of −π to + π with the other phase as a reference. In order to be within the range, each phase data is corrected to shift the phase by an integral multiple of 2π. This phase connection process is based on the premise that the difference in surface height between two adjacent measurement points does not exceed a quarter wavelength of object light.
JP 2003-329422 A JP 2000-292351 A

  However, in the conventional shape measurement using an optical interferometer, when there is a part (hereinafter referred to as a singular part) in which the reflection characteristic of the object light is significantly different from the other part of the shape measurement region of the object to be measured. There is a problem in that an erroneous phase connection process result that does not represent the original shape of the measurement object is obtained. The object light reflected on the singular part exhibits a different behavior from the object light reflected on other measurement parts due to the reflection characteristics different from those of the measurement parts other than the singular part, and this is reflected in the interference light intensity signal. This is noise. Examples of the singular part include a part where dust adheres to or is scratched on the surface of the object to be measured, and a part where a member that supports the object to be measured exists.

FIG. 10 is a schematic diagram in which phase data after phase connection processing obtained by conventional shape measurement using an optical interferometer and phase data representing the original shape are graphed for the object to be measured having the singular part. FIG. In FIG. 10, the horizontal axis of the graph is the coordinate axis of the number for identifying each measurement point, and the vertical axis is the coordinate axis of the phase θ ′ after the phase connection process. In addition, the measurement points having consecutive numbers are adjacent to each other on the surface of the object to be measured. In FIG. 10, the (Kx + 1) th and (Kx + 2) th measurement points are measurement points in the singular part.
When the singular part is present in a part of the object to be measured, in the phase data obtained for a plurality of measurement points through an optical interferometer, the light reflection characteristics between the measurement point in the singular part and the surrounding measurement points A phase difference noise caused by the difference occurs.
However, when conventional phase connection processing is performed on such phase data including noise, as shown in FIG. 10, at the measurement points in the singular part ((Kx + 1) th and (Kx + 2) th in FIG. 10). , The phase after processing is greatly different from the phase representing the original shape.
Further, the phase connection is also performed at all subsequent measurement points (from (Kx + 3) in FIG. 10) on which the phase connection is performed in a chain manner with reference to the phase of the measurement point ((Kx + 2) in FIG. 10) in the singular part. The phase after processing is significantly different from the phase representing the original shape.
Therefore, the present invention has been made in view of the above circumstances, and the object of the present invention is to have a portion (singular portion) having a different light reflection characteristic (singular portion) such as a dust adhesion portion or a scratched portion. A shape calculation device capable of measuring a shape with high accuracy by performing phase connection processing based on phase data obtained by an optical interferometer, a program therefor, a method therefor, and a shape measurement device are provided. There is to do.

In order to achieve the above object, a shape calculation device according to the present invention is a device for calculating a shape value of a device under test based on phase data obtained for each of a plurality of measurement sites of the device under test. Each component shown in 1) to (3) is provided. Here, the plurality of pieces of phase data are obtained by changing the relative position in the two-dimensional direction between the object to be measured and the optical interferometer that irradiates the measurement site on the surface thereof with the interference light in the optical interferometer. This is data obtained by an interference light phase detection device that detects the phase of the intensity signal. Note that “changing the relative position” means that the position of the optical interferometer is fixed and only the position of the measured object is moved, and the position of the measured object is fixed and the optical interference is fixed. This includes moving only the position of the meter and moving the positions of both the optical interferometer and the device under test.
(1) Amplitude data acquisition means for acquiring a plurality of amplitude data obtained by detecting the amplitude of the intensity signal of the interference light that is the detection target each time the phase is detected.
(2) The phase data obtained by the interference light phase detection device is converted into processing target data whose amplitude data corresponding to the phase data is within a preset amplitude tolerance range and other non-processing target data. Data sorting means to sort.
(3) Phase connection means for calculating a shape value of the object to be measured by performing phase connection processing based only on the processing target data among the phase data obtained by the interference light phase detection device.
For example, it is conceivable that the phase connecting means executes the following processes (3-1) and (3-2).
(3-1) The measurement sites are adjacent to each other along a predetermined scanning line that sequentially extends from the measurement site that is the starting point of the plurality of measurement sites corresponding to the plurality of phase data to the next measurement site. The phase connection process based on the two data to be processed is sequentially executed.
(3-2) When the measurement part of the non-processing target data and the measurement part of the processing target data are adjacent to each other along the predetermined scanning line, the processing target data and the processing target data are On the other hand, the phase connection process based on the other data to be processed adjacent in the direction in which the measurement site intersects the predetermined single scanning line is executed.

The shape calculation apparatus according to the present invention converts each phase data obtained for each of a plurality of measurement parts of the object to be measured, based on the amplitude of the intensity signal of the interference light (that is, the signal intensity). It is divided into the non-processing object data obtained in the singular part, such as the part where the error occurs, and the other processing object data. This utilizes a phenomenon in which the difference in the light reflection characteristics on the surface of the object to be measured appears as the difference in the amplitude signal of the interference light.
And the shape calculation apparatus which concerns on this invention performs a phase connection process only based on the phase data (the said process target data) obtained in measurement parts other than the said specific | specification part. As a result, the shape calculation apparatus according to the present invention can calculate a shape value with high accuracy even for the object to be measured in which the singular part having a light reflection characteristic different from others is present in part.

In addition, the amplitude allowable range may be set to a predetermined range for each type of object to be measured. However, the amplitude of the intensity signal of the interference light is caused by the difference in the inclination of the surface of the measurement site with respect to the incident direction of the object light. The fluctuation range is relatively large depending on On the other hand, the amplitude detected at the measurement site adjacent in the region of the singular part has a small fluctuation because the surface inclination of the measurement site approximates.
Therefore, it is preferable that the shape calculation apparatus according to the present invention includes the constituent element shown in the following (4).
(4) The allowable amplitude range corresponding to each of the plurality of phase data obtained by the interference light phase detection device is obtained at one or a plurality of other measurement sites around the measurement site corresponding to the phase data. An amplitude allowable range setting unit that dynamically sets the amplitude data as a reference.
As a result, the accurate amplitude tolerance range is dynamically set for each phase data (that is, for each measurement site), and the accuracy of the phase data classification, that is, the accuracy of specifying the singular part is increased.

Further, the present invention can also be understood as a shape calculation program for causing a computer to execute the procedure of processing executed by each means included in the shape calculation apparatus according to the present invention described above.
Similarly, in the present invention, the procedure of processing executed by each unit included in the above-described shape calculation apparatus according to the present invention can also be understood as a shape calculation method executed by a computer.

Moreover, this invention can also be grasped | ascertained as a shape measuring apparatus containing the shape calculation apparatus which concerns on this invention shown above.
That is, the shape measuring apparatus according to the present invention includes an optical interferometer that irradiates object light onto a measurement site on the surface of the object to be measured, and further includes the components shown in the following (a) to (c). Prepare.
(A) A movable support means for changing a relative position between the object to be measured and the optical interferometer in a two-dimensional direction.
(B) Phase detection means for obtaining phase data by detecting the phase of the intensity signal of the interference light in the optical interferometer according to a change in the relative position between the object to be measured and the optical interferometer by the movable support means.
(C) The shape calculation apparatus (shape calculation according to the present invention) that acquires the phase data for each of a plurality of measurement sites obtained by the phase detection means, and calculates the shape value of the device under test based on the phase data. apparatus).

In addition, it is further preferable that the shape measuring apparatus according to the present invention further includes each component shown in the following (d) to (f).
(D) a first measurement light and a second measurement light emitted from a predetermined light source, each of which has a different frequency, and a measurement part of the front surface that is a part opposite to the front and back of the object to be measured; Light guiding means for guiding the measurement area on the back surface in each direction.
(E) One of the pair of optical interferometers, which irradiates the measurement site on the front surface as object light with the first measurement light guided in the direction of the measurement site on the front surface. , Irradiating the first reference surface with the second measurement light guided in the direction of the measurement region on the front surface, and the reflected light of the first measurement light from the measurement region on the front surface And a reflected light of the second measurement light from the first reference surface, and a heterodyne optical interferometer on the front surface side that outputs an intensity signal of the interference light.
(F) the other of the pair of optical interferometers, irradiating the second measurement light guided in the direction of the measurement part of the back surface as the object light on the measurement part of the back surface; Irradiating the second reference surface with the first measurement light guided in the direction of the measurement site, and from the reflected light of the second measurement light from the measurement site on the back surface and the second reference surface A heterodyne interferometer on the back surface side that interferes with the reflected light of the first measurement light and outputs an intensity signal of the interference light.
Then, the phase detection means is configured to change the level of the intensity signal output from each of the two heterodyne interferometers according to a change in relative position between the object to be measured and the two heterodyne optical interferometers by the movable support means. The result of detecting the phase difference is defined as the phase data.
Further, the shape calculation device acquires the phase data for each of a plurality of measurement sites obtained by the phase detection means, and calculates a shape value representing the thickness distribution of the surface of the object to be measured based on the phase data. .

According to the known heterodyne interferometer principle, the phase of the output signal of the front heterodyne interferometer depends on the surface position (height) of the measurement part of the front surface of the object to be measured. However, the phase of the signal reflects the component of the shape of the measurement portion of the front surface itself and the component of the displacement due to the vibration of the object to be measured.
Similarly, the phase component of the output signal of the heterodyne interferometer on the back surface side reflects the component of the shape of the measurement site itself of the back surface and the component of the displacement due to the vibration of the object to be measured.
Further, in the front surface side heterodyne interferometer and the back surface side heterodyne interferometer, the correspondence of which of the first measurement light and the second measurement light is the reference light or the object light The relationship is reversed. That is, the first measurement light that is object light on one surface (front surface) of the object to be measured is reference light on the other surface (back surface), and The second measurement light serving as reference light is object light on the other surface.
For this reason, the phase difference (phase data) detected by the phase detection means cancels out the displacement component due to the vibration of the object to be measured, as shown in equations (1) to (3) described later. The amount of displacement in which only the component of the shape of the measurement part of the front surface and the component of the shape of the measurement part of the back surface are reflected, that is, the measurement part of the front surface of the object to be measured And it becomes a measured value corresponding to the thickness of the position of the measurement site on the back surface. As a result, more accurate thickness measurement can be performed without being affected by displacement due to vibration or the like of the object to be measured.

  According to the present invention, when there is a part (singular part) having different light reflection characteristics, such as a part where dust is attached or a part having a scratch, in a part of the shape measurement region of the object to be measured, Of the phase data obtained by the meter, the phase connection process based on only the data other than the phase data obtained by the singular part (the data to be processed) is performed. As a result, it is possible to measure the shape of the object to be measured in which the singular part is partially present with high accuracy.

Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings so that the present invention can be understood. The following embodiment is an example embodying the present invention, and does not limit the technical scope of the present invention.
Here, FIG. 1 is a schematic configuration diagram of a shape measuring apparatus X according to an embodiment of the present invention, FIG. 2 is a configuration diagram of an interference light detection unit Y provided in the shape measuring apparatus X, and FIG. FIG. 4 is a schematic diagram showing an example of the distribution of measurement points on the object to be measured. FIG. 5 is a diagram showing the intensity signal of the interference light obtained by measuring the object to be measured having a singular part. FIG. 6 is a schematic diagram showing changes in amplitude, FIG. 6 is a schematic diagram in which data after phase connection processing at measurement points around the singular part of the object to be measured, calculated by the shape measuring apparatus X, is graphed, and FIG. 7 is a shape measurement. FIG. 8 is a schematic diagram showing another example of the distribution of measurement points on the measurement object, and FIG. 9 is an oblique incidence interference applicable to the shape measurement apparatus X. FIG. Fig. 10 is a schematic diagram of the total, and Fig. 10 It is a schematic diagram and the phase of the data is graphed representing the phase of the data and the original shape after phase unwrapping process obtained by the conventional shape measurement using an optical interferometer for Jobutsu.

Hereinafter, the shape measuring apparatus X according to the embodiment of the present invention will be described with reference to the configuration diagram shown in FIG.
As shown in FIG. 1, the shape measuring apparatus X includes an interference light detection unit Y including a pair of optical interferometers a20 and b20, two phase detectors 4 and 5, a calculator 6, and a movable support device Z. It has.
The computer 6 includes a CPU 6a, a memory 6b, and a signal input / output interface 6c (shown as I / O in the figure). When the CPU 6a executes a predetermined program, various operations and the signal input / output interface 6c are performed. For example, transmission / reception of signals to / from an external device and recording of data in the memory 6b are performed.
The pair of optical interferometers a20 and b20 irradiate the measurement points 1a and 1b on the front and back surfaces of the thin plate-like object 1 such as a semiconductor wafer with the reflected light (object light) and a predetermined reference. The optical device outputs intensity signals Sg1 and Sg2 of interference light on which light is superimposed. The pair of optical interferometers a20 and b20 also output the other interference light intensity signals Ref1 and Ref2 used for correcting the shape value of the DUT 1, which will be described later.
One of the phase detectors 4 (hereinafter referred to as the first phase detector 4) performs phase detection of the intensity signals Sg1 and Sg2 of the interference light output from the pair of optical interferometers a20 and b20, This is a signal processing device that outputs phase data ΔΦs (hereinafter referred to as first phase data ΔΦs) that is the detection result. Further, the phase detector 4 detects the amplitudes As1 and As2 of the interference light intensity signals Sg1 and Sg2 at the time of phase detection, and outputs the detection results.
The other phase detector 5 (hereinafter referred to as the second phase detector 5) is configured to output intensity signals Ref1 and Ref2 of the other interference light for correction output from the pair of optical interferometers a20 and b20. This is a signal processing device that performs phase detection and outputs phase data ΔΦr (hereinafter referred to as second phase data ΔΦr) as a result of the detection.

Then, the calculator 6 corrects phase data ΔΦsr (corrected) which is a difference between the first phase data ΔΦs and the second phase data ΔΦr obtained for each of the plurality of measurement points 1a, 1b on the front and back surfaces of the DUT 1. = ΔΦs−ΔΦr), the shape value of the DUT 1 is calculated. Here, the plurality of post-correction phase data ΔΦsr change the relative positions of the device under test 1 and the pair of optical interferometers a20 and b20 in the two-dimensional direction (directions along the front and back surfaces of the device under test 1). The data obtained by detecting the phase of the intensity signal of the interference light in the pair of optical interferometers a20 and b20 by the two phase detectors 4 and 5.
However, based on the amplitudes As1 and As2 obtained for each of the plurality of measurement points 1a and 1b, the calculator 6 determines whether or not the corresponding corrected phase data ΔΦsr is appropriate data for the phase connection process. (Whether it is processing target data or non-processing target data to be described later). Then, as will be described later, the computer 6 executes the phase connection process based only on the corrected phase data ΔΦsr classified as suitable for the phase connection process.
Hereinafter, the detail of each component with which the shape measuring apparatus X is provided is demonstrated.

The movable support device Z supports the device under test 1 between the object light emitting portions from the pair of optical interferometers a20 and b20, and moves the support position in a two-dimensional direction (horizontal direction). This is a device that changes the relative position of the pair of optical interferometers a20 and b20 (an example of movable support means).
As shown in FIG. 1, the movable support device Z includes a rotation shaft 41, a support portion 44 connected to the rotation shaft 41, a rotation drive portion 42, a linear movement mechanism 43, and a movement control device 7.
A disk-shaped object to be measured 1 such as a semiconductor wafer is supported at three points by support portions 44 arranged at three locations on the circumference at the edge (edge portion). These three support portions 44 are connected to a rotation shaft 41 extending toward the center of the circumference.
Further, the rotary shaft 41 is rotationally driven by the rotational drive unit 42 such as a stepping motor. As a result, the DUT 1 is rotated with its center portion as the center of rotation.
Further, the rotation shaft 41 and the rotation drive unit 42 are linearly moved within a predetermined movement range in a direction parallel to the front and back surfaces of the DUT 1 (direction perpendicular to the thickness direction) by the linear movement mechanism 43. Is done. That is, the linear movement mechanism 43 moves the DUT 1 along its radial direction.
The movement control device 7 is a device that controls the movement of the rotation driving unit 42 and the linear movement mechanism 43. Further, the movement control device 7 detects the irradiation position of the object light on the object 1 to be measured, that is, the positions of the measurement points 1 a and 1 b that change as needed, and transmits the detection results to the computer 6. The positions of the measurement points 1a and 1b are detected based on, for example, the history of operation commands for the rotation drive unit 42 and the linear movement mechanism 43 (that is, the movement history of the DUT 1) or the rotation drive unit 42. And it detects based on the detection result of the position detection sensor (not shown) provided in each of the said linear movement mechanism 43. FIG.

Then, the shape measuring apparatus X uses the rotation of the DUT 1 by the rotation driving unit 42 and the movement of the DUT 1 in the linear direction by the linear movement mechanism 43 to measure the DUT 1 at the DUT 1. The first phase data ΔΦs and the second phase data ΔΦr at a plurality of measurement points 1a, 1b are detected while sequentially changing the positions of the points 1a, 1b (object light irradiation spots).
For example, the movement control device 7 continuously rotates and linearly moves the device under test 1 at a constant speed, and at a constant cycle or whenever the positions of the measurement points 1a and 1b become predetermined positions. A data acquisition command is transmitted to the computer 6. Then, the computer 6 acquires the first phase data ΔΦs and the second phase data ΔΦr from each of the phase detectors 4 and 5 in response to reception of the data acquisition command, and obtains a difference between them. The corrected phase data ΔΦsr (= ΔΦs−ΔΦr) for a plurality of measurement points 1a and 1b is obtained.
FIG. 4 is a schematic diagram illustrating an example of the distribution of measurement points in the DUT 1.
When the phase of the interference light is sequentially detected while rotating the measured object 1 and moving linearly, the positions of the measurement points 1a and 1b are spiral lines on the surface of the measured object 1 as shown in FIG. It changes sequentially along the (dashed line).
Hereinafter, the measurement points corresponding to each of the plurality of corrected phase data ΔΦsr obtained in the computer 6 are identified by the measurement point numbers (1, 2, 3,...) Assigned in accordance with the data acquisition order. To do. Therefore, as shown in FIG. 4, when a certain measurement point number is Kb, the Kb-th measurement point and the (Kb + 1) -th measurement point are adjacent points, and the (Kb + 1) -th measurement point. The (Kb + 2) -th measurement point is an adjacent point.
As described above, in the shape measuring apparatus X, the object light from each of the pair of interferometers a20 and b20 is scanned along one scanning line R on the DUT 1. In other words, the one scanning line R is a predetermined one that sequentially extends from the measurement point that is the starting point at the plurality of measurement points corresponding to each of the plurality of corrected phase data ΔΦsr to the next measurement point. Is a line.
In FIG. 4, the hatched portion is the singular portion where the light reflection characteristic on the surface of the DUT 1 is significantly different from other portions. FIG. 4 shows an example in which the (Kb + 1) th and (Kb + 2) th measurement points exist in the singular part.

Next, the interference light detection unit Y including the pair of optical interferometers a20 and b20 will be described with reference to the configuration diagram shown in FIG.
As shown in FIG. 2, the interference light detection unit Y includes a two-polarized light source 2, a polarization beam splitter 3 (hereinafter referred to as PBS3), a plurality of mirrors a11 to a13, b11, b12, and A A surface-side heterodyne interferometer a20 and a B-side heterodyne interferometer b20, an A-side correction interferometer a30 and a B-side correction interferometer b30, a first phase detector 4 and a second phase detector 5; It has. Here, the A-side correction interferometer a30 and the B-side correction interferometer b30 are the pair of optical interferometers.
Hereinafter, for the sake of convenience, one surface (the upper surface in FIG. 2) of the DUT 1 is the A surface (corresponding to the front surface), and the other surface in front and back is the B surface (the back surface). Equivalent). Further, the surface portion on the A surface side at the measurement position of the thickness of the DUT 1 is the A surface measurement point 1a (corresponding to the measurement portion of the front surface), and the B surface opposite to the A surface measurement point 1a. The surface portion is referred to as a B-surface measurement point 1b (corresponding to the measurement portion of the back surface).
Although not shown in FIG. 2, the DUT 1 is supported by the movable support device Z (see FIG. 1).

The dual-polarized light source 2 is a laser light source that emits two light beams P1 and P2 having slightly different frequencies. 2 is, for example, a Zeeman laser that emits two light beams P1 and P2 having slightly different frequencies coaxially or substantially coaxially (in a state where two light beams partially overlap). Etc. Hereinafter, one light beam is referred to as a first light beam P1, and the other light beam is referred to as a second light beam P2.
In FIG. 2, for the sake of convenience, it is described that both beam lights P1 and P2 are emitted along different axes. However, in this embodiment, both beam lights P1 and P2 have the same axis (the same It is assumed that the light is emitted along an optical path) or a partially overlapping optical path. The first beam light P1 and the second beam light P2 are single wavelength light, and their frequencies are not particularly limited. For example, the frequency ω of one beam light is about 5 × 10 ⁸ MHz ( This is an example in which a visible laser light source is used, and the difference in frequency between the two light beams is about several tens of kHz. Further, the first beam light P1 and the second beam light P2 emitted from the dual-polarized light source 2 have different polarization plane directions. Here, it is assumed that the polarization planes of the two light beams P1 and P2 are orthogonal. The first beam light P1 and the second beam light P2 are examples of the first measurement light and the second measurement light having different frequencies.
The two light beams P1 and P2 include a light source that emits one light beam, a beam splitter that divides the emitted light beam into two, and an acoustooptic device that converts the frequency of one of the two branched light beams. It is also conceivable that they are generated by the above.

The PBS 3 splits each of the first beam light P1 (frequency: ω) and the second beam light (frequency: ω + Δω) emitted from the dual-polarized light source 2 into two branches. Then, the first beam light P1 and the second beam light P2 branched by the PBS 3 are guided in the direction of the A-plane measurement point 1a of the DUT 1 by the three mirrors a11, a12, and a13. The other first beam light P1 and second beam light P2 branched by the PBS 3 are guided in the direction of the B surface measurement point 1b of the object 1 to be measured by the two mirrors b11 and b12. The PBS 3 and the mirrors a11 to a13, b11, b12 are an example of the light guide means.
In the example shown in FIG. 2, since the two light beams P1 and P2 are guided along the same optical path, even if fluctuations occur in the light guide paths of the two light beams, the fluctuations equally affect the two light beams P1 and P2. Therefore, it is possible to prevent the measurement accuracy from deteriorating due to fluctuations.
In addition, as an optical device for guiding the first beam light P1 and the second beam light P2 to the A-surface measurement point 1a and the B-surface measurement point 1b, an optical fiber or the like can be considered in addition to a mirror.

  As shown in FIG. 2, the A-side heterodyne interferometer a20 (corresponding to the front-side heterodyne interferometer) includes a polarization beam splitter a21 (hereinafter referred to as A-side PBS (a21)), two Quarter-wave plates a22 and a23, A-side reference plate a24, polarizing plate a25 (hereinafter referred to as A-side first polarizing plate a25) and photodetector a26 (hereinafter referred to as A-side first photodetector a26). Is provided). One quarter-wave plate a22 is disposed between the A-plane side PBS (a21) and the A-plane measurement point 1a, and the other quarter-wave plate a23 is disposed on the A-plane side PBS ( a21) and the A-side reference plate a24.

The A surface side PBS (a21) transmits the first beam light P1 guided in the direction of the A surface measurement point 1a by the mirrors a11 to a13, and thereby transmits the first beam light P1 to the A surface measurement point. While irradiating 1a and reflecting the 2nd beam light P2 similarly guided to the direction of the A surface measurement point 1a, the second beam light P2 is reflected on the surface of the A surface side reference plate a24 (the first surface). Equivalent to the reference plane). In the example shown in FIG. 2, the first beam light P1 and the second beam light P2 are perpendicularly incident on the surface of the A surface measurement point 1a and the surface of the A surface side reference plate a24, respectively.
Further, the A-plane side PBS (a21) reflects the reflected light of the first beam light P1 from the A-plane measurement point 1a and reflects the second beam light P2 from the surface of the A-plane side reference plate a24. Both light and light are guided in the direction (the same direction) of the A-plane side first polarizing plate a25.
The state of polarization of the first light beam P1 emitted from the A-plane side PBS (a21) to the A-plane measurement point 1a side (P-polarized light or S-polarized light) due to the presence of the quarter-wave plate a22. And the state of polarization of the first beam light P1 reflected on the A-plane measurement point 1a and incident on the A-plane side PBS (a21) is switched. Similarly, due to the presence of the quarter-wave plate a23, the polarization state of the second beam light P2 emitted from the A-plane side PBS (a21) to the A-plane measurement reference plate a24 side, and the A-plane measurement The polarization state of the second light beam P2 that is reflected by the reference plate a24 and incident on the A-side PBS (a21) is switched.
The A-plane-side first polarizing plate a25 transmits only the light having a polarization plane in a predetermined direction (an intermediate direction between the polarization planes of both light beams), whereby the first beam from the A-plane measurement point 1a is transmitted. The reflected light of the light P1 is caused to interfere with the reflected light of the second beam light P2 from the surface of the A-plane side reference plate a24. The interference light (hereinafter referred to as A-surface measurement interference light) is input (incident) into the A-surface side first photodetector a26.
The A surface side first photodetector a26 outputs the intensity signal Sig1 (electric signal) of the A surface measurement interference light by receiving the A surface measurement interference light and performing photoelectric conversion.

On the other hand, the B-side heterodyne interferometer b20 has the same configuration as the A-side heterodyne interferometer a20, but both of them use either the first beam light P1 or the second beam light P2 as reference light ( The correspondence relationship between the irradiation light on the reference plate) and the object light (irradiation light on the measurement point) is reversed.
That is, the B-side heterodyne interferometer b20 (corresponding to the back-side heterodyne interferometer) includes a polarization beam splitter b21 (hereinafter referred to as B-side PBS (b21)), two pieces as shown in FIG. Quarter-wave plates b22 and b23, B-side reference plate b24, polarizing plate b25 (hereinafter referred to as B-side first polarizing plate b25) and photodetector b26 (hereinafter referred to as B-side first photodetector b26). Is provided). One quarter-wave plate b22 is disposed between the B-side PBS (b21) and the B-plane measurement point 1b, and the other quarter-wave plate b23 is disposed on the B-side PBS ( b21) and the B-side reference plate b24.

The B-side PBS (b21) transmits the second beam light P2 guided in the direction of the B-plane measurement point 1b by the mirrors b11 to b12, and thereby transmits the second beam light P2 to the B-plane measurement point. While irradiating 1b and reflecting the 1st beam light P1 similarly guided to the said B surface measurement point 1b direction, the 1st beam light P1 is made into the surface (said 2nd said 2nd surface side reference board b24). Equivalent to the reference plane). In the example shown in FIG. 2, the second beam light P2 and the first beam light P1 are perpendicularly incident on the surface of the B surface measurement point 1b and the surface of the B surface side reference plate b24, respectively.
Further, the B surface side PBS (b21) reflects the reflected light of the second beam light P2 from the B surface measurement point 1b and the reflected light of the first beam light P1 from the surface of the B surface side reference plate b24. Both the light and the light are guided in the direction (the same direction) of the B-side first polarizing plate b25.
The state of polarization of the second light beam P2 emitted from the B-side PBS (b21) to the B-side measurement point 1b due to the presence of the quarter-wave plate b22 (S-polarized light or P-polarized light) And the state of polarization of the second light beam P2 reflected on the B-surface measurement point 1b and incident on the B-surface side PBS (b21) is switched. Similarly, due to the presence of the quarter-wave plate b23, the polarization state of the first beam light P1 emitted from the B-side PBS (b21) to the B-side measurement reference plate b24 side, and the B-side measurement. The polarization state of the first light beam P1 reflected on the reference plate b24 and incident on the B-side PBS (b21) is switched.
The B-side first polarizing plate b25 transmits only the light having the polarization plane in a predetermined direction (the intermediate direction between the polarization planes of both light beams), so that the second beam from the B-plane measurement point 1b is transmitted. The reflected light of the light P2 is caused to interfere with the reflected light of the first beam light P1 from the surface of the B-side reference plate b24. The interference light (hereinafter referred to as B-surface measurement interference light) is input (incident) into the B-surface side first photodetector b26.
The B surface side first photodetector b26 outputs the intensity signal Sig2 (electric signal) of the B surface measurement interference light by receiving the B surface measurement interference light and performing photoelectric conversion.

  The first phase detector 4 (corresponding to the phase detection means) calculates the phase difference ΔΦs between the intensity signals Sig1 and Sig2 output from the A-side heterodyne interferometer a20 and the B-side heterodyne interferometer b20, respectively. The detected value is output as an electric signal (detection signal). For example, a lock-in amplifier can be adopted as the first phase detector 4. Here, the value (phase difference ΔΦs) of the output signal of the first phase detector 4 is the difference between the surface position of the A-surface measurement point 1a and the surface position of the B-surface measurement point 1b in the device under test 1. That is, it changes according to the thickness of the position of both measurement points 1a and 1b on the DUT 1.

According to the known principle of the heterodyne interferometer, the phase of the output signal Sig1 of the A-plane side heterodyne interferometer a20 is determined according to the surface position (height) of the A-plane measurement point 1a, but the phase of the output signal Sig1 This reflects both the shape (unevenness) component of the A surface measurement point 1a itself and the displacement amount component due to the vibration of the DUT 1.
Similarly, the phase of the output signal Sig2 of the B-side heterodyne interferometer b20 is determined according to the surface position (height) of the B-plane measurement point 1b, but the phase of the output signal Sig2 includes the B-plane measurement. Both the shape (irregularity) component of the point 1b itself and the displacement component due to the vibration of the DUT 1 are reflected.
Further, as described above, in the A-side heterodyne interferometer a20 and the B-side heterodyne interferometer b20, either the first beam light P1 or the second beam light P2 is used as reference light or object light. The corresponding relationship is reversed.
For this reason, the phase difference ΔΦs detected by the first phase detector 4 cancels out the displacement component due to the vibration of the DUT 1 between the A plane side and the B plane side, and the A plane measurement point 1a itself The amount of displacement reflecting only the shape (unevenness) component and the shape (unevenness) component of the B-side measurement point 1b itself, that is, the A-side measurement point 1a and the B-side measurement point 1b of the object 1 to be measured. The measured value corresponds to the thickness of the position.

On the other hand, as shown in FIG. 2, the A-side correction interferometer a30 includes a beam splitter a31 (hereinafter referred to as A-side BS (a31)) and a polarizing plate a32 (hereinafter referred to as A-side second polarizing plate a32). And a photodetector a33 (hereinafter referred to as A-side second photodetector a33).
The A plane side BS (a31) is a position immediately before the first beam light P1 and the second beam light P2 guided in the direction of the A plane measurement point 1a are input to the A plane heterodyne interferometer a20. In FIG. 2, the light beam (hereinafter referred to as main light) input to the A-side heterodyne interferometer a20 and the other light beam (hereinafter referred to as sub-light) are branched (main surface side main sub-light). Spectroscopic means).
The A-plane-side second polarizing plate a32 is branched by the A-plane-side BS (a31) by transmitting only light having a polarization plane in a predetermined direction (a middle direction between the polarization planes of both light beams). The auxiliary light (that is, the branched first beam light P1 and second beam light P2) is caused to interfere (sub-light interference means on the front surface side).
The A-side second photodetector a33 receives the interference light obtained by the A-side second polarizing plate a32 and performs photoelectric conversion, thereby outputting an intensity signal Ref1 (electric signal) of the interference light. (Sub-light intensity detecting means on the front side).

The B-side correction interferometer b30 has the same configuration as the A-side correction interferometer a30 on the B side of the DUT 1.
That is, as shown in FIG. 2, the B-side correction interferometer b30 includes a beam splitter b31 (hereinafter referred to as B-side BS (b31)) and a polarizing plate b32 (hereinafter referred to as B-side second polarizing plate b32). And a photodetector b33 (hereinafter referred to as a B-side second photodetector b33).
The B surface side BS (b31) is a position immediately before the first beam light P1 and the second beam light P2 guided in the direction of the B surface measurement point 1b are input to the B surface heterodyne interferometer b20. In FIG. 2, the light beam (main light) input to the B-side heterodyne interferometer b20 is branched into the other light beam (sub-light) (back side main sub-spectral means).
The B-side second polarizing plate b32 is branched by the B-side BS (b31) by transmitting only light having a polarization plane in a predetermined direction (the intermediate direction between the polarization planes of both light beams). The auxiliary light (that is, the branched first beam light P1 and second beam light P2) is caused to interfere (sub-light interference means on the back surface side).
The B-side second photodetector b33 receives the interference light obtained by the B-side second polarizing plate b32 and performs photoelectric conversion to output an intensity signal Ref2 (electric signal) of the interference light. (Sub-light intensity detecting means on the back side).

The second phase detector 5 detects the phase difference ΔΦr between the intensity signals Ref1 and Ref2 output from the A-side second photodetector a33 and the B-side second photodetector b33, respectively. The detection value is output as an electric signal (detection signal), and, for example, a lock-in amplifier can be employed. Here, the value (phase difference ΔΦr) of the output signal of the second phase detector 5 is a measured value for correcting the thickness of the DUT 1 (corresponding to the second measured value), as will be described later. It is.
The first beam light P1 and the second beam light P2 are guided from the dual-polarized light source 2 to the A-plane measurement point 1a and the B-plane measurement point 1b by the PBS 3 and mirrors a11 to a13, b11, and b12, respectively. When phase fluctuation occurs in the path, the influence of the fluctuation is reflected in the phase difference ΔΦs detected by the first phase detector 4, which becomes a measurement error.
However, the sum of the phase fluctuation components on the A plane side and the B plane side is reflected in the phase difference ΔΦr detected by the second phase detector 5. Therefore, the corrected phase data ΔΦsr (= ΔΦs) obtained by performing correction by subtracting the phase difference ΔΦr detected by the second phase detector 5 from the phase difference ΔΦs detected by the first phase detector 4. −ΔΦr) is measurement data from which the influence of the phase fluctuations of the two light beams P1 and P2 up to the two heterodyne interferometers a20 and b20 is removed.
Here, the corrected phase data ΔΦsr changes according to the thickness of the DUT 1, but ΔΦs and ΔΦr are values detected within a range of −π to + π, and therefore, two adjacent measurement points are measured. Even if the difference in thickness (shape change) is small, the difference between the corrected phase data ΔΦsr obtained at these two measurement points may be large. Therefore, phase data (hereinafter, referred to as “phase data”) that continuously changes in accordance with a continuous thickness change of the DUT 1 by applying a phase connection process to the corrected phase data ΔΦsr obtained at a plurality of measurement points. It is necessary to calculate post-processing phase data θ).

  Therefore, the computer 6 obtains a plurality of sets of phase differences ΔΦs and ΔΦr through the phase detectors 4 and 5 for a plurality of measurement points 1a and 1b, and calculates the corrected phase data ΔΦsr therefrom. Further, the computer 6 performs a phase connection process based on the corrected phase data ΔΦsr obtained for each of the plurality of measurement points 1a and 1b. Each of the plurality of post-processing phase data θ obtained by this phase connection processing is data representing the thickness of the device under test 1 at each measurement point, and is data representing the thickness distribution of the device under test 1 as a whole. Further, the calculator 6 outputs the processed phase data θ (data representing the thickness distribution of the DUT 1) calculated by the phase connection process. The data output is, for example, writing to a storage unit (hard disk or the like) included in the computer 6, transmitting to an external device through a predetermined communication interface, or a calculated value on a predetermined display unit such as a liquid crystal display device. Means to display the information.

Next, the measurement principle of the shape measuring apparatus X will be described using mathematical expressions.
First, symbols used in mathematical expressions will be described.
ω: frequency of the first beam light L1.
ω + Δω: frequency of the second beam light L2.
λ: wavelength of the first beam light L1. Since the value of Δω / ω is very small, it can be approximated that λ is equal to the wavelength of the second beam light L2.
L1: Optical path length from the A-side PBS (a21) to the A-side measurement point 1a.
L2: Optical path length from the B-side PBS (b21) to the B-side measurement point 1b.
M1: Optical path length from the A-side PBS (a21) to the A-side reference plate a24 surface.
M2: optical path length from the B-side PBS (b21) to the B-side reference plate b24 surface.
ΔL1: Surface displacement amount (surface height) based on the shape of the A-plane measurement point 1a.
ΔL2: Surface displacement amount (surface height) based on the shape of the B surface measurement point 1b.
ΔN: A displacement amount due to vibration of the DUT 1.
Note that t represents time and i represents a natural number variable.

また，前記Ｂ面側ヘテロダイン干渉計ｂ２０の出力信号Ｓｉｇ２の強度Ｉ２は，次の（２）式により表される。

ここで，被測定物１の振動に起因する変位量ΔＮは，前記Ａ面側と前記Ｂ面側とで正負が逆となって影響し，さらに，前記Ａ面側と前記Ｂ面側とで，ヘテロダイン干渉計における物体光と参照光との対応関係が逆になっているため，この（１）式及び（２）式において，変位量ΔＮの符号は同じとなる。
前記（１）式及び（２）式（最終行に記載の式）において，"４π／λ"以降の項が，各周波数Δωの周期変化の位相を決定する。そして，前記入力光に位相揺らぎの差がない状態における前記第１位相検波器４により検出される出力信号Ｓｉｇ１，Ｓｉｇ２の位相差Φs’は，次の（３）式により表される。

前記（３）式において，Ａ面側及びＢ面側の相殺効果により，被測定物１の振動に起因する変位量ΔＮが除去（相殺）されていることがわかる。 Before the measurement, the shape measuring apparatus X performs beam light P1, P2 (that is, in the process from the dual-polarized light source 2 to the A-side correction interferometer a30 and the B-side correction interferometer b30). , The input light to both heterodyne interferometers a20 and b20) is set to a state in which there is no difference in phase fluctuation (hereinafter referred to as a state in which there is no difference in phase fluctuation in the input light), or a state that can be approximated. It shall be.
As a method of making the input light have no phase fluctuation difference or a state that can be approximated, for example, both beam lights P1 and P2 guided in the direction of the A-plane measurement point 1a and the B-plane measurement point 1b. It is conceivable that the optical paths of the two light beams P1 and P2 are kept constant as much as possible by providing the same optical path as much as possible, or providing a heat insulating material covering the periphery of the optical path. Further, a light guide means such as a mirror or an optical fiber for guiding both light beams P1 and P2 to the A-surface measurement point 1a and the B-surface measurement point 1b is fixed to a fixed and robust (high-strength) support member. Is also effective.
First, the intensity I1 of the output signal Sig1 of the A-plane side heterodyne interferometer a20 is expressed by the following equation (1).

The intensity I2 of the output signal Sig2 of the B-side heterodyne interferometer b20 is expressed by the following equation (2).

Here, the amount of displacement ΔN caused by the vibration of the DUT 1 is affected by the opposite in the positive and negative directions on the A surface side and the B surface side, and further on the A surface side and the B surface side. Since the correspondence relationship between the object beam and the reference beam in the heterodyne interferometer is reversed, the sign of the displacement amount ΔN is the same in the equations (1) and (2).
In the equations (1) and (2) (the equation in the last row), the term after “4π / λ” determines the phase of the period change of each frequency Δω. Then, the phase difference Φs ′ of the output signals Sig1 and Sig2 detected by the first phase detector 4 in a state where there is no phase fluctuation difference in the input light is expressed by the following equation (3).

In the equation (3), it can be seen that the displacement amount ΔN due to the vibration of the DUT 1 is removed (cancelled) by the canceling effect on the A plane side and the B plane side.

In the equation (3), ΔΦs ′ is a measured value, and other numerical values other than (ΔL1−ΔL2) are known invariants. Therefore, if there is no phase fluctuation difference in the input light or a state that can be approximated, the calculation 6 applies the measurement value ΔΦs to the above (3) (substitute for ΔΦs ′). By doing so, the thickness (ΔL1-ΔL2) of the DUT 1 can be calculated.
Note that (ΔL1−ΔL2) does not represent the absolute value of the thickness of the DUT 1, but is an index for evaluating a relative value with respect to the thickness of other measurement points (a value indicating a relative thickness). In measuring the shape of an object to be measured such as a semiconductor wafer, it is important to obtain a distribution of values representing relative thicknesses.

By the way, the displacement ΔN caused by the vibration of the device under test 1 is sufficiently larger than the wavelength λ of the light beams P1 and P2, so that if the influence of the displacement ΔN is not removed, the thickness of the device under test 1 is substantially reduced. Cannot be measured automatically.
For example, when the phase detector detects the phase difference ΔΦa between the detection signal Sig1 of the A-plane side first photodetector a26 and the detection signal Ref1 of the A-plane side second photodetector a33, each measurement point In the distribution of the phase difference ΔΦa at, so-called phase jumps occur many times, and it is difficult to accurately obtain a distribution of continuous values representing the thickness distribution of the DUT 1.

  As described above, in the shape measuring apparatus X, the component of the displacement amount ΔN due to the vibration of the DUT 1 is canceled out in the measured value Φs, and the shape component of the A plane measurement point 1a itself and the B plane measurement A measurement value corresponding to the thickness of the DUT 1 reflecting only the component having the shape of the point 1b itself is obtained. Moreover, since the shape measuring apparatus X does not propagate the interference light from the A surface (front surface) of the DUT 1 to the B surface (back surface), it is necessary to adjust the optical path in the propagation path (adjustment of optical equipment). In addition, the interference light does not fluctuate in the propagation path. Furthermore, since nothing other than the optical system is inserted in the optical path, the interference light is not disturbed.

By the way, the beam lights P1 and P2 input to the heterodyne interferometers a20 and b20 are respectively transmitted from the dual-polarized light source 2 to the A-plane side correction interferometer a30 and the B-plane side correction interferometer b30. If a phase fluctuation occurs due to changes in temperature, humidity, etc., the influence of the fluctuation is reflected in the measured value ΔΦs (phase difference), which becomes a measurement error.
On the other hand, the phase difference (ΔΦs−ΔΦr) obtained by subtracting the correction phase difference ΔΦr from the phase difference ΔΦs is the fluctuation of the phases of the two light beams P1 and P2 up to the two heterodyne interferometers a20 and b20. This is a measurement value from which the influence of is removed.
Therefore, the computer 6 calculates the thickness (ΔL1−) of the DUT 1 based on the following equation (4) in which ΔΦs ′ in the equation (3) is replaced with the corrected phase data ΔΦsr (= ΔΦs−ΔΦr). ΔL2) is calculated. The calculated value is a value that is not affected by the fluctuation of the phases of the two light beams P1 and P2.

Next, an example of the procedure for measuring the shape of the DUT 1 using the shape measuring device X will be described with reference to the flowchart shown in FIG. It is assumed that the device under test 1 is held at a predetermined initial position by the movable support device Z before the processing shown in FIG. Further, S1, S2,... Shown below represent identification codes of processing procedures.
First, the computer 6 initializes a variable i representing a measurement point number (i = 1), and outputs a measurement start command to the dual-polarized light source 2 and the movement control device 7 (S1). . Thereby, when the emission of the light beams P1 and P2 by the dual-polarized light source 2, that is, the irradiation of the object light to the object to be measured 1 is started, the object to be measured 1 moves in the two-dimensional direction. Be started. As a result, scanning of the object light along the scanning line R (see FIG. 4) is started on the front and back surfaces of the DUT 1.
Thereafter, steps S2 to S9 shown below are sequentially repeated while the variable i representing the number of the measurement point is sequentially counted up (S10).

Subsequently, the calculator 6 acquires the first phase difference ΔΦs and the second phase difference ΔΦr, the amplitudes As1, As2, and the position data of the measurement points at the i-th measurement point (S2). Further, the calculator 6 stores the corrected phase data ΔΦsr, which is the difference between the first phase difference ΔΦs and the second phase difference ΔΦr, in the memory 6b together with the acquired amplitudes As1, As2 and position data of the measurement points. Record (S2).
In this way, every time the phase of the interference light intensity signal is detected (each time the corrected phase data is acquired), the amplitudes of the interference light intensity signals Sig1 and Sig2 that are detection targets are sequentially detected. The computer 6 acquires a plurality of amplitude data As1 and As2 obtained by the detection (S2, amplitude data acquisition procedure).
The first phase difference ΔΦs and the amplitudes As1 and As2 are data acquired from the first phase detector 4, and the position data of the measurement point is data acquired from the movement control device 7. is there.

Next, the calculator 6 obtains an allowable range of each of the amplitudes As1 and As2 at the i-th measurement point (hereinafter referred to as an amplitude allowable range) at one or a plurality of other measurement points around the i-th measurement point. The amplitudes As1 and As2 are set as a reference (S3, amplitude allowable range setting procedure). The allowable amplitude range is individually set for each of the measurement points 1a on the A plane side and the measurement points 1b on the B plane side.
For example, the calculator 6 sets the allowable amplitude range for the i-th measurement point back in the direction opposite to the measurement order along the scanning line R among the measurement points within the predetermined range from the i-th measurement point. A plurality of measurement points (for example, (i-3) th to (i-1) th measurement points) at the selected position are selected. Further, the calculator 6 calculates a weighted average value of the amplitudes As1 and As2 obtained at a plurality of selected measurement points, and sets a lower limit value setting coefficient and an upper limit value setting coefficient that are predetermined for the weighted average value. The values obtained by multiplying them are set as the lower limit value AL and the upper limit value AH of the amplitude allowable range. In the weighted average, a high weight is set to data at a measurement point closer to the i-th measurement point. Alternatively, one measurement point ((i-1) th measurement point) located at an adjacent position retroactive along the scanning line R with respect to the i-th measurement point is selected, and the measurement point obtained at the measurement point is selected. A value obtained by multiplying the amplitudes As1 and As2 by the lower limit value setting coefficient and the upper limit value setting coefficient may be set as the lower limit value AL and the upper limit value AH of the allowable amplitude range.

FIG. 5 is a schematic diagram showing changes in the amplitude signal of the interference light obtained by measuring the DUT 1 having the singular part.
As shown in FIG. 5, the amplitude signal of the interference light in the singular part has a large difference from the amplitude obtained at the peripheral positions. In the present invention, using this phenomenon, the phase data obtained at the measurement point in the singular part (the data to be processed) and the phase data obtained at other measurement points (the data to be processed) Is divided.
Further, in FIG. 5, the allowable amplitude range (AL to AH) at the (Kb + 1) th measurement point set based on the amplitude data at the Kbth measurement point is indicated by a wavy line. Note that the measurement points shown in FIG. 5 correspond to the measurement points shown in FIG. 4, and the (Kb + 1) th and (Kb + 2) th measurement points are the measurement points existing in the singular part.
In this way, the calculator 6 determines the amplitude allowable range corresponding to each of the plurality of corrected phase data ΔΦsr obtained through the first phase detector 4 and the second phase detector 5 as the corrected phase data. The amplitude data As1 and As2 obtained at other measurement points around the measurement point corresponding to ΔΦsr are dynamically set (S3).

Next, the computer 6 determines whether or not the amplitude data As1 and As2 obtained in step S2 are within the amplitude allowable range set in step S3 for the i-th measurement point. According to the result, the corrected phase data ΔΦsr is classified into either processing target data or non-processing target data, and the result of the classification is recorded in the memory 6b (S4, data classification procedure).
By the process of step S4, a plurality of the corrected phase data ΔΦsr sequentially obtained by the process of step S2, the amplitude target data As1 and As2 corresponding to the data, the processing target data within the amplitude allowable range, It is divided into other non-processing target data.

Next, the computer 6 performs a phase connection process based only on a plurality of the corrected phase data ΔΦsr obtained by repeating the process of step S2 that is classified as the process target data in step S4. Thus, a process of calculating the shape value representing the thickness of the DUT 1, that is, the post-processing phase data θ is executed (S5 to S9). The contents will be described below.
First, the computer 6 switches processing depending on whether the corrected phase data ΔΦsr at the i-th measurement point is the processing target data or the non-processing target data (S5).
That is, when the corrected phase data ΔΦsr at the i-th measurement point is the non-processing target data, the computer 6 does not perform the phase connection process on the corrected phase data ΔΦsr and counts up the variable i (S10). Only do.

When the corrected phase data ΔΦsr at the i-th measurement point is the processing target data, the calculator 6 determines whether the corrected phase data ΔΦsr at the (i−1) -th measurement point is the processing target data. Is determined (S6).
When both the corrected phase data ΔΦsr at both the (i−1) -th measurement point and the i-th measurement point that are adjacent along the scanning line R are the processing target data, the calculator 6 1) The phase data at the measurement point (data θ after the phase connection process) is set as the reference phase data for the phase connection process (S7). Here, the reference phase data is data used as a reference to be compared with the phase data (here, the corrected phase data ΔΦsr at the i-th measurement point) that is the target of the phase connection process in the phase connection process. It is. That is, at the time of phase connection processing in step S9 described later, the corrected phase data ΔΦsr at the i-th measurement point is an integral multiple of 2π so that the phase difference with respect to the reference phase data falls within the range of −π to + π. The phase is shifted by this amount, and the data after the phase shift is calculated as the post-processing phase data θ.

On the other hand, the corrected phase data ΔΦsr at the i-th measurement point is the data to be processed, and the corrected phase at the (i−1) -th measurement point adjacent to the measurement point along the scanning line R. When the data ΔΦsr is the non-processing target data, the calculator 6 includes the data classified into the processing target data among the phase data for the measurement points in the vicinity of the i-th measurement point (however, after the phase connection processing) Is set as the reference phase data (S8).
For example, the computer 6 is divided into the processing target data from the phase data at the measurement points adjacent to the i-th measurement point in the direction crossing the scanning line R (however, the phase connection process) The later data θ) is set as the reference phase data.
Then, the calculator 6 performs a phase connection process based on the reference phase data set in step S7 or S8 on the corrected phase data ΔΦsr at the i-th measurement point, and converts the processed phase data θ to i The shape value representing the thickness at the th measurement point is recorded in the memory 6b (S10).

FIG. 6 is a schematic diagram in which the data θ after the phase connection process at the singular part of the device under test 1 and the measurement points in the vicinity thereof calculated by the computer 6 is graphed. Note that the measurement points shown in FIG. 6 correspond to the measurement points shown in FIG. 4, and the (Kb + 1) th and (Kb + 2) th measurement points are the measurement points existing in the singular part. FIG. 10 is a schematic diagram in which data θ ′ after conventional phase connection processing at the same measurement point is graphed.
As shown in FIG. 6, in the shape measuring apparatus X, the measurement points ((Kb + 1) th and (Kb + 2) th) in the singular part are discriminated based on the amplitude, and the phase data at the measurement points in the singular part. Is not subjected to the phase connection process (the process of step S5 in FIG. 3).
Further, if phase connection processing is performed in the normal order along the scanning line R, the phase data corresponding to the measurement point ((Kb + 2) th) in the singular part becomes the reference phase data. For (Kb + 3) th), a phase connection process is performed. In this case, the phase data of the Kath measurement point (see FIG. 4) adjacent to the (Kb + 3) th measurement point in the direction crossing the scanning line R (however, the phase connection divided into the processing object data) (Data after processing) is set as the reference phase data, and phase connection processing of the phase data of the (Kb + 3) th measurement point is performed using the reference phase data as a reference (processing in steps S8 and S9 in FIG. 3). .
Then, the computer 6 counts up the variable i representing the number of the measurement points (S10), and repeats the processes of steps S2 to S9 described above until the measurement for all the measurement points is completed (S11). ).

As described above, the computer 6 in the shape measuring apparatus X uses the plurality of measurement points 1a and 1b of the DUT 1 based on the amplitudes As1 and As2 (that is, the signal strength) of the intensity signal of the interference light. Each of the phase data ΔΦsr obtained for each is divided into the non-processing object data obtained in the singular part, such as a dust adhering part or a scratched part, and the other processing object data (S3, S3). S4). This utilizes a phenomenon in which a difference in the reflection characteristics of light (object light) on the surface of the DUT 1 appears as a difference in the amplitudes As1 and As2 of the intensity signals of the interference light.
Then, the computer 6 executes the phase connection process based only on the phase data ΔΦsr (the processing target data) obtained at the measurement points other than the singular part (S5 to S9). As a result, the calculator 6 can calculate the shape value θ with high accuracy even for the DUT 1 in which the singular part having a different light reflection characteristic from others exists.

Further, the calculator 6 corresponds the amplitude tolerance range corresponding to each of the plurality of phase data ΔΦsr obtained through the phase detectors 4 and 5 (that is, the amplitude tolerance range for each measurement point) to the phase data ΔΦsr. Is dynamically set based on the amplitude data As1 and As2 obtained at one or a plurality of other measurement points around the measurement point (S3, amplitude allowable range setting procedure).
As a result, the accurate amplitude allowable range is dynamically set for each phase data ΔΦsr (that is, for each measurement site), and the phase data ΔΦsr can be properly classified, that is, the singular part can be specified appropriately.

FIG. 7 is a graph of actual measurement data obtained by measuring a semiconductor wafer (an example of the object 1 to be measured 1) with dust attached to a part thereof using the shape measuring apparatus X. The measured data are the amplitude As1 of the intensity signal Sig1 of the interference light and the phase data θ after the phase connection processing calculated by the calculator 6. In FIG. 7, the phase data θ ′ after the conventional phase connection process that does not exclude the measurement data at the singular part is also graphed as a comparison target.
As shown in FIG. 7, in the situation where the singular part does not exist, the data θ after the phase connection processing obtained by the shape measuring apparatus X and the data θ ′ after the conventional phase connection processing have the same result.
However, when the phase data obtained at the singular part is referred to during the phase connection process, the data calculated for the measurement points to be subjected to the subsequent phase connection process are the processing result θ and the shape measurement device X. There is a large difference with the result θ ′ of the conventional phase connection processing. In this case, the processing result θ by the shape measuring apparatus X represents the actual shape of the wafer to be measured. As described above, according to the shape measuring apparatus X, it is possible to calculate the shape value θ with high accuracy for the DUT 1 in which the singular part having a different light reflection characteristic is present in part.

In the embodiment described above, an example has been described in which the calculator 6 uses the difference between the measured value ΔΦs and the measured value ΔΦr as the phase data and performs phase connection processing based on the phase data. However, when the input light to each of the optical interferometers a20 and b20 has no phase fluctuation difference or a state that can be approximated, the A-side correction interferometer a30 and A shape measuring device (hereinafter referred to as a shape measuring device X ′) excluding the B-side correction interferometer b30 may be considered.
Then, the computer 6 in the shape measuring apparatus X ′ uses the phase difference ΔΦs, which is a measurement value output from the first phase detector 4, as the phase data, performs phase connection processing based on the phase data, The processing result is calculated as a shape value representing the thickness of the DUT 1.

In the shape measuring apparatus X, the heterodyne interferometers a20 and b20 shown in FIG. 2 allow measurement light to enter substantially perpendicular to the surface (A surface, B surface) of the DUT 1. Even if an interferometer of a type in which the measurement light is incident obliquely with respect to the surface (A surface, B surface) of the object to be measured 1 and the interference light is obtained by the reflected light by changing the optical system is adopted. Good.
FIG. 9 is a schematic configuration diagram of a grazing incidence type heterodyne interferometer (hereinafter referred to as an A-side heterodyne interferometer a20 ′ and a B-side heterodyne interferometer b20 ′) that can be employed in the shape measuring apparatus X.
As shown in FIG. 9, the A-side heterodyne interferometer a20 ′ includes the A-side PBS (a21), the A-side reference plate a24, a beam splitter a28, and the A-side first polarization. A plate a25 and the A-side first photodetector a26 are provided.
In the A-plane side heterodyne interferometer a20 ′, the A-plane side PBS (a21) causes the first beam P1 transmitted therethrough to be inclined with respect to the surface (plane) of the A-plane side measurement point 1a. It arrange | positions so that it may inject.
The A-side PBS (a21) transmits the first beam P1 guided by an optical path different from that of the second beam P2, thereby transmitting the first beam P1 to the A-plane measurement point. 1a is irradiated (obliquely incident) and the second beam light P2 is reflected to irradiate the surface (first reference surface) of the A-plane side reference plate a24 with the second beam light P2 ( Obliquely incident).
The beam splitter a28 transmits the reflected light (regularly reflected light) of the first beam light P1 from the A-plane measurement point 1a in the direction of the A-plane side first polarizing plate a25, and the A-plane. The reflected light (regular reflected light) of the second beam light P2 from the surface of the side reference plate a24 is reflected in the direction of the A-plane side first polarizing plate a25. As a result, the reflected light of the first beam light P1 from the A-plane measurement point 1a transmitted through the beam splitter a28 and the second light from the surface of the A-plane side reference plate a24 reflected by the beam splitter a28. The reflected light of the light beam P2 interferes with each other by passing through the first A-side polarizing plate a25 while traveling along the same optical path (in a state where the optical axes overlap). The interference light (the A surface measurement interference light) is input (incident) into the A surface side first photodetector a26, and an intensity signal Sig1 of the A surface measurement interference light is obtained.

Similarly, the B-side heterodyne interferometer b20 ′ includes the B-side PBS (b21), the B-side reference plate b24, a beam splitter b28, the B-side first polarizing plate b25 and the B-side. The surface side 1st photodetector b26 is provided.
In the B-side heterodyne interferometer b20 ', the B-side PBS (b21) causes the second beam light P2 transmitted therethrough to be inclined with respect to the surface (plane) of the B-side measurement point 1b. It arrange | positions so that it may inject.
The B-side PBS (b21) transmits the second beam light P2 guided by an optical path different from that of the first beam light P1, thereby transmitting the second beam light P2 to the B-surface measurement point. 1b is irradiated (obliquely incident) and the first beam light P1 is reflected to irradiate the surface (second reference surface) of the B-side reference plate b24 with the first beam light P1 (second reference surface). Obliquely incident).
The beam splitter b28 transmits reflected light (regularly reflected light) of the second beam light P2 from the B-surface measurement point 1b in the direction of the B-surface side first polarizing plate b25, and the B-surface. The reflected light (regular reflected light) of the first beam light P1 from the surface of the side reference plate b24 is reflected in the direction of the B-side first polarizing plate b25. Thereby, the reflected light of the second beam light P2 from the B surface measurement point 1b that has passed through the beam splitter b28 and the first surface from the surface of the B surface side reference plate b24 reflected by the beam splitter b28. The reflected light of the light beam P1 interferes with each other by passing through the first B-side polarizing plate b25 while traveling along the same optical path (with the optical axes overlapping). The interference light (the B surface measurement interference light) is input (incident) to the B surface side first photodetector b26, and an intensity signal Sig2 of the B surface measurement interference light is obtained.
As shown in FIG. 9, a shape measuring apparatus employing a heterodyne interferometer that obliquely enters measurement light with respect to the DUT 1 is also an example of an embodiment of the present invention.

In the above-described embodiment, as shown in FIG. 4, the example in which the positions of the measurement points 1 a and 1 b sequentially change along the spiral scanning line R on the surface of the DUT 1 has been described.
However, for example, as the movable support device Z, a device that moves the support portion 44 of the DUT 1 along two intersecting straight lines such as an XY plotter may be used. In this case, it is also conceivable to sequentially change the positions of the measurement points 1a and 1b along the scanning line R as shown in FIG. FIG. 8 shows an example in which the (Kb + 1) th and (Kb + 2) th measurement points exist in the singular part.

  Further, the shape measuring apparatus X is provided with a pair of optical interferometers a20 and b20 that irradiate object light on the measurement points 1a and 1b opposite to each other on the object 1 to measure the thickness of the object 1 to be measured. However, a shape measuring apparatus in which only one optical interferometer a20 is provided is also conceivable. In this case, it is conceivable that the first phase detector 4 receives the two interference light intensity signals Sig1 and Ref1, and detects the phase difference between them as the phase data used in the phase connection process. In this case, the data calculated by the phase connection process based on the phase data obtained by the first phase detector 4 is data representing the height of the A surface of the DUT 1.

  In the shape measuring apparatus X, a heterodyne interferometer is employed. In addition, one laser beam (single-wavelength laser beam) is branched into object light and reference light, and reflected light and reference light of the object light. It is also conceivable that other general interferometers that output the intensity signal of the interference light that interferes with each other, such as a Michelson interferometer, a Fizeau interferometer, or an oblique incidence interferometer, may be employed.

  The present invention is applicable to a shape measuring apparatus for an object to be measured such as a semiconductor wafer.

The schematic block diagram of the shape measuring apparatus X which concerns on embodiment of this invention. The block diagram of the interference light detection part Y with which the shape measuring apparatus X is provided. 5 is a flowchart showing an example of a shape measuring procedure of an object to be measured by the shape measuring apparatus X. The schematic diagram showing an example of distribution of the measurement point in a to-be-measured object. The schematic diagram showing the change of the amplitude signal of the intensity signal of the interference light obtained by measuring the to-be-measured object which has a specific part. The schematic diagram by which the data after the phase connection process in the measurement point of the circumference | surroundings of the singular part of the to-be-measured object computed by the shape measuring apparatus X was graphed. The graph of the measurement data of the to-be-measured object which has a specific part by the shape measuring apparatus X. The schematic diagram showing another example of distribution of the measurement point in a to-be-measured object. 2 is a schematic configuration diagram of an oblique incidence interferometer applicable to the shape measuring apparatus X. FIG. The schematic diagram by which the phase data after the phase connection process obtained by the conventional shape measurement which used the optical interferometer about the to-be-measured object which has a specific part, and the phase data showing an original shape were graphed.

Explanation of symbols

X: shape measuring device Y according to an embodiment of the present invention Y: interference light detection unit Z: movable support device 1: object to be measured 1a: A surface measurement point 1b: B surface measurement point 2, 2 ′: dual polarized light source 3 : Polarizing beam splitter 4: First phase detector 5: Second phase detector 6: Computer 7: Movement control devices a11 to a13, b11, b12: Mirrors a20, a20-1 to a20-4: A-plane side heterodyne interference Total a30, a30-1, a30-2: A-side correction interferometers b20, b20-1 to b20-4: B-side heterodyne interferometers b30, b30-1, b30-2: B-side correction interference Total P1: First beam light P2: Second beam light R: Scan lines S1, S2,...: Processing procedure (step)

Claims (7)

Interference light phase detection for detecting the phase of the intensity signal of the interference light in the optical interferometer while changing the relative position in the two-dimensional direction of the object to be measured and the optical interferometer that irradiates the measurement site on the surface with the object light A shape calculation device for calculating a shape value of the object to be measured based on phase data for each of a plurality of measurement sites obtained by the device,
Amplitude data acquisition means for acquiring a plurality of amplitude data obtained by detecting the amplitude of the intensity signal of the interference light that is the detection target for each detection of the phase;
Data that classifies the phase data obtained by the interference light phase detection device into processing target data in which the amplitude data corresponding to the phase data is within a preset allowable amplitude range and other non-processing target data Classification means;
Phase connection means for calculating a shape value of the object to be measured by performing phase connection processing based only on the processing target data among the phase data obtained by the interference light phase detection device;
A shape calculation apparatus comprising: The phase connecting means comprises:
The two processing objects that are adjacent to each other along a predetermined scanning line that sequentially extends from the measurement site that is the starting point of the plurality of measurement sites corresponding to the plurality of phase data to the next measurement site. Perform phase connection processing based on data sequentially,
When the measurement part of the non-processing target data and the measurement part of the processing target data are adjacent to each other along the predetermined one scanning line, the measurement part is relative to the processing target data and the processing target data. The shape calculation apparatus according to claim 1, wherein a phase connection process based on the other data to be processed adjacent to each other in a direction crossing the predetermined scanning line is executed.   The amplitude data obtained in one or a plurality of other measurement sites around the measurement site corresponding to the phase data, the amplitude tolerance range corresponding to each of the plurality of phase data obtained by the interference light phase detection device The shape calculation apparatus according to claim 1, further comprising an amplitude allowable range setting unit that dynamically sets the reference value. A shape measuring device including an optical interferometer that irradiates a measurement site on the surface of an object to be measured with object light,
Movable support means for changing a relative position in a two-dimensional direction between the object to be measured and the optical interferometer;
Phase detection means for obtaining phase data by detecting a phase of an intensity signal of interference light in the optical interferometer according to a change in a relative position between the object to be measured and the optical interferometer by the movable support means;
The shape calculation device according to any one of claims 1 to 3, wherein the phase data for each of a plurality of measurement sites obtained by the phase detection means is acquired, and a shape value of the object to be measured is calculated based on the phase data. When,
A shape measuring apparatus comprising: The first measurement light and the second measurement light that are emitted from a predetermined light source and having different frequencies are branched to measure the front surface measurement region and the back surface of the measurement object. A light guiding means for guiding each direction of the measurement site;
One of the pair of optical interferometers, the first measurement light guided in the direction of the measurement part on the front surface is irradiated to the measurement part on the front surface as object light, and The second measurement light guided in the direction of the measurement part on the front surface is irradiated onto the first reference surface, and the reflected light of the first measurement light from the measurement part on the front surface and the first A heterodyne optical interferometer on the front surface side that interferes with the reflected light of the second measurement light from the reference surface of 1 and outputs an intensity signal of the interference light;
The second measurement light, which is the other of the pair of optical interferometers and guided in the direction of the measurement part of the back surface, is irradiated as object light to the measurement part of the back surface, and the measurement part of the back surface The second reference surface is irradiated with the first measurement light guided in the direction of, and the reflected light of the second measurement light from the measurement part of the back surface and the first reference light from the second reference surface A backside heterodyne interferometer that interferes with the reflected light of the measurement light of 1 and outputs an intensity signal of the interference light,
The phase detection means comprises:
The result of detecting the phase difference between the intensity signals output from each of the two heterodyne interferometers according to the change in the relative position between the object to be measured and the two heterodyne optical interferometers by the movable support means is the phase. As data,
The shape calculation device acquires the phase data for each of a plurality of measurement sites obtained by the phase detection means, and calculates a shape value representing the thickness distribution of the surface of the object to be measured based on the phase data. The shape measuring apparatus according to claim 4. Interference light phase detection for detecting the phase of the intensity signal of the interference light in the optical interferometer while changing the relative position in the two-dimensional direction of the object to be measured and the optical interferometer that irradiates the measurement site on the surface with the object light A shape calculation program for causing a computer to execute a procedure for calculating a shape value of the object to be measured based on phase data for each of a plurality of measurement sites obtained by an apparatus,
An amplitude data acquisition procedure for acquiring a plurality of amplitude data obtained by detecting the amplitude of the intensity signal of the interference light that is the detection target for each detection of the phase;
Data that classifies the phase data obtained by the interference light phase detection device into processing target data in which the amplitude data corresponding to the phase data is within a preset allowable amplitude range and other non-processing target data Classification procedure;
A phase connection procedure for calculating a shape value of the device under test by performing phase connection processing based only on the processing target data among the phase data obtained by the interference light phase detection device;
Calculation program for causing a computer to execute. Interference light phase detection for detecting the phase of the intensity signal of the interference light in the optical interferometer while changing the relative position in the two-dimensional direction of the object to be measured and the optical interferometer that irradiates the measurement site on the surface with the object light A shape calculation method for executing, by a computer, a procedure for calculating a shape value of the object to be measured based on phase data for each of a plurality of measurement sites obtained by an apparatus,
An amplitude data acquisition procedure for acquiring a plurality of amplitude data obtained by detecting the amplitude of the intensity signal of the interference light that is the detection target for each detection of the phase;
Data that classifies the phase data obtained by the interference light phase detection device into processing target data in which the amplitude data corresponding to the phase data is within a preset allowable amplitude range and other non-processing target data Classification procedure;
A phase connection procedure for calculating a shape value of the device under test by performing phase connection processing based only on the processing target data among the phase data obtained by the interference light phase detection device;
A shape calculation method for executing the above by a computer.

Images