JP6371926B1 - Optical measuring apparatus and optical measuring method - Google Patents

Abstract

An optical measurement apparatus and an optical measurement method are provided that can measure the in-plane distribution of film thicknesses of various samples at higher speed and with higher accuracy.
An optical measurement apparatus includes an irradiation optical system that linearly irradiates measurement light having a predetermined wavelength range to a measurement target, and a straight line that is transmitted light or reflected light generated from the measurement target by irradiation of the measurement light. A measurement optical system that receives the measurement interference light in the form of a shape and a processing device. The measurement optical system picks up the measurement interference light in a direction orthogonal to the longitudinal direction of the measurement interference light, and receives the measurement interference light expanded in wavelength by the diffraction grating and outputs a two-dimensional image. Part. The processing device calculates a correction element corresponding to an incident angle from each measurement point to the measurement optical system in association with a region on the two-dimensional image corresponding to each measurement point to be measured irradiated with measurement light. And a second calculation unit that calculates an optical characteristic of the measurement target after applying a corresponding correction element to each pixel value included in the two-dimensional image.
[Selection] Figure 12

Description

  The present invention relates to an optical measurement apparatus and an optical measurement method capable of measuring optical characteristics such as film thickness and refractive index.

  A technique for measuring the film thickness of a sample such as a functional resin film or a semiconductor substrate is known. For example, Japanese Patent Laying-Open No. 2009-092454 (Patent Document 1) discloses a multilayer film analysis apparatus and a multilayer film analysis method capable of measuring the film thickness of a multilayer film sample having wavelength dependency with higher accuracy. . Japanese Unexamined Patent Publication No. 2013-079921 (Patent Document 2) discloses a film thickness measuring apparatus and a film thickness measuring method capable of accurately measuring the film thickness of a dielectric thin film whose refractive index is unknown.

  In general, a sample to be measured has a certain area, and there is a need to measure the film thickness distribution (film thickness in-plane distribution) on the measurement target surface at high speed. In response to such needs, Japanese Patent Application Laid-Open No. 2004-279296 (Patent Document 3) describes the thickness of a thin film formed when a thin film is formed on a flat plate in a manufacturing process of a liquid crystal display device or the like. A method for obtaining a distribution at high speed with a simple device configuration is disclosed. More specifically, in Patent Document 3, incident light from a light source is incident on a film provided on a substrate to be measured, and reflected light that causes interference from the film is reflected on the main surface of the film. A method is disclosed in which the film thickness of the coating is obtained from the incident angle of the irradiation light that is measured by the light receiving device while changing the incident angle stepwise and takes the maximum value and the minimum value in the fluctuation of the received light intensity of the measured reflected light.

JP 2009-092454 A JP2013-079921A JP 2004-279296 A

  As the sample size increases, there is a need to measure the in-plane distribution of film thickness for a wider sample at a higher speed and with higher accuracy. The configurations disclosed in Patent Document 1 and Patent Document 2 described above basically measure light by irradiating one point on the sample, and measure the in-plane distribution of film thickness at high speed. The needs cannot be met sufficiently.

  Further, Patent Document 3 employs a so-called peak-valley method in which the film thickness is calculated using the position where the maximum value and the minimum value of the interference waveform occur. The peak valley method may not be able to accurately measure the film thickness due to the influence of noise caused by an optical system or the like. Moreover, the peak-valley method cannot measure the film thickness of each layer of a sample in which a plurality of layers are stacked. Therefore, although Patent Document 3 can be applied to a manufacturing process of a liquid crystal display device or the like, the in-plane distribution of film thicknesses of various samples cannot be measured for general purposes.

  An object of the present invention is to provide an optical measuring apparatus and an optical measuring method capable of measuring in-plane distribution of film thicknesses of various samples at higher speed and with higher accuracy. Another object of the present invention is to provide an optical measuring device and an optical measuring method capable of measuring optical characteristics of a sample such as a refractive index without using a dedicated measuring device.

  An optical measurement apparatus according to an aspect of the present invention includes an irradiation optical system that linearly irradiates measurement light having a predetermined wavelength range on a measurement target, and transmitted light or reflected light generated from the measurement target by irradiation of the measurement light. A measurement optical system for receiving a certain linear measurement interference light and a processing device are included. The measurement optical system picks up the measurement interference light in a direction orthogonal to the longitudinal direction of the measurement interference light, and receives the measurement interference light expanded in wavelength by the diffraction grating and outputs a two-dimensional image. Part. The processing device calculates a correction element corresponding to an incident angle from each measurement point to the measurement optical system in association with a region on the two-dimensional image corresponding to each measurement point to be measured irradiated with measurement light. And a second calculation unit that calculates an optical characteristic of the measurement target after applying a corresponding correction element to each pixel value included in the two-dimensional image.

  Preferably, the correction element includes a wave number that is a parameter including the wavelength of the measurement light and the refractive index of the measurement target. The wave number is calculated in consideration of the magnitude of the corresponding incident angle for each pixel position of the two-dimensional image.

  Preferably, the second calculation means converts a column of values converted according to a relational expression for linearizing a pixel value of a two-dimensional image corresponding to a target measurement point with respect to a phase factor, for a corresponding column of wave numbers. Means for Fourier transform, means for determining the film thickness at the measurement point of interest based on the peak position appearing in the power spectrum obtained by Fourier transform, and film thickness distribution by collecting the film thicknesses determined for a plurality of measurement points And outputting means.

Preferably, the wave number is calculated in consideration of the wavelength dependence of the refractive index of the measurement target.
Preferably, the correction element includes a value indicating the magnitude of the incident angle corresponding to each measurement point. The second calculation means uses the film thickness at each measurement point as a variation parameter, the refractive index of the measurement object, a value indicating the magnitude of the incident angle corresponding to each measurement point, each measurement point and the two-dimensional image. Means for calculating the theoretical value of each pixel corresponding to the two-dimensional image based on the correspondence relationship with the pixel position of the pixel, and the similarity between the calculated theoretical value of each pixel and each pixel value of the two-dimensional image is And a means for determining the film thickness at each measurement point by adjusting the variation parameter so as to be higher.

  An optical measurement method according to another aspect of the present invention linearly irradiates measurement light having a predetermined wavelength range with respect to a measurement target, and transmits light or reflected light generated from the measurement target by irradiation of the measurement light. Receiving the measurement interference light in the form of a wave, expanding the wavelength of the measurement interference light in a direction orthogonal to the longitudinal direction of the measurement interference light, and receiving the measurement interference light expanded in wavelength and outputting a two-dimensional image A step of calculating a correction element according to an incident angle from each measurement point in association with a region on the two-dimensional image corresponding to each measurement point of the measurement target irradiated with the measurement light, and a two-dimensional image And applying a corresponding correction element to each pixel value included in, and calculating an optical characteristic of the measurement target.

  Preferably, the correction element includes a wave number that is a parameter including the wavelength of the measurement light and the refractive index of the measurement target. The wave number is calculated in consideration of the magnitude of the corresponding incident angle for each pixel position of the two-dimensional image.

  Preferably, in the step of calculating the optical characteristics, a column of values converted according to a relational expression for linearizing a pixel value of a two-dimensional image corresponding to a target measurement point with respect to a phase factor is converted into a column of corresponding wave numbers. A step of Fourier transforming, a step of determining a film thickness at a measurement point of interest based on a peak position appearing in a power spectrum obtained by the Fourier transform, and a film thickness obtained by collecting the film thicknesses determined for a plurality of measurement points Outputting as a distribution.

Preferably, the wave number is calculated in consideration of the wavelength dependence of the refractive index of the measurement target.
Preferably, the correction element includes a value indicating the magnitude of the incident angle corresponding to each measurement point. The step of calculating the optical characteristics includes using the film thickness at each measurement point as a variable parameter, the refractive index of the measurement object, the value indicating the magnitude of the incident angle corresponding to each measurement point, each measurement point and the two-dimensional A step of calculating a theoretical value of each pixel corresponding to the two-dimensional image based on a correspondence relationship with a pixel position of the image, and a similarity between the calculated theoretical value of each pixel and each pixel value of the two-dimensional image Adjusting the variation parameter so as to increase the thickness, and determining the film thickness at each measurement point.

  According to still another aspect of the present invention, an irradiation optical system that linearly irradiates measurement light having a predetermined wavelength range on a measurement target, and transmitted light or reflected light generated from the measurement target by irradiation of the measurement light. There is provided an optical measurement method using an optical measurement apparatus including a measurement optical system that outputs a two-dimensional image by developing a wavelength of a certain linear measurement interference light in a direction orthogonal to the longitudinal direction of the measurement interference light. . In the optical measurement method, the same measurement reference is sequentially arranged at the measurement points of the measurement target irradiated with the measurement light, and the actual measurement values corresponding to the two-dimensional image are obtained by sequentially acquiring the actual measurement values at the measurement points. A step of obtaining, and a step of calculating a correction element according to an incident angle from each measurement point to the measurement optical system in association with a region on the two-dimensional image corresponding to each measurement point of the measurement object irradiated with the measurement light And calculating optical characteristics including the refractive index of the measurement reference based on the pixel value group for one or a plurality of columns along any one direction of the measured value distribution and the corresponding correction element. .

  According to an embodiment of the present invention, the in-plane distribution of film thicknesses of various samples can be measured with higher speed and higher accuracy. According to another embodiment of the present invention, the optical characteristics of a sample such as a refractive index can be measured without using a dedicated measuring device or the like.

It is a schematic diagram which shows schematic structure of the optical measurement apparatus of the transmission system according to this Embodiment. It is a schematic diagram which shows schematic structure of the optical measurement apparatus of the reflection type according to this Embodiment. It is a schematic diagram which shows schematic structure of the measurement optical system employ | adopted as the optical measuring device according to this Embodiment. It is a schematic diagram which shows schematic structure of the position adjustment mechanism employ | adopted as the optical measuring device according to this Embodiment. It is a schematic diagram which shows schematic structure of the processing apparatus according to this Embodiment. It is a figure for demonstrating incidence | injection of the measurement interference light to the measurement optical system of the optical measuring device according to this Embodiment. It is a figure for demonstrating the principle of the film thickness measuring method according to this Embodiment. It is a figure which shows an example of the two-dimensional image handled in the optical measuring device according to this Embodiment. It is a figure for demonstrating the binning process at the time of calculating the incident angle used for the film thickness measuring method according to this Embodiment. It is a figure for demonstrating the calculation method of the incident angle used for the film thickness measuring method according to this Embodiment. It is a flowchart which shows the process sequence (the 1) of the film thickness measuring method according to this Embodiment. It is a figure for demonstrating the processing content in the process sequence (the 1) of the film thickness measuring method shown in FIG. It is a figure which shows an example of the wavelength distribution of the refractive index of a polyethylene thin film. It is a figure which shows an example of the film thickness trend obtained by the film thickness measuring method according to this Embodiment. It is a figure which shows an example of the two-dimensional image which shows the transmittance | permeability spectrum according to the theoretical formula according to this Embodiment. It is a graph which shows the transmittance | permeability spectrum T ((lambda)) corresponding to the position direction pixel number j in the two-dimensional image (theoretical value) shown in FIG. It is a graph showing the transmittance spectrum T (lambda) wavenumber conversion transmittance was calculated from the T '(K ₁₎ shown in FIG. 16. It is a diagram illustrating an example of a thickness trend that is calculated from the wave number conversion transmittance T '(K ₁₎ shown in FIG. 17. It is a schematic diagram for demonstrating the processing content in the process sequence (the 2) of the film thickness measuring method according to this Embodiment. It is a flowchart which shows the process sequence (the 2) of the film thickness measuring method according to this Embodiment. It is a schematic diagram for demonstrating the outline | summary of the refractive index measuring method according to this Embodiment. It is a schematic diagram for demonstrating the outline | summary of the refractive index measuring method according to this Embodiment. It is a graph which shows an example of the film thickness trend computed according to the refractive index measuring method (the 1) based on the information of the wavelength direction according to this Embodiment. It is a flowchart which shows the process sequence of the refractive index measuring method (the 1) based on the information of the wavelength direction according to this Embodiment. It is a figure for demonstrating the determination method of the refractive index in the refractive index measuring method (the 2) based on the information of the wavelength direction according to this Embodiment. It is a figure for demonstrating the determination method of the refractive index in the refractive index measuring method (the 3) based on the information of the wavelength direction according to this Embodiment. It is a figure for demonstrating the method of determining the more probable value of the film thickness in the refractive index measuring method (the 2) based on the information of the position direction according to this Embodiment.

  Embodiments of the present invention will be described in detail with reference to the drawings. Note that the same or corresponding parts in the drawings are denoted by the same reference numerals and description thereof will not be repeated.

<A. Configuration of Optical Measuring Device>
First, the device configuration of the optical measurement device according to the present embodiment will be described. The optical measurement device according to the present embodiment is a measurement device using an imaging spectrometer, and irradiates a measurement object (hereinafter also referred to as “sample”) with a linear measurement light, and the linear measurement device. Wavelength information at each measurement point on the measurement line irradiated with the measurement light by spectroscopically analyzing the light generated when the measurement light is transmitted through the sample or the reflected light generated when the line-shaped measurement light is reflected by the sample To get. The transmitted light or reflected light generated from the sample indicates the result of causing interference in the sample, and is hereinafter also referred to as “measurement interference light”.

Hereinafter, a typical apparatus configuration of the optical measurement apparatus according to the present embodiment will be described.
(A1: Transmission system)
FIG. 1 is a schematic diagram showing a schematic configuration of a transmissive optical measurement apparatus 1 according to the present embodiment. Referring to FIG. 1, an optical measurement apparatus 1 includes a measurement optical system 10, a light source 20 that generates measurement light, a line light guide 22 that irradiates the sample S with measurement light generated by the light source 20, and a processing apparatus 100. Including.

  The light source 20 and the line light guide 22 correspond to a linear light source unit (irradiation optical system) that linearly irradiates the sample S with light having a predetermined wavelength range. The wavelength range of the measurement light is determined according to the range of wavelength information to be acquired from the sample S. For example, a halogen lamp is used as the light source 20.

  The line light guide 22 is typically disposed immediately below the surface on which the sample S is transported, and irradiates the sample S with measurement light from the light source 20 toward the sample S. On the irradiation surface of the line light guide 22, a diffusion member or the like for suppressing unevenness in the amount of light is disposed. The measurement light from the line light guide 22 is incident on the sample S to generate a measurement line 24 that is irradiated with the measurement light.

  The measurement optical system 10 receives linear measurement interference light that is transmitted light or reflected light generated from the sample S by irradiation of the measurement light. More specifically, the measurement optical system 10 determines the wavelength distribution characteristics of the transmittance or reflectance at each measurement point based on the measurement interference light transmitted through the sample S or the measurement interference light reflected by the sample S. get. The measurement optical system 10 is disposed at a position facing the line light guide 22 with the sample S interposed therebetween. Of the measurement light emitted from the line light guide 22, light (measurement interference light) transmitted through the sample S enters the measurement optical system 10. The measurement optical system 10 is fixed by the base member 4 and the support member 6.

  The measurement optical system 10 includes an objective lens 12, an imaging spectroscope 14, and an imaging unit 16. The transmitted light from the sample S is converged by the objective lens 12 and guided to the imaging spectroscope 14.

The imaging spectroscope 14 collectively measures spectral information at each measurement point on the line of the sample S. More specifically, the imaging spectrometer 14 develops the wavelength of the incident line-shaped transmitted light and outputs it to the imaging unit 16. The imaging unit 16 is configured by an imaging element having a two-dimensional light receiving surface. Such an image sensor is composed of, for example, a CCD (Charge Coupled Device) image sensor or a CMOS (Complementary Metal Oxide Semiconductor) image sensor. The imaging unit 16 outputs a two-dimensional image by receiving the transmitted light from the imaging spectroscope 14 with an imaging element. The output two-dimensional image includes wavelength information and position information. Details of the measurement optical system 10 will be described later.

  The processing apparatus 100 performs a process as described later on the two-dimensional image output from the measurement optical system 10 (imaging unit 16), thereby obtaining a sample such as a film thickness at each measurement point on the measurement line 24. The characteristic value of S is calculated. Details of the measurement process performed by the processing apparatus 100 will be described later.

(A2: reflection system)
FIG. 2 is a schematic diagram showing a schematic configuration of a reflection-type optical measurement device 2 according to the present embodiment. Referring to FIG. 2, the optical measurement device 2 is different from the optical measurement device 1 in the positional relationship between the measurement optical system 10 and the line light guide 22. Specifically, the line light guide 22 is arranged so that the measurement light with respect to the sample S has an incident angle Θ (> 0) with respect to a plane including the measurement line 24 and the vertical direction 28. The measurement optical system 10 is disposed at a position where light (measurement interference light) generated by reflection of measurement light incident on the sample S can be received. The measurement optical system 10 is arranged so that its optical axis has the same incident angle Θ with respect to the plane including the measurement line 24 and the vertical direction 28.

  Since the other configuration of optical measuring device 2 is substantially the same as optical measuring device 1, detailed description will not be repeated.

  For convenience of explanation, the details will be basically described by taking the optical measurement apparatus 1 employing a transmission system as an example.

(A3: Measurement optical system)
Next, measurement optical system 10 employed in the optical measurement device according to the present embodiment will be described.

  FIG. 3 is a schematic diagram showing a schematic configuration of the measurement optical system 10 employed in the optical measurement apparatus according to the present embodiment. With reference to FIG. 3, in the measurement optical system 10, measurement interference light from the sample S forms an image with the objective lens 12 and then enters the imaging spectroscope 14.

  The imaging spectroscope 14 includes a slit 142, a first lens 144, a diffraction grating 146, and a second lens 148 in order from the sample S.

  The slit 142 shapes the beam cross section of the measurement interference light incident through the objective lens 12 into a predetermined shape. The length of the slit 142 in the longitudinal direction is set according to the measurement line 24 generated on the sample S, and the width in the short direction of the slit 142 is set according to the resolution of the diffraction grating 146 and the like.

  The first lens 144 is typically composed of a collimator lens, converts the measurement interference light that has passed through the slit 142 into parallel light, and then guides it to the diffraction grating 146.

  The diffraction grating 146 develops the wavelength of the measurement interference light in a direction orthogonal to the longitudinal direction of the measurement interference light. More specifically, the diffraction grating 146 expands the wavelength of the line-shaped measurement interference light that has passed through the objective lens 12 and the slit 142 in a direction orthogonal to the line direction. The two-dimensional image 150 corresponding to the longitudinal direction of the measurement line 24 and the direction orthogonal to the longitudinal direction is generated on the light receiving surface of the imaging device 160 of the imaging unit 16 by the wavelength development by the diffraction grating 146. The imaging unit 16 receives the measurement interference light whose wavelength is expanded by the diffraction grating 146 and outputs a two-dimensional image. Although FIG. 3 shows an example in which a transmission type diffraction grating is employed as the diffraction grating 146, a reflection type diffraction grating may be employed.

  In the following description, the direction of the two-dimensional image 150 corresponding to the longitudinal direction of the measurement line 24 on the sample S is referred to as “position direction”, and the wavelength expanded direction orthogonal to the position direction is referred to as “wavelength direction”. Called. Each point in the position direction corresponds to each measurement point on the measurement line 24, and each point in the wavelength direction corresponds to each wavelength at the corresponding measurement point.

  As shown in FIG. 3, the measurement optical system 10 takes the measurement interference light from the sample S linearly through the objective lens 12 and the slit 142. The linear measurement interference light is converted into parallel light by the first lens 144, and the linear measurement interference light is positioned in the position direction by the transmission type or reflection type diffraction grating 146 disposed at the subsequent stage of the first lens 144. Wavelength development (spectroscopy) in a direction (wavelength direction) orthogonal to. The second lens 148 disposed in the subsequent stage forms an image of the measurement interference light that has been wavelength-developed as a two-dimensional optical spectrum that reflects wavelength information and position information. The two-dimensional image sensor 160 receives the formed image.

In the following description, the light receiving surface of the imaging element 160 has a C _x channel as the resolution of the wavelength direction, and has a C _y channel as the resolution of the position-direction.

  As described above, the two-dimensional image 150 reflects wavelength information and position information. By using such a two-dimensional image 150, wavelength information at a plurality of measurement points set on the sample S can be acquired at once.

(A4: Position adjustment mechanism of measurement optical system)
Next, a position adjustment mechanism of measurement optical system 10 that can be mounted on the optical measurement apparatus according to the present embodiment will be described. In order to appropriately guide the measurement interference light transmitted through the sample S or the measurement interference light reflected by the sample S to the measurement optical system 10, it is necessary to appropriately adjust the position of the measurement optical system 10 with respect to the sample S. Hereinafter, some of the position adjustment mechanism of the measurement optical system 10 and a position adjustment method using the position adjustment mechanism will be described.

  FIG. 4 is a schematic diagram showing a schematic configuration of a position adjustment mechanism 170 employed in the optical measurement device according to the present embodiment. The position adjustment mechanism 170 illustrated in FIG. 4 is disposed between the slit 142 and the first lens 144. The position adjustment mechanism 170 includes a shutter 172 and a light source 174 that generates observation light. The observation light is light for adjusting the focal position of the measurement optical system 10 with respect to the sample S and adjusting the observation position of the measurement optical system 10 with respect to the sample S.

  The shutter 172 is disposed at the intersection of the optical axis of the slit 142 and the first lens 144 and the optical axis of the light source 174. The shutter 172 can transition between an open state and a closed state. The shutter 172 passes light from the slit 142 toward the first lens 144 in the open state. On the other hand, when the shutter 172 is closed, the optical path from the slit 142 toward the first lens 144 is blocked, and a mirror attached to the back surface of the shutter 172 reflects the observation light from the light source 174 toward the slit 142. . That is, the observation light from the light source 174 is irradiated onto the sample S when the shutter 172 is closed.

  The user can appropriately adjust the distance (focal position) from the sample S to the measurement optical system 10 by adjusting the position of the measurement optical system 10 while observing the state of the observation light appearing on the sample S, and performing measurement. The position on the sample S (the position of the measurement site) observed by the optical system 10 can be adjusted appropriately. That is, while the light source 174 is turned on and the shutter 172 is closed, the position of the measurement optical system 10 and the focus of the objective lens 12 are adjusted so that the contrast of the observation light appearing on the sample S is maximized. By doing so, the measurement location of the measurement optical system 10 can be confirmed, and focusing of the measurement optical system 10 can be realized.

  As another method for adjusting the position of the measurement optical system 10, a test chart is arranged instead of the sample S, and a two-dimensional image 150 obtained by imaging the test chart with the imaging unit 16 is evaluated. Thus, the position of the measurement optical system 10 may be adjusted. As the test chart, for example, a pattern such as Ronchi-Ruring or a black and white striped pattern drawn at equal intervals can be used. When such a pattern is used, the distance (focal position) from the sample S to the measurement optical system 10 is adjusted so that the contrast ratio of light and dark displayed on the actually captured two-dimensional image 150 is maximized. May be.

<B. Apparatus configuration of processing apparatus>
Next, the apparatus configuration of processing apparatus 100 included in the optical measurement apparatus according to the present embodiment will be described. Processing apparatus 100 according to the present embodiment is typically implemented using a general-purpose computer.

  FIG. 5 is a schematic diagram showing a schematic configuration of processing apparatus 100 according to the present embodiment. Referring to FIG. 5, the processing device 100 includes a processor 102, a main memory 104, an input unit 106, a display unit 108, a storage 110, a communication interface 120, a network interface 122, and a media drive 124. Including.

The processor 102 is typically a CPU (Central Processing Unit) and a GPU.
An arithmetic processing unit such as (Graphics Processing Unit), which reads one or more programs stored in the storage 110 into the main memory 104 and executes them.

The main memory 104 is a volatile memory such as DRAM (Dynamic Random Access Memory) or SRAM (Static Random Access Memory), and functions as a working memory for the processor 102 to execute a program.

  The input unit 106 includes a keyboard, a mouse, and the like, and accepts an operation from the user. The display unit 108 outputs the execution result of the program by the processor 102 to the user.

  The storage 110 is composed of a nonvolatile memory such as a hard disk or a flash memory, and stores various programs and data. More specifically, the storage 110 holds an operating system 112 (OS: Operating System), a measurement program 114, two-dimensional image data 116, and a measurement result 118.

  The operating system 112 provides an environment in which the processor 102 executes a program. The measurement program 114 realizes a film thickness measurement method, a refractive index measurement method, and the like according to the present embodiment as described later. The two-dimensional image data 116 is data acquired by the imaging unit 16 of the measurement optical system 10. The measurement result 118 includes a result obtained by executing the measurement program 114.

  The communication interface 120 mediates data transmission between the processing apparatus 100 and the measurement optical system 10, acquires two-dimensional image data from the measurement optical system 10, or gives various instructions to the measurement optical system 10. . The network interface 122 mediates data transmission between the processing device 100 and an external server device, transmits measurement results to the server device, or receives a program from the server device.

  The media drive 124 reads necessary data from a recording medium 126 (for example, an optical disk) that stores a program executed by the processor 102 and stores the data in the storage 110. Note that the measurement program 114 or the like executed in the processing apparatus 100 may be installed via the recording medium 126 or the like, or may be downloaded from the server apparatus via the network interface 122 or the like.

  The measurement program 114 may be a program module that is provided as a part of the operating system 112 and calls necessary modules in a predetermined arrangement at a predetermined timing to execute processing. In such a case, the measurement program 114 that does not include the module is also included in the technical scope of the present invention. The measurement program 114 may be provided by being incorporated in a part of another program.

  Note that all or part of the functions provided by the processor 102 of the processing device 100 executing the measurement program 114 may be realized by dedicated hardware.

<C. Overview of optical property measurement method>
Next, an outline of the optical property measurement method used in the optical measurement apparatus including the imaging spectrometer shown in FIG. 1 or FIG. 2 will be described. The optical measurement apparatus according to the present embodiment measures optical characteristics such as the in-plane distribution of the film thickness of the sample S using the two-dimensional image 150 including wavelength information and position information.

  When the in-plane distribution of the film thickness of the sample S is measured using an imaging spectroscope, a plurality of measurement points are arranged in a line, so that the positional relationship with respect to the measurement optical system 10 differs between the measurement points.

FIG. 6 is a diagram for explaining the incidence of measurement interference light on the measurement optical system 10 of the optical measurement device according to the present embodiment. Referring to FIG. 6, measurement interference light Lc generated on the sample S from the center portion of the measurement line 24 propagates through an optical path substantially similar to the optical axis of the measurement optical system 10. On the other hand, the measurement interference light Le from the end of the measurement line 24 enters the measurement optical system 10 with a certain incident angle θ. Due to the presence of the incident angle θ, information appearing in the two-dimensional image 150 differs between measurement points on the measurement line 24.

  Therefore, in the optical characteristic measurement method according to the present embodiment, the incident angle θ when the measurement interference light from the sample S enters the measurement optical system 10 is considered. That is, when the in-plane distribution of the film thickness of the sample S is measured, the positional information of the two-dimensional image 150 output from the measurement optical system 10 is substantially corrected. More specifically, as will be described later, the processing apparatus 100 associates the measurement optical system 10 from each measurement point with a region on the two-dimensional image 150 corresponding to each measurement point of the measurement target irradiated with the measurement light. A correction factor corresponding to the incident angle to is calculated. Then, the processing apparatus 100 calculates the optical characteristics of the sample S after applying a corresponding correction element to each pixel value included in the two-dimensional image 150.

  Further, depending on the sample S, the refractive index of the sample S has wavelength characteristics. In this case, such wavelength characteristics are considered. That is, when measuring the in-plane distribution of the film thickness of the sample S, the correction may be performed on the information in the wavelength direction of the two-dimensional image 150 output from the measurement optical system 10 substantially. .

  For convenience of explanation, an optical property measurement method considering both (1) the influence of the incident angle of the measurement interference light and (2) the wavelength property of the refractive index of the sample S will be described in detail. In some cases, it is not necessary to consider the wavelength characteristics of the refractive index.

  Hereinafter, as a typical example of the optical property measurement method according to the present embodiment, a film thickness measurement method for measuring the film thickness (or in-plane distribution of the film thickness) of the sample S, and refraction for measuring the refractive index of the sample S. The rate measurement method will be described. However, the optical property measuring method according to the present embodiment can be applied not only to measuring the film thickness and / or refractive index but also to measuring any optical property.

<D. Theoretical explanation of film thickness measurement method>
Next, a theoretical description of the film thickness measuring method according to the present embodiment will be given.

FIG. 7 is a diagram for explaining the principle of the film thickness measuring method according to the present embodiment. Referring to FIG. 7A, consider a case where a thin film sample (film thickness d ₁ ) is arranged in air (medium 0). In consideration of multiple reflection occurring in the sample S (medium 1), the intensity transmittance T (1-R) and the intensity reflectance R are expressed by the following expressions (1) and (2), respectively.

However, n ₁ is the refractive index of the sample S (medium 1), n ₀ is the refractive index of air (medium 0), lambda is a wavelength. In the above equation, the amplitude reflectance r ₀₁ represents the amplitude reflectance in the optical path of medium 0 → medium 1 → medium 0. The phase difference factor β ₁ generated by the propagation of light through the sample S shown in FIG. 7A can be expressed by the following equation (3).

Here, as shown in FIG. 7B, considering the case where the incident angle of light with respect to the sample S is θ ₀ , the refraction angle of light generated in the sample S is θ ₁ . Here, a relationship of n ₀ · sin θ ₀ = n ₁ · sin θ ₁ (Snell's law) is established between the incident angle θ ₀ and the refraction angle θ ₁ . Here, by utilizing the relationship between the incident angle θ ₀ and the refraction angle θ ₁ , a wave number K ₁ as shown in the following equation (4) is introduced. The wave number K ₁ corresponds to a parameter for facilitating Fourier transform for measuring the film thickness. By using the wave number K ₁ , the phase angle β ₁ in the sample S can be defined as shown in the following equation (5).

The wave number K ₁ shown in the above equation (4) includes the incident angle θ _0. By using such a wave number K ₁ , the film thickness considering the difference in the incident angle θ ₀ corresponding to each measurement point can be obtained. It can be calculated.

Further, when considering the Fourier transform for the phase angle β ₁ , the phase factor (cos ₂ β ₁ ) is non-linear with respect to the intensity reflectance R, and FFT (Fast Fourier Transform) is applied as it is. I can't. Therefore, by introducing a unique variable, the phase factors Cos2beta ₁ after having converted to a function having a linearity, implementing the Fourier transform. As an example, a wave number conversion transmittance T ′ (≡1 / T) or a wave number conversion reflectance R ′ (≡R / (1-R)), which is a linear expression for the phase factor cos 2β ₁ , is introduced. Specifically, the wave number conversion transmittance T ′ and the wave number conversion reflectance R ′ are derived from the above equations (1) and (2) as in the following equations (6) and (7).

Further, the power spectrum P (K ₁ ) obtained by Fourier transforming the wave number conversion transmittance T ′ shown in the equation (6) or the wave number conversion reflectance R ′ shown in the above equation (7) A peak appears at a position corresponding to the film thickness d ₁ of the sample S. That is, the film thickness d ₁ of the sample S is determined by calculating the position of the peak appearing in the power spectrum P (K ₁ ).

  For details of the wave number conversion transmittance T ′ and the wave number conversion reflectance R ′, refer to Japanese Unexamined Patent Application Publication No. 2009-092454 (Patent Document 1) and the like.

As described above, the correction factor corresponding to the incident angle from each measurement point to the measurement optical system includes the wave number K ₁ that is a parameter including the wavelength λ of the measurement light and the refractive index n ₁ of the sample S. Note that the wave number K ₁ may be calculated in consideration of the wavelength dependence of the refractive index of the sample S. A sequence of values converted according to a relational expression (for example, R / (1-R), 1 / T, etc.) for linearizing the pixel value of the two-dimensional image corresponding to the measurement point of interest with respect to the phase factor. The film thickness is obtained by Fourier transforming (wave number conversion transmittance distribution T ′ (i, j) or wave number conversion reflectance distribution R ′ (i, j)) with respect to the corresponding column of wave numbers K ₁ (i, j). d is determined.

As described above, when the incident angle theta ₀ of the measured interference light to the sample S can not be regarded as zero, by introducing a wave number K ₁ including the incident angle theta _0, considering the effect of incident angle theta ₀ The film thickness of the sample S can be calculated.

Next, a method for calculating the incident angle θ ₀ will be described. As described above, in the film thickness measurement method according to the present embodiment, it is necessary to calculate the incident angle θ ₀ corresponding to each measurement point. Each measurement point can be set in a pixel unit of the two-dimensional image 150 output from the measurement optical system 10 or a pixel set unit in which a plurality of adjacent pixels are collected.

  FIG. 8 is a diagram showing an example of a two-dimensional image 150 handled in the optical measurement apparatus according to the present embodiment. Since the light receiving surface of the image sensor 160 and the two-dimensional image 150 correspond one-to-one, coordinates indicating an arbitrary position on the light receiving surface of the image sensor 160 indicate a corresponding position of the two-dimensional image 150.

Originally, it is necessary to calculate the incident angle θ ₀ for each channel of the image sensor 160. When the number of channels of the image sensor 160 is sufficiently large with respect to the angle of view φ of the measurement optical system 10, a set of a plurality of adjacent channels is regarded as one pixel of the two-dimensional image 150, and for each pixel The incident angle θ ₀ may be calculated. Collecting a plurality of such adjacent channels is also referred to as “binning” below.

FIG. 9 is a diagram for illustrating a binning process when calculating the incident angle θ ₀ used in the film thickness measuring method according to the present embodiment. As shown in FIG. 9, among a plurality of channels constituting the image sensor 160, a predetermined number of adjacent channels arranged in the position direction are processed together. The image sensor 160 has a resolution of C _x channel × C _y channel, and can output a two-dimensional image 150 corresponding to the number of channels. However, the processing speed can be increased by grouping adjacent channels.

The number of channels collected in the position direction is also referred to as “binning number B _y ”. Binning number B _y is preferably a common divisor of the number of channels C _y. By setting the binning number B _y to "1", so that the incident angle theta ₀ corresponding to the number of channels of the image sensor 160 is calculated.

Although FIG. 9 shows an example in which a plurality of channels are set to one pixel in the position direction, a plurality of channels may be set to one pixel also in the wavelength direction. That is, the binning number B _x in the wavelength direction may be introduced.

In the following description, the position of an arbitrary pixel among a plurality of pixels defined in the two-dimensional image 150 is represented by a wavelength direction pixel number (hereinafter represented by “variable i”) and a position direction pixel number (hereinafter, In combination with “variable j” or “variable j ′”). Here, the position direction pixel number j is an integer satisfying 1 ≦ j ≦ C _y / B _y .

FIG. 8A shows a coordinate system when the lower left corner of the page is the origin coordinate (1, 1). In this case, the coordinates at the upper right of the paper surface are (C _x / B _x , C _y / B _y ). FIG. 8B shows a coordinate system in the case where the center in the position direction is the origin coordinate (1, 0). In the coordinate system shown in FIG. 8B, the incident angle θ ₀ corresponding to the center of the position direction, that is, the pixel on the straight line connecting the coordinates (1, 0) and the coordinates (C _x / B _x , 0) is It becomes zero. Here, a relationship of j ′ = j−C _y / 2B _y is established between the position direction pixel number j ′ and the position direction pixel number j.

The coordinate system shown in FIG. 8A has an advantage that the processing for the wavelength information and the position information included in the two-dimensional image 150 can be simplified, and the coordinate system shown in FIG. 8B corresponds to each measurement point. There is an advantage that the processing can be simplified when calculating the incident angle θ ₀ .

Hereinafter, the incident angle θ ₀ of the measurement point corresponding to the position direction pixel number j (or position direction pixel number j ′) of the two-dimensional image 150 will be considered.

FIG. 10 is a diagram for describing a method of calculating the incident angle θ ₀ used in the film thickness measuring method according to the present embodiment. FIG. 10 shows, as an example, a case where the measurement line is regarded as an arc (measurement line 24 ′) and a case where the measurement line is regarded as a straight line (measurement line 24) in the transmission optical measurement apparatus 1 shown in FIG. . Either case is employed to calculate the incident angle θ ₀ .

  In FIG. 10, b indicates the length of the image sensor 160 in the position direction, f indicates the focal length of the objective lens 12, and h indicates the height h of the objective lens 12. Note that the length b, the focal length f, and the height h are in the same unit (for example, mm). Φ represents the angle of view φ (= Atan (b / 2f)) of the measurement optical system 10.

When the measurement line is regarded as an arc (measurement line 24 ′), the incident angle θ ₀ corresponding to the position direction pixel number j ′ is derived as in the following equation (8-1). Expression (8-1) defines the incident angle θ ₀ using the position direction pixel number j ′ when the field angle φ and the center of the position direction are set to zero. Here, using the relational expression about the angle of view φ (φ = Atan (b / 2f)) and the relational expression about the position direction pixel number (j ′ = j−C _y / 2B _y ), (8− By transforming equation (1), equation (8-2) can be derived.

When the measurement line is regarded as a straight line (measurement line 24), the incident angle θ ₀ corresponding to the position direction pixel number j ′ is derived as shown in the following equation (9-1). Expression (9-1) defines the incident angle θ ₀ using the position direction pixel number j ′ when the field angle φ and the center of the position direction are set to zero. Here, using the relational expression for the angle of view φ (φ = Atan (b / 2f)) and the relational expression for the position direction pixel number (j ′ = j−C _y / 2B _y ), (9− By transforming equation (1), equation (9-2) can be derived.

  It should be noted that in the above equations (8-1), (8-2), (9-1), and (9-2), the size of each channel of the image sensor 160 is ignored and treated as a point. ing. Strictly speaking, the transmitted light or reflected light to be observed should be expressed by a value obtained by integrating the angle change corresponding to the size of one channel, but the size of one channel is compared with the length of the imaging range. Since it is sufficiently small and the change in angle is negligible, it can be represented by the value of transmitted light or reflected light from one point.

In the above (Formula 4), when n ₀ = 1 (refractive index of air) and the wave number K ₁ is defined using the wavelength direction pixel number i and the position direction pixel number j of the two-dimensional image, the following (10) An expression can be derived.

  In the equation (10), the wavelength conversion equation λ (i, j) indicating the relationship between the pixel position of the two-dimensional image 150 and the wavelength can be determined in advance by calibrating the wavelength of the measurement optical system 10. Here, the wavelength calibration includes an operation of assigning a value of the corresponding wavelength λ to each of the wavelength direction pixel numbers i for each of the position direction pixel numbers j.

The refractive index n ₁ (i, j) (that is, the refractive index n ₁ (λ)) of the sample S can be obtained in advance by a measuring apparatus capable of analyzing optical constants (for example, a microspectrophotometer). In the equation (10), the incident angle θ ₀ (j) corresponding to each measurement point is defined according to the above equation (8-2) or (9-2). By substituting these values into equation (10), the wave number K ₁ (i, j) at the pixel position (i, j) can be determined. Thus, the wave number K ₁ is calculated for each pixel position (i, j) of the two-dimensional image in consideration of the magnitude of the corresponding incident angle θ ₀ (j).

Based on the relationship between the wave number K ₁ (i, j) at the pixel position (i, j) and the actual measurement value, (1) the influence of the incident angle of the measurement interference light, and (2) the refractive index wavelength of the sample S The film thickness can be determined in consideration of both characteristics.

As a more specific calculation procedure, the transmittance distribution T (i, j) or the reflectance distribution R (i, j) of the sample S is obtained using the optical measurement device according to the present embodiment. Subsequently, for each position direction pixel number j, the horizontal axis is the wave number K ₁ (i, j), and the vertical axis is the wave number conversion transmittance distribution T ′ (i, j) or the wave number conversion reflectance distribution R ′ (i, j). The wave number distribution characteristic as j) is generated. The power spectrum P (K ₁ ) is calculated by Fourier transforming the generated wave number distribution characteristics, and the wavelength dependence of the incident angle and the refractive index is calculated based on the peak appearing in the calculated power spectrum P (K ₁ ). The film thickness distribution (in-plane distribution of film thickness) of the sample S taken into consideration can be determined.

As described above, the wave number K ₁ is expressed by the above-described equation based on the incident angle θ ₀ (j) at each measurement point on the measurement line, the refractive index n ₁ (λ) considering wavelength dispersion, and the wavelength λ. Accordingly, the pixel position is calculated for each pixel position (i, j) of the image sensor having a two-dimensional light receiving surface. A relational expression (for example, R / (1) for linearizing the phase factor cos2β from the transmittance distribution T (i, j) or the reflectance distribution R (i, j) of the sample S obtained by actual measurement. -R), 1 / T, etc.) is used to generate the wavenumber converted transmittance distribution T ′ (i, j) or the wavenumber converted reflectance distribution R ′ (i, j). By applying the wave number K ₁ (i, j) to the generated wave number conversion transmittance distribution T ′ or the wave number conversion reflectance distribution T ′, a power spectrum P (m, j) (wherein Fourier transform is performed) The parameter m can be obtained as a discrete value corresponding to the horizontal axis of the power spectrum. Based on the peak appearing in the power spectrum P (m, j), the film thickness value at each measuring point of the sample S is calculated.

  As a method for specifying a wave number component (peak) having a large amplitude from the wave number distribution characteristics, typically, a method using discrete Fourier transform such as FFT (Fast Fourier Transform), and a maximum Any of optimization methods such as an entropy method (hereinafter referred to as “MEM”) can be employed. When discrete Fourier transform is used, powers of 2 such as 512, 1024, 2048, 4096,... Are used as discrete values in the frequency domain.

<E. Specific example of film thickness measurement method>
Next, a film thickness measurement method based on the theoretical description of the film thickness measurement method described above will be described. In the following description, the method of determining the film thickness of the sample S based on the peak appearing in the power spectrum P (K ₁ ) obtained by Fourier transforming the wave number conversion transmittance T ′ or the wave number conversion reflectance R ′. (The so-called FFT method), the obtained wavelength distribution characteristic (measured value of transmittance spectrum or reflectance spectrum), and a model formula (theoretical formula) including incident angle, refractive index, wavelength, and film thickness as parameters. A method (so-called optimization method) for determining the film thickness of the sample S by comparing the shape with the wavelength distribution characteristic (fitting) will be described.

  Only one of these film thickness measurement methods may be mounted, but it is preferable that the film thickness measurement method can be appropriately selected depending on the film thickness or material of the sample S.

(E1: Processing procedure of film thickness measurement method (1))
First, the processing procedure (part 1) of the film thickness measuring method according to the present embodiment will be described. The processing procedure (part 1) of the film thickness measurement method is a method of determining the film thickness of the sample S based on the peak appearing in the power spectrum P (K ₁ ) for the wave number K ₁ .

  FIG. 11 is a flowchart showing a processing procedure (No. 1) of the film thickness measuring method according to the present embodiment. FIG. 12 is a diagram for explaining the processing contents in the processing procedure (part 1) of the film thickness measurement method shown in FIG.

Referring to FIG. 11, first, processing apparatus 100 calculates an incident angle θ ₀ corresponding to each measurement point for measurement interference light incident on measurement optical system 10 (step S100).

Specifically, the processing apparatus 100 sets each measurement point set on the measurement line (corresponding to each pixel in the position direction of the two-dimensional image 150 determined depending on the number of channels and the number of binning of the image sensor 160). the incident angle theta ₀ corresponding to calculated for each position the direction pixel number j to. That is, the processing device 100 calculates θ ₀ (j) in the above equation (10) for all the position direction pixel numbers j. Note that θ ₀ (j) may be a radian value or a trigonometric function value (for example, sin θ ₀ (j) or cos θ ₀ (j)). That is, as long as it is a value indicating the magnitude of the incident angle, any value corresponding to the subsequent calculation process may be adopted.

The value of the incident angle θ ₀ corresponding to each measurement point obtained as a result of step S100 need not be recalculated as long as the setting or configuration of the optical measurement device is the same. Therefore, when the incident angle θ ₀ corresponding to each measurement point is calculated in advance, the process of step S100 may be skipped.

The processing apparatus 100 obtains the refractive index n ₁ (λ) in consideration of the chromatic dispersion of the sample S from the measurement result of the sample S by a measuring apparatus (for example, a microspectrophotometer, etc.) capable of optical constant analysis. (Step S102).

Note that the refractive index n ₁ (λ) of the sample S acquired in step S102 need not be acquired again as long as the material of the sample S is the same. Therefore, as long as the identity of the sample S corresponding to the refractive index n ₁ (λ) acquired previously is maintained, the process of step S102 may be skipped. When the refractive index can be regarded as constant regardless of the wavelength, a constant value may be set as the refractive index n ₁ (λ).

  The processing apparatus 100 calculates a wavelength conversion formula λ (i, j) indicating the relationship between the pixel position of the two-dimensional image 150 and the wavelength λ from the result of wavelength calibration for the measurement optical system 10 (step S104). In step S104, the wavelength λ is associated with the pixel position (i, j) of the two-dimensional image 150. That is, the wavelength λ can be expressed in a matrix with the pixel position (i, j) of the two-dimensional image 150 as a parameter (that is, wavelength λ = λ (i, j)).

  Note that the wavelength conversion formula λ (i, j) calculated in step S104 basically does not need to be recalculated as long as the setting or configuration of the measurement optical system 10 is the same. Therefore, as long as the previously calculated wavelength conversion formula λ (i, j) can be used effectively, the process of step S104 may be skipped.

  Further, the execution order of the processes of steps S100, S102, and S104 is not particularly limited. In addition, the execution timings of steps S100, S102, and S104 may be different.

Subsequently, the processing apparatus 100 develops the wave number K ₁ shown in the above-described equation (4) for the pixel position (i, j) of the two-dimensional image 150. That is, the processing device 100 calculates the wave number K ₁ (i, j) for each pixel position (i, j) of the two-dimensional image 150 (step S106).

As shown in the above equation (4), in the film thickness measuring method according to the present embodiment, a wave number K ₁ having a wavelength λ, a refractive index n ₁ , and an incident angle θ ₀ as variables is introduced.

Of the variables included in the wave number K ₁ , the refractive index n ₁ is a function of the wavelength λ. Since the wavelength λ can be defined by the wavelength conversion equation λ (i, j), the refractive index n ₁ can be defined using the pixel position (i, j) of the two-dimensional image 150 as a parameter (ie, refractive index n ₁ = n _1). (Λ) = n ₁ (i, j) and wavelength λ = λ (i, j)). Further, the incident angle theta ₀ is calculated in step S100, it is possible to use the incident angle theta ₀ corresponding to each measurement point. The incident angle θ ₀ is defined only by the position direction pixel number j (that is, the incident angle θ ₀ = θ ₀ (j)).
).

As described above, since the wavelength λ, the refractive index n ₁ , and the incident angle θ ₀ can all be defined using the measurement points (i, j) corresponding to the pixels of the two-dimensional image 150, the pixel position (i, By specifying j), each value is uniquely determined. By applying the values of the wavelength λ, the refractive index n ₁ , and the incident angle θ ₀ determined for each pixel position (i, j), the processing device 100 determines the value of the wave number K ₁ for each pixel position (i, j). To generate a wave number K ₁ (i, j). As shown in FIG. 12, the generated wave number K ₁ (i, j) corresponds to each pixel of the two-dimensional image 150.

The processes in steps S100 to S106 described above correspond to a preparation process.
The processing apparatus 100 sets the sample S in the optical measurement apparatus 1 and acquires the two-dimensional image 150 captured in a state where the sample S is irradiated with measurement interference light (step S110). That is, the processing apparatus 100 acquires the actual measurement value of the transmittance distribution T (i, j) (or the reflectance distribution R (i, j)) for a plurality of measurement points on the measurement line 24 of the sample S. . As shown in FIG. 12, a two-dimensional image 150 having a wavelength direction and a position direction is acquired. Here, for a specific position direction pixel number j, the wavelength direction corresponds to the wavelength λ.

  Subsequently, the processing apparatus 100 sets the position direction pixel number j = 1 (step S112). Setting the position direction pixel number j means to target a pixel row corresponding to a specific position direction pixel number j of the two-dimensional image 150 as shown in FIG.

The processing device 100 refers to the wave number K ₁ (i, j) calculated in step S106, and obtains a horizontal value from the acquired transmittance distribution T (i, j) (or reflectance distribution R (i, j)). Generate wave number distribution characteristics with the wave number K ₁ (i, j) on the axis and the wave number conversion transmittance distribution T ′ (i, j) (or wave number conversion reflectance distribution R ′ (i, j)) on the vertical axis. (Step S114).

More specifically, as illustrated in FIG. 12, the processing apparatus 100 extracts a column value corresponding to the current position direction pixel number j from the wave number K ₁ (i, j) and extracts it from the two-dimensional image 150. Applied to the value of the selected pixel column.

  In this way, the processing apparatus 100 converts the acquired wavelength distribution characteristics (correspondence between each wavelength and the transmittance or reflectance value at that wavelength) into transmittance or reflectance calculated according to the wave number distribution characteristics. Convert to correspondence with value. The wave number distribution characteristic includes a correspondence relationship between a wave number determined by a function including an incident angle, a refractive index, and a wavelength as parameters, and a value of transmittance or reflectance at the wave number. Alternatively, the wave number distribution characteristic is a transmittance or reflectance conversion calculated according to a relational expression (for example, R / (1-R), 1 / T, etc.) for linearization with respect to the wave number and the phase factor cos 2β. Includes correspondence with values.

Subsequently, the processing apparatus 100 performs a Fourier transform on the wave number K ₁ (i, j), which is generated in step S114, with the horizontal axis as the wave number K ₁ (i, j), so that the power spectrum P (K ₁ ) is generated (step S116). In other words, the processing apparatus 100 performs a Fourier transform on a column of values converted according to a relational expression for linearizing a pixel value of the two-dimensional image 150 corresponding to a target measurement point with respect to a phase factor, with respect to a corresponding column of wave numbers. To do.

The processing device 100 calculates the peak position appearing in the power spectrum P (K ₁ ) generated in step S116, thereby calculating the film thickness d ₁ (j) for the measurement point corresponding to the current position direction pixel number j. (Step S118). That is, the processing apparatus 100 determines the film thickness at the measurement point of interest based on the peak position appearing in the power spectrum P (K ₁ ) obtained by Fourier transform. Note that a wave number component having a large amplitude (that is, the film thickness d ₁ (j)) may be determined by using an optimization method instead of the Fourier transform.

  The processing device 100 determines whether or not the current position direction pixel number j is a final value (step S120). If current position-direction pixel number j is not the final value (NO in step S120), processing device 100 increments current position-direction pixel number j by 1 (step S122), and repeats the processes in and after step S114. .

If the current position direction pixel number j is the final value (YES in step S120), the processing apparatus 100 sets the film thickness d ₁ (j) calculated for each of the position direction pixel number j from 1 to the final value. Collectively, a film thickness distribution on the measurement line 24 of the sample S is generated (step S124). That is, the processing apparatus 100 collects the film thicknesses determined for a plurality of measurement points and outputs them as a film thickness distribution.

  The processing apparatus 100 determines whether or not the film thickness measurement end condition for the sample S is satisfied (step S126). If the film thickness measurement termination condition for sample S is not satisfied (NO in step S126), processing device 100 repeats the processing in step S110 and subsequent steps.

  On the other hand, if the film thickness measurement end condition for sample S is satisfied (YES in step S126), processing device 100 integrates the film thickness distribution sequentially calculated in step S124, and A film thickness distribution (in-plane distribution of film thickness) on the measurement surface is output (step S128). Then, the process ends.

In the processing procedure (part 1) of the above-described film thickness measurement method, the description has been given focusing on the wave number K ₁ as a correction element according to the incident angle θ ₀ from the measurement point to the measurement optical system 10. However, it is not limited to this. For example, the correction element is a concept that may include the above-described wave number conversion transmittance T ′ (≡1 / T) or the wave number conversion reflectance R ′ (≡R / (1-R)).

(E2. Measurement example)
Next, a measurement example obtained by the film thickness measurement method (part 1) according to the present embodiment will be described.

  FIG. 14 is a diagram showing an example of a film thickness trend obtained by the film thickness measuring method according to the present embodiment. As sample S, a 1 mm square (outside dimension: 1 mm × 1 mm) polyethylene thin film was used.

For two patterns with and without correction of the incident angle θ ₀ (this embodiment), the measurement point of the sample S (that is, the incident angle θ ₀ ) was sequentially changed, and the film thickness of each pattern was measured. . As the objective lens 12, a lens having a focal length f = 16 mm was used.

As the refractive index n ₁ , the wavelength distribution of the refractive index of the polyethylene thin film measured by a microspectrophotometer was adopted. FIG. 13 is a diagram showing an example of the wavelength distribution of the refractive index n ₁ (λ) of the polyethylene thin film. As an example of the refractive index n ₁ (λ) shown in FIG. 13, the Cauchy dispersion formula shown in the following formula (11) is used.

In the example shown in FIG. 13, the coefficients C ₀ = 1.533731, C ₁ = 429.0333, and C ₂ = 2.09247 × 10 ⁸ .

FIG. 14 shows the change in the measured film thickness accompanying the change in the incident angle θ ₀ . As shown in FIG. 14, it can be seen that the film thickness trend is flatter by considering the incident angle θ ₀ . That is, it is shown that the film thickness distribution (in-plane distribution of film thickness) of the sample S can be measured more accurately.

(E3. Simulation example)
For example, the theoretical value of the transmittance spectrum T (λ) at each measurement point can be calculated by the above-described equations (1), (5), (8-2), and (10). FIG. 15 is a diagram showing an example of a two-dimensional image (1200 pixels × 1920 pixels) showing a transmittance spectrum according to the theoretical formula according to the present embodiment. The two-dimensional image showing the transmittance spectrum shown in FIG. 15 is obtained when the film thickness d ₁ = 10 [μm] (uniform) and the amplitude reflectance | r ₀₁ | = 0.2.

As the refractive index n ₁ , the wavelength distribution of the refractive index of the polyethylene thin film measured with a microscopic spectrophotometer as shown in FIG. 13 was used. As an example of the refractive index n ₁ (λ) shown in FIG. 13, the Cauchy dispersion formula as shown in the above formula (11) is used.

In the example shown in FIG. 13, the coefficients C ₀ = 1.533731, C ₁ = 429.0333, and C ₂ = 2.09247 × 10 ⁸ .

FIG. 16 is a graph showing the transmittance spectrum T (λ) corresponding to the position direction pixel number j in the two-dimensional image (theoretical value) shown in FIG. FIG. 16 shows the transmittance spectrum T (λ) in each line of the position direction pixel numbers j = 1, 100, 200,. In FIG. 16, the transmittance spectra T (λ) do not match because the incident angle θ ₀ corresponding to each measurement point is different.

FIG. 17 is a graph showing the wave number conversion transmittance T ′ (K ₁ ) calculated from the transmittance spectrum T (λ) shown in FIG. FIG. 17A shows the wave number conversion transmittance T ′ (K ₁ ) calculated using the wave number K ₁ assuming that the incident angle θ ₀ is zero for all the position direction pixel numbers j. FIG. 17B shows the wave number conversion transmittance T ′ (K ₁ ) calculated using the wave number K ₁ considering the incident angle θ ₀ corresponding to all the position direction pixel numbers j.

In the transmittance spectrum T (λ) shown in FIG. 16, the period of the interference waveform differs depending on the difference in the position direction pixel number j (that is, the incident angle θ ₀ ). Therefore, when the wave number K ₁ that does not consider the difference in the incident angle θ ₀ is used, it is understood that the wave number conversion transmittance T ′ (K ₁ ) is not uniform as shown in FIG. .

On the other hand, by using the wave number K ₁ considering the incident angle θ ₀ corresponding to the position direction pixel number j, as shown in FIG. 17B, the wave number conversion transmittance T ′ for all the position direction pixel numbers j. It can be seen that (K ₁ ) is aligned. Regardless of the pixel number j in the position direction, the wave number conversion transmittance T ′ (K ₁ ) is substantially the same, so that the correct film thickness can be calculated from any wave number conversion transmittance T ′ (K ₁ ).

FIG. 18 is a diagram illustrating an example of a film thickness trend calculated from the wave number conversion transmittance T ′ (K ₁ ) illustrated in FIG. Referring to FIG. 18, when the incident angle correction shown in FIG. 17A is not performed, the film thickness becomes maximum at the position direction pixel number j = 600 at which the incident angle ₀ is zero, and toward the both ends. As the incident angle θ ₀ increases, the film thickness decreases.

  On the other hand, when the incident angle correction shown in FIG. 17B is performed, it is understood that the original film thickness of 10 [μm] is correctly calculated in all the position direction pixel numbers j. .

As described above, by adopting the theoretical formula according to the present embodiment, the physical characteristics in consideration of the incident angle θ ₀ corresponding to the measurement point can be accurately reproduced. In other words, accurate fitting can be realized by using the mathematical formula as described above.

(E4: Processing procedure of film thickness measurement method (2))
In the processing procedure (part 1) of the above-described film thickness measuring method, the wave number K ₁ is introduced, and then the wave number distribution characteristic calculated from the measured two-dimensional image 150 is Fourier transformed to calculate the film thickness. The method of doing was illustrated. A method for calculating the film thickness by performing fitting between the theoretically generated two-dimensional image and the measured two-dimensional image 150 instead of such a method will be described.

  FIG. 19 is a schematic diagram for illustrating the processing contents in the processing procedure (part 2) of the film thickness measurement method according to the present embodiment. Each element shown in FIG. 19 is typically realized by the processor 102 of the processing apparatus 100 executing the measurement program 114.

Referring to FIG. 19, the processing device 100 includes buffers 152 and 156, a modeling module 154, and a fitting module 158. In the configuration shown in FIG. 19, the modeling module 154 calculates the theoretical value of the transmittance spectrum T (λ) (or the reflectance spectrum R (λ)) and obtains the acquired transmittance spectrum T (λ) ( Alternatively, the film thickness d ₁ that defines the theoretical value is adjusted so that the correlation with the actually measured value of the reflectance spectrum R (λ) is high. Finally, the transmittance spectrum T (λ) (or the reflectance spectrum R (λ) having the highest correlation with the measured value of the transmittance spectrum T (λ) (or the reflectance spectrum R (λ)). )) Is generated as a measurement result.

For convenience of explanation, the subscript “ _meas ” is added to the measured value of the transmittance spectrum or the reflectance spectrum, and the subscript “ _theo ” is added to the theoretical value of the transmittance spectrum or the reflectance spectrum.

  More specifically, the buffer 152 stores a two-dimensional image 150 (actual measurement value) captured by the measurement optical system 10. On the other hand, the buffer 156 stores a two-dimensional image (theoretical value) generated by the modeling module 154. The fitting module 158 performs a shape comparison (fitting) between the two-dimensional image 150 (actually measured value) stored in the buffer 152 and the two-dimensional image (theoretical value) stored in the buffer 156 to obtain the degree of similarity. And a parameter update command is output to the modeling module 154 so that the calculated similarity is maximized. As a similarity, the case where a correlation value or a correlation matrix is used is illustrated.

The fitting module 158 outputs the film thickness d ₁ (j) at that time as a measurement result when the calculated similarity is equal to or greater than a predetermined threshold value.

The modeling module 154 includes an initial value of the film thickness d ₁ (j), an optical constant considering the wavelength dispersion of the sample S (refractive index n ₁ (λ) and extinction coefficient k ₁ (λ)), and measurement. A wavelength conversion formula λ (i, j) determined by wavelength calibration of the optical system 10 and an incident angle θ ₀ (j) for each measurement point on the measurement line 24 are input. The modeling module 154 determines the transmittance distribution T _theo for the pixel position (i, j) and the film thickness d ₁ (j) based on the input information.
(I, j, d ₁ (j)) or reflectance distribution R _theo (i, j, d ₁ (j)) is calculated. In addition, the modeling module 154 updates the film thickness d ₁ (j) as appropriate in accordance with the parameter update command from the fitting module 158. For details of the transmittance distribution T _theeo and the reflectance distribution R _theo , see also the equation (20) described later.

As the refractive index n ₁ , the wavelength distribution of the refractive index of the polyethylene thin film measured with a microscopic spectrophotometer as shown in FIG. 13 was used. As an example of the refractive index n ₁ (λ) shown in FIG. 13, the Cauchy dispersion formula as shown in the above formula (11) is used.

In the example shown in FIG. 13, the coefficients C ₀ = 1.533731, C ₁ = 429.0333, and C ₂ = 2.09247 × 10 ⁸ .

FIG. 20 is a flowchart showing a processing procedure (No. 2) of the film thickness measuring method according to the present embodiment. Referring to FIG. 20, first, processing apparatus 100 calculates an incident angle θ ₀ corresponding to each measurement point for measurement interference light incident on measurement optical system 10 (step S100). That is, as a correction element corresponding to the incident angle from each measurement point to the measurement optical system, a value indicating the magnitude of the incident angle corresponding to each measurement point is used.

Next, the processing apparatus 100 is a refractive index n ₁ (λ) in consideration of the wavelength dispersion of the sample S from the measurement result of the sample S by a measuring apparatus (for example, a microscopic differential film thickness meter or the like) capable of optical constant analysis. Is acquired (step S102). Subsequently, the processing apparatus 100 calculates a wavelength conversion formula λ (i, j) indicating the relationship between the pixel position of the two-dimensional image 150 and the wavelength λ from the result of wavelength calibration for the measurement optical system 10 (step S104). ).

  Since the processing of steps S100 to S104 is the same as steps S100 to S104 in the flowchart of the processing procedure (part 1) of the film thickness measurement method shown in FIG. 11, detailed description will not be repeated. The processes in steps S100 to S104 described above correspond to a preparation process.

The processing apparatus 100 sets the sample S in the optical measurement apparatus 1 and acquires the two-dimensional image 150 captured in a state where the sample S is irradiated with the measurement light (step S130). That is, the processing apparatus 100 acquires the transmittance distribution T _meas (i, j) (or the reflectance distribution R _meas (i, j)).

Based on the information calculated in steps S100 to S104 and the initial value of the film thickness d ₁ (j), the processing apparatus 100 transmits the transmittance distribution T _theo (i, j, d ₁ (j)) (or reflectivity). Distribution R _theo (i, j, d ₁ (j))) is calculated (step S132). That is, the processing apparatus 100 uses the film thickness d ₁ (j) at each measurement point as a variable parameter, and also sets the refractive index n ₁ of the sample S and a value corresponding to the incident angle corresponding to each measurement point. The theoretical value of each pixel corresponding to the two-dimensional image 150 is calculated based on the correspondence between each measurement point and the pixel position (i, j) of the two-dimensional image.

Subsequently, the processing apparatus 100, the transmittance distribution _T meas acquired in step S130 (i, j) (or, reflectance distribution _R meas (i, j)) and the calculated transmittance distribution _{T theo} in step S132 ( i, j, d ₁ (j)) (or reflectance distribution R _theo (i, j, d ₁ (j))) is compared to calculate the similarity between the two (step S134).

More specifically, the processing apparatus 100 transmits the transmittance distribution T _meas (i, j) (or the reflectance distribution R _meas (i, j)) and the transmittance distribution T _theo (i, j, d ₁ ( j)) (or reflectance distribution R _theo (i, j, d ₁ (j))) and a correlation matrix or correlation coefficient. By using the correlation matrix, the similarity for each position direction pixel number j can be calculated. However, when it is estimated that the variation in the film thickness d ₁ (j) in the position direction is sufficiently small, it is assumed that d ₁ (j) = d ₁ and a one-dimensional value that summarizes the entire spectrum (ie, (Correlation value) may be calculated.

The processing device 100 determines whether or not the similarity calculated in step S134 is equal to or greater than a predetermined threshold (step S136). If the calculated degree of similarity is less than a predetermined threshold value (NO in step S136), processing apparatus 100 updates film thickness d ₁ (j) (step S138), and performs the processes in and after step S132. Repeat. The update to the film thickness d ₁ (j) may be performed according to the corresponding degree of similarity for each position direction pixel number j, or a predetermined amount may be uniformly added or subtracted.

If the calculated similarity is greater than or equal to a predetermined threshold value (YES in step S136), processing apparatus 100 aggregates the current film thickness d ₁ (j) on measurement line 24 of sample S. It outputs as a film thickness distribution (step S140).

  As described above, the processing apparatus 100 adjusts the variation parameter so that the similarity between the calculated theoretical value of each pixel and each pixel value of the two-dimensional image 150 is high, so that the film thickness at each measurement point is adjusted. To decide. That is, the variation parameter is adjusted so that a correlation close to a similarity is found between the calculated theoretical waveform and the actually measured actual waveform.

  Subsequently, the processing apparatus 100 determines whether or not a film thickness measurement termination condition for the sample S is satisfied (step S142). If the film thickness measurement termination condition for sample S is not satisfied (NO in step S142), processing device 100 repeats the processes in and after step S130.

  On the other hand, if the film thickness measurement termination condition for sample S is satisfied (YES in step S142), processing device 100 integrates the film thickness distributions sequentially calculated in step S140, and A film thickness distribution (in-plane distribution of film thickness) on the measurement surface is output (step S144). Then, the process ends.

As described above, in the processing procedure (part 2) of the film thickness measurement method, the measured value of the transmittance spectrum or the reflectance spectrum, which is the wavelength distribution characteristic acquired from the sample S, the incident angle θ ₀ , the refractive index. Shape comparison (fitting) between the transmittance spectrum or the theoretical value of the reflectance spectrum determined in accordance with a model equation (theoretical equation) having n ₁ (λ), wavelength λ, and film thickness d ₁ (j) as parameters. Thus, the film thickness (or film thickness distribution) of the sample S is determined.

More specifically, in the transmittance spectrum T _theo (i, j, d ₁ (j)) (or reflectance distribution R _theo (i, j, d ₁ (j))), the film thickness d ₁ (j ) While calculating the correlation matrix (or correlation value) between the transmittance distribution T _meas (i, j) (or the reflectance distribution R _meas (i, j)) and the most correlated. The film thickness d ₁ (j) that is high (that is, the correlation coefficient is closest to 1) is output as the final result.

The film thickness distribution (in-plane distribution of film thickness) of the sample S can be measured by the above processing.
In the processing procedure (part 2) of the above-described film thickness measurement method, the transmittance is calculated in consideration of the incident angle θ ₀ as a correction factor corresponding to the incident angle θ ₀ from the measurement point to the measurement optical system 10. Although the description has been given focusing on the distribution T _theo (i, j, d ₁ (j)) or the reflectance distribution R _theo (i, j, d ₁ (j)), the correction element is limited to this. is not. For example, the correction element is a concept may include the wave number K ₁ described above.

In steps S136 and S138 described above, the transmittance distribution T _theeo (i, j, d ₁ (j)) (or the reflectance distribution R _theo (i, j, d ₁ (j))) is changed. By setting the range and pitch of the film thickness parameter d ₁ (j), the transmittance distribution T _theo (i, j, d ₁ (j) with respect to the film thickness value d ₁ (j) within the set fluctuation range. ) (Or reflectance distribution R _theo (i, j, d ₁ (j))) is calculated in advance. Then, the transmittance distribution T _theo (i, j, d ₁ (j)) calculated in advance (or the reflectance distribution R _theo (i, j, d ₁ (j))) and the actually measured transmittance distribution. A correlation matrix or correlation coefficient is calculated with _{respect to} T _meas (i, j) (or reflectance distribution R _meas (i, j)), and a similarity (correlation) is calculated from the calculated results. The film thickness value d ₁ (j) that gives the highest (number) may be determined as the film thickness at each measurement point.

(E5: multilayer film sample)
For convenience of explanation, the process of measuring the film thickness of one layer has been mainly described, but the present invention is not limited thereto, and the film thickness of each layer of the multilayer film sample can be measured. The refractive index of each layer of the multilayer film sample can also be measured.

In the processing procedure (part 1) of the film thickness measurement method described above, when measuring the film thickness of each layer of the multilayer film sample, the wave number conversion transmittance T ′ or the wave number conversion reflectance R ′ is obtained by Fourier transform. A plurality of peaks corresponding to the film thickness of each layer appear in the power spectrum P (K ₁ ). By analyzing a plurality of peaks appearing in the power spectrum P (K ₁ ), the film thickness of each layer constituting the target sample can be calculated.

  Further, in the processing procedure of the film thickness measurement method (part 2), when measuring the film thickness of each layer of the multilayer film sample, the optical constants (refractive index and extinction coefficient) of each layer considering wavelength dispersion, and By performing fitting on each layer using a model formula including the film thickness of each layer, the film thickness of each layer constituting the target sample can be calculated.

(E6: Inline measurement / offline measurement)
In the above description, the processing example in which the film thickness measurement is performed following the imaging of the two-dimensional image of the sample S is mainly shown. However, the present invention is not limited to such in-line measurement or real-time measurement. A two-dimensional image of the sample S may be sequentially captured, and a film thickness trend (in-plane distribution of film thickness) may be output afterwards.

<F. Refractive index measurement method>
In the above-described film thickness measurement method, the refractive index n ₁ (λ) of the sample S is measured in advance using a microspectrophotometer or the like, but an optical measurement device according to the present embodiment is used. Thus, the refractive index n ₁ (λ) of the sample S can also be measured.

(F1: Overview)
First, a small piece (for example, 1 mm square) of the same sample S is arranged at each measurement point on the measurement line, and an actual measurement value (transmittance distribution T _meas (i, j) or reflectance distribution R _meas ( i, j)) are obtained sequentially. That is, for the same sample S, the transmittance spectrum or reflectance spectrum in the wavelength direction when the position direction pixel number j (that is, the incident angle θ ₀ ) is varied is measured.

Applying prior information of the same film thickness d ₁ to the transmittance distribution T _meas (i, j) (or reflectance distribution R _meas (i, j)) measured from the same sample S Thus, the unknown refractive index n ₁ (λ) is determined.

21 and 22 are schematic diagrams for illustrating the outline of the refractive index measurement method according to the present embodiment. FIG. 21 shows an example in which the refractive index n ₁ (λ) of the sample S is measured using the intensity distribution at a specific position direction pixel number j. FIG. 22 shows an example in which the refractive index n ₁ (λ) of the sample S is measured using the intensity distribution at a specific wavelength direction pixel number i.

Referring to FIG. 21, when attention is paid to the position direction pixel number j, the refractive index n ₁ (λ) of the sample S is set to a provisional value, and the transmittance distribution T at a specific position direction pixel number j is set. Film thickness d ₁ (j) (j = j ₁ , j ₂ , j ₃ ,...) is calculated from _meas (i, j) (or reflectance distribution R _meas (i, j)). Here, since the film thickness d ₁ is the same, the refractive index n ₁ (λ) of the sample S is determined so that the calculated film thicknesses d ₁ (j) match.

Referring to FIG. 22, when focusing on a specific wavelength direction pixel number i, unknown information is obtained by applying prior information of the same film thickness d ₁ to the difference between the theoretical value and the actual measurement value. The refractive index n ₁ (λ) of is determined.

More specifically, first, the transmittance distribution T _theo (i, j, d ₁ , n ₁ (i)) (or the reflectance distribution R _theo (i, j, d ₁ , n ₁ (i))) And the transmittance distribution T _meas (i, j) (or the reflectance distribution R _meas (i, j)) at the corresponding position direction pixel number j.

The transmittance distribution T _theo (or the reflectance distribution R _theo ) is a value that depends on the pixel position (i, j), the film thickness d ₁ , and the refractive index n ₁ (λ). Pixel location (i, j) are known, the thickness d ₁ is the same irrespective of the wavelength direction pixel number i. Therefore, in advance, the film thickness d ₁ is the same as the comparison result between the theoretical value and the actual measurement value for the plurality of wavelength direction pixel numbers i (i = i ₁ , i ₂ , i ₃ ,...). By applying the information, the refractive index n ₁ (λ) can be determined.

Referring to FIG. 22, similarly to the case where attention is paid to the position direction pixel number j, the refractive index n ₁ (λ) of the sample S is set to the provisional value even when attention is paid to the specific wavelength direction pixel number i. After setting, the film thickness d ₁ (i) (i = i ₁ ) from the transmittance distribution T _meas (i, j) (or the reflectance distribution R _meas (i, j)) at a specific wavelength direction pixel number i. , I ₂ , i ₃ ,... Here, since the film thickness d ₁ is the same, the refractive index n ₁ (λ) of the sample S is determined so that the calculated film thicknesses d ₁ (i) coincide with each other.

The outline of the refractive index measurement method according to the present embodiment will be described as follows. That is, according to the refractive index measurement method according to the present embodiment, the refractive index n _{1 of the} sample S can be obtained using the optical measurement device according to the present embodiment without using a dedicated measurement device such as a microspectrophotometer. Can be measured.

For example, after setting an arbitrary refractive index n ₁ to an arbitrary initial value (for example, 1 for all wavelengths) according to an arithmetic process having a correction function of the incident angle θ ₀ as described above, The film thickness trend is acquired by changing the arrangement position. When the set value of the refractive index n ₁ is different from the actual refractive index of the sample S, the film thickness trend does not become flat. By using a least squares method, the refractive index n ₁ by appropriately changing a thickness trend that most becomes flat, i.e. the refractive index n ₁ as the film thickness variance is minimized it can be determined.

The refractive index n ₁ may be obtained as a constant of the average of all wavelengths, or when it is desired to obtain it more strictly, considering chromatic dispersion, for example, Cauchy's dispersion formula n ₁ (λ) = E + (F / λ ² ) + (G / λ ⁴ ), and the coefficient of each term may be obtained by the least square method or the like.

  Alternatively, attention may be paid to a specific line having a relatively large incident angle, and the refractive index may be calculated by a method that minimizes the residual square value of the film thickness.

In the present embodiment, the same measurement standard is sequentially arranged at the measurement points of the sample S irradiated with the measurement light, and the actual measurement values at the measurement points are sequentially acquired, thereby the actual measurement corresponding to the two-dimensional image 150. A value distribution (transmittance distribution T _meas (i, j) or reflectance distribution R _meas (i, j)) is acquired. Further, in association with the region on the two-dimensional image 150 corresponding to each measurement point of the sample S irradiated with the measurement light, a correction element corresponding to the incident angle from each measurement point to the measurement optical system 10 (for each measurement point) The value indicating the magnitude of the corresponding incident angle) is calculated. Based on the pixel value group for one or a plurality of columns along any one direction of the actually measured value distribution and the corresponding correction element, the optical characteristic including the refractive index of the measurement reference is calculated. At this time, the measured value distribution, a priori information between the film thickness d ₁ is the same is utilized.

Hereinafter, each case will be described in more detail.
(F2: Refractive Index Measurement Method Based on Wavelength Direction Information (Part 1))
First, a refractive index measurement method (part 1) based on information in the wavelength direction will be described. First, the case where n ₁ (λ) = n ₁ (constant value) will be described _first without considering the wavelength dependence of the refractive index of the sample S.

In the refractive index measurement method (part 1) based on the information on the wavelength direction, the refractive index n ₁ of the sample S is set to some initial value, and a plurality of wavelength direction pixel numbers i are set based on the refractive index n ₁ . transmission spectrum (or reflectance spectra) will be calculated thickness d _1, respectively. Then, to evaluate the thickness trend of a plurality of film thickness d ₁ that is calculated. As the thickness trend is flat, fitting the refractive index n _1. That is, if the actual refractive index n ₁ of the sample S is different from the set refractive index n ₁ , the film thickness trend cannot be kept flat. This is because the calculation method in consideration of the influence of the incident angle θ ₀ of the measurement interference light as described above is adopted, and the film thickness of the fixed point of the sample S is constant regardless of the incident angle θ _0. Based on the assumption that it should be a value.

In accordance with such a premise, the flatness of the film thickness trend is adopted as a cost function, and the refractive index n ₁ that minimizes the value of this cost function is determined. When the fixed point of the sample S is measured at the position direction pixel number j, the film thickness is defined as d ₁ (j). The film thickness trend curve d ₁ (j) when the position direction pixel number j is changed is approximated to a constant function f (j) = μ (μ is a constant value). At this time, the residual sum of squares S can be defined as the following equation (12). The value of the constant value μ in the equation (12) is determined by the least square method. More specifically, when a condition that minimizes the residual sum of squares S, that is, a constant value μ when ∂S / ∂μ = 0 holds, the following equation (13) is obtained.

The constant value μ calculated according to the equation (12) corresponds to the average value of the film thickness d ₁ (j). Then, the residual sum of squares S is, since the residual square sum to the average value of the film thickness d _{1 (j),} the variance of the film thickness d _{1 (j)} (hereinafter, also referred to as "film thickness dispersion".) To Equivalent to.

Next, since the film thickness trend, the film thickness average value, and the film thickness dispersion (the residual sum of squares of the film thickness) all depend on the refractive index n ₁ , the film thickness d ₁ (j) is expressed as a cost function. The film thickness dispersion D (n ₁ ) can be defined as in the following formula (14).

With respect to the above equation (14), the refractive index n ₁ is sequentially changed to change the refractive index when the condition that the film thickness dispersion D (n ₁ ) is minimized, that is, ∂D / ∂n ₁ = 0 holds. n ₁ can be determined.

FIG. 23 is a graph showing an example of the film thickness trend calculated according to the refractive index measurement method (part 1) based on the information in the wavelength direction according to the present embodiment. The graph in FIG. 23A shows a film thickness trend for each refractive index n ₁ (constant value) of the polyethylene thin film. In order to facilitate comparison of film thickness trend changes, the graph of FIG. 23B shows the result of normalization so that the film thickness d ₁ = 10 [μm] at the position direction pixel number j = 600. In calculating the film thickness trend of FIG. 23, the theoretical value of the transmittance spectrum T (λ) used when generating the two-dimensional image shown in FIG. 15 was used.

As the refractive index n ₁ of the sample S is increased from 1.51 to 0.01, the film thickness trend changes from convex downward to convex upward at an increase from 1.56 to 1.57. I understand that It can also be seen that when the refractive index n ₁ = 1.56, the film thickness dispersion D (n ₁ ) is the smallest and the film thickness trend is the flattest. Values relating to the film thickness trend shown in FIG. 23 are shown in the following table.

From the above calculation results, if the refractive index n _{1 is obtained} with an accuracy of 1/100, it can be determined that the refractive index n ₁ = 1.56.

In the above-described equations (12), (13), and (14), it is described that all the position direction pixel numbers j are used when calculating the film thickness dispersion D (n ₁ ). However, it is not always necessary to use all of them, and a predetermined number of pixel columns may be used according to the required accuracy. In this case, the measurement points at which the sample S is arranged may be arranged at coarse intervals as compared with the resolution in the film thickness measurement method.

FIG. 24 is a flowchart showing a processing procedure of the refractive index measurement method (part 1) based on information in the wavelength direction according to the present embodiment. Referring to FIG. 24, first, the user arranges a small piece of sample S at each measurement point on the measurement line, and obtains an actual measurement value from the arranged sample S by operating optical measurement apparatus 1. Repeat (step S200). Thereby, the processing apparatus 100 acquires the transmittance distribution T _meas (i, j) (or the reflectance distribution R _meas (i, j)) measured from the same sample S.

Subsequently, the processing apparatus 100 calculates an incident angle θ ₀ corresponding to each measurement point for the measurement interference light incident on the measurement optical system 10 (step S202). Next, the processing apparatus 100 calculates a wavelength conversion formula λ (i, j) indicating the relationship between the pixel position of the two-dimensional image 150 and the wavelength λ from the result of wavelength calibration for the measurement optical system 10 (step S204). ).

  Since the processes in steps S202 and S204 are the same as steps S100 and S104 in the flowchart of the refractive index measurement method processing procedure (part 1) shown in FIG. 11, detailed description thereof will not be repeated.

Subsequently, the processing apparatus 100, the refractive index _{n 1} is set to an arbitrary initial value (step S206). Then, the processing apparatus 100 determines the film thickness d ₁ (j) from the transmittance distribution T _meas (i, j) (or the reflectance distribution R _meas (i, j)) for each of the plurality of position-direction pixel numbers j. Calculate (step S208).

The processing apparatus 100 calculates an average value of the film thickness from each calculated film thickness d ₁ (j) (step S210), and uses the average value of the film thickness calculated in step S210 to determine the film thickness dispersion D. (N ₁ ) is calculated (step S212). Then, the processing unit 100 determines whether the change in the predetermined range of the refractive index n ₁ is completed (step S214). If the change in refractive index n ₁ within the predetermined range has not been completed (NO in step S214), processing device 100 updates refractive index n ₁ (step S216), and repeats the processes in and after step S206. .

If the change in refractive index n ₁ within a predetermined range is completed (YES in step S214), processing device 100 determines the smallest _{one of} film thickness dispersions D (n ₁ ) calculated in step S212. The refractive index n ₁ corresponding to the determined film thickness dispersion D (n ₁ ) is determined as the refractive index of the sample S (step S220). Then, the process ends.

As described above, the refractive index n ₁ of the sample S can be determined based on the information in the wavelength direction.
(F3: Refractive index measurement method based on wavelength direction information (2))
In the refractive index measurement method (part 1) based on the information on the wavelength direction described above, the method of analytically determining the refractive index n ₁ of the sample S is exemplified, but the refractive index is determined by fitting using a predetermined polynomial. n ₁ may be determined.

FIG. 25 is a diagram for describing a method of determining the refractive index n ₁ in the refractive index measurement method (part 2) based on information in the wavelength direction according to the present embodiment. FIG. 25 shows a graph in which the film thickness dispersion D (n ₁ ) calculated when the refractive index n ₁ is changed is plotted. For example, a cubic polynomial as shown in the following equation (15) can be fitted to the relationship between the refractive index n ₁ and the film thickness dispersion D (n ₁ ) as shown in FIG.

That is, the coefficients A ₃ , A ₂ , A ₁ , A ₀ of the equation (15) are fitted so as to pass through the respective film thickness dispersions D (n ₁ ) shown in FIG. The refractive index n ₁ can be determined corresponding to the point at which the fitted cubic polynomial as shown in FIG. 25 takes the minimum value (minimum value). That is, the refractive index n ₁ can be calculated based on the coefficients A ₃ , A ₂ , A ₁ , A ₀ according to the following equation (16).

  The results obtained by the fitting shown in FIG. 25 are shown in the following table.

From the above calculation results, when the refractive index n _{1 is obtained} with an accuracy of 1/10000, it can be determined that the refractive index n ₁ = 1.5634. By using a polynomial for the fitting, the refractive index can be determined with higher accuracy than the fluctuation range of the refractive index n ₁ (in this example, in increments of 0.01 (that is, accuracy of 1/100)).

  In the processing procedure of the refractive index measurement method (part 2) based on the information in the wavelength direction, fitting using a polynomial as shown in FIG. 25 is executed instead of steps S218 and S220 in the flowchart shown in FIG. Since the other points are the same as those in the refractive index measurement method (part 1) based on the information in the wavelength direction described above, detailed description will not be repeated.

(F4: Refractive index measurement method based on wavelength direction information (part 3))
By focusing on the information in the wavelength direction as described above that is more greatly affected by the incident angle θ ₀ , the refractive index n ₁ of the sample S can be determined more efficiently. More specifically, first, a residual square value y (n ₁ , j), which is a deviation from an average value of the film thickness d ₁ (j) at an arbitrary pixel number j in the position direction, is expressed by the following equation (17). Specify as shown.

FIG. 26 is a diagram for describing a method of determining the refractive index n ₁ in the refractive index measurement method (part 3) based on information in the wavelength direction according to the present embodiment. FIG. 26 shows a graph in which the residual square value y calculated when the refractive index n ₁ is changed is plotted. With respect to the relationship between the refractive index n ₁ and the residual square value y as shown in FIG. 26, for example, a cubic polynomial as shown in the above equation (15) can be fitted.

That is, each coefficient A in equation (15) is passed through each residual square value y shown in FIG.
_3. Fitting A ₂ , A ₁ , A ₀ The refractive index n ₁ can be determined corresponding to the point at which the fitted cubic polynomial as shown in FIG. 26 takes the minimum value (minimum value). That is, the refractive index n ₁ can be calculated based on the coefficients A ₃ , A ₂ , A ₁ , A ₀ according to the above equation (16).

That is, the coefficients A ₃ , A ₂ , A ₁ , A ₀ of the equation (15) are fitted so as to pass through the respective residual square values y shown in FIG. The refractive index n ₁ can be determined corresponding to the point at which the fitted cubic polynomial as shown in FIG. 26 takes the minimum value (minimum value). The results obtained by the fitting shown in FIG. 26 are shown in the following table.

In the above table, the refractive index n ₁ was calculated for four position direction pixel numbers (j = 50, 100, 1100, 1150) that are considered to have a relatively large incident angle θ ₀ .

In the refractive index measurement method (part 3) based on the information in the wavelength direction according to the present embodiment, it is only necessary to measure the spectrum by arranging the sample S at at least two measurement points. The refractive index n ₁ can be easily calculated. At this time, the measurement point where the sample S is arranged is preferably selected at a position where the angle difference between the incident angles is large.

Furthermore, focusing on a specific position direction pixel number j having the largest possible incident angle, the refractive index n ₁ can also be calculated from the point where the residual square value y takes a minimum value (minimum value). For example, when the refractive index n ₁ is obtained with an accuracy of 1/10000 for the position direction pixel number j = 50 in the above table, the refractive index n ₁ = 1.5587 can be determined. By using a polynomial for the fitting, the refractive index can be determined with higher accuracy than the fluctuation range of the refractive index n ₁ (in this example, in increments of 0.01 (that is, accuracy of 1/100)).

The processing procedure of the refractive index measurement method ( _No. 3) based on the information in the wavelength direction is compared with the processing procedure of the refractive index measurement method ( _No. 2) based on the information in the wavelength direction described above. Only the point that the residual square value y is used instead of. Since the other points are the same as those in the refractive index measurement method (part 1) based on the information in the wavelength direction described above, detailed description will not be repeated.

(F5: Refractive index measurement method based on wavelength direction information (part 4))
In the description of the refractive index measurement method (part 1) based on the information in the wavelength direction described above, it is assumed that the refractive index n ₁ (λ) = n ₁ (constant value). However, in practice, the refractive index n ₁ (λ) has wavelength dependency. In this case, the refractive index n ₁ (λ) is defined using a polynomial, and the refractive index n ₁ (λ) in consideration of wavelength dependence can be determined by using each coefficient of the polynomial as a fitting target.

As an example, a Cauchy dispersion formula as shown in the following formula (18) may be used. While changing the coefficient (E, F, G) of each term of the equation (18), the value of the evaluation object (in the refractive index measurement method (part 2) based on the information on the wavelength direction described above), the film thickness dispersion D ( A set of coefficients for which n ₁ )) takes a minimum value (minimum value) may be determined by a least square method or the like.

That is, the coefficient satisfying ∂D / ∂E = ∂D / ∂F = ∂D / ∂G = 0 in the following equation (19) in which the film thickness dispersion D is made to depend on the coefficients (E, F, G). By obtaining a set of (E, F, G), the refractive index n ₁ (λ) considering wavelength dependence can be obtained from the equation (18).

Further, in the refractive index measurement method (part 3) based on the information in the wavelength direction described above, the refractive index n ₁ is calculated from the point where the residual square value y takes the minimum value (minimum value). Also, the refractive index n ₁ (λ) can be obtained in consideration of wavelength dependency. Similar to the above equation (19), y (D, E, F) is derived, and coefficients (E, F, G) satisfying ∂y / ∂E = ∂y / ∂F = ∂y / ∂G = 0. ), The refractive index n ₁ (λ) considering wavelength dependence can be obtained from the equation (18).

The refractive index n ₁ (λ) considering the wavelength dependency can be obtained by the procedure as described above.

(F6: Refractive Index Measurement Method Based on Position Direction Information (Part 1))
Next, a refractive index measurement method based on position direction information will be described. As described above with reference to FIG. 22, when the refractive index n ₁ of the sample S is measured based on the position direction information, the transmittance distribution T _meas ( i, j) (or reflectance distribution R _meas (i, j)) and a transmittance distribution T _theo (i, j, n ₁ (i)) calculated according to a function including the refractive index n ₁ (or For the reflectance distribution R _theo (i, j, n ₁ (i))), for one or more wavelength direction pixel numbers i, by comparing the trend along the position direction pixel number j, the refractive index n (i ).

First, for the sample S (film thickness d ₁ ) of the thin film in the air (medium 0) as shown in FIG. 7 described above, the transmittance distribution T _theo when considering multiple reflections in the sample S is as follows: (20). Note that the refractive index of air n ₀ = 1.

In the above equation (20), the amplitude reflectance r ₀₁ is expressed by the following equation (21-1) for each of s-polarized light and p-polarized light. Furthermore, since the relationship of n ₀ · sin θ ₀ = n ₁ · sin θ ₁ (Snell's law) is established between the incident angle θ ₀ and the refraction angle θ ₁ , the equation (21-1) is expressed as (21− 2) It can be transformed as shown in the equation. Note that the refractive index of air n ₀ = 1. That is, the amplitude reflectance r ₀₁ can be defined only by the refractive index n ₁ and the incident angle θ ₀ of the sample S.

Here, the intensity reflectance R ₀₁ = | r ₀₁ | ² in the case where no polarization occurs includes both s-polarized light and p-polarized light components, and therefore can be defined as in the following equation (22).

By substituting the equations (21-2) and (22) into the above equation (20) and erasing the amplitude reflectance r ₀₁ , the transmittance distribution T _theo is the refractive index n ₁ (i) of the sample S. , Incident angle θ ₀ (j), film thickness d _{1 of} sample S, and wavelength λ (i).

As described above, the transmittance distribution T _meas (i, j) (or the reflectance distribution R _meas (i, j)) is measured by placing a small piece of the same sample S at each measurement point on the measurement line. The film thickness d ₁ is a constant value independent of the wavelength direction pixel number i and the position direction pixel number j.

Focusing on a specific wavelength direction pixel number i, the transmittance distribution T _meas (i, j) (or reflectance distribution R _meas (i, j)) and the transmittance distribution T _meas (i, j) (or reflection) A residual sum of squares Q indicating an error from the rate distribution R _meas (i, j)) can be defined as the following equation (23).

The wavelength direction pixel of interest is solved by solving the condition for minimizing the residual sum of squares Q defined in the above equation (23), that is, ∂Q / ∂d ₁ = ∂Q / ∂n ₁ (i) = 0. The refractive index n ₁ (i) and the corresponding film thickness d ₁ for number i can be determined.

As described above, the film thickness d ₁ are the constant value independent of the wavelength direction pixel number i and the position-direction pixel number j, without considering the wavelength dependence of the refractive index n ₁ (i.e., the refractive index n ₁ May be a constant value), the refractive index n ₁ (i) calculated for one wavelength direction pixel number i and the corresponding film thickness d ₁ may be output as final values.

In order to further improve the measurement accuracy, a set of the refractive index n ₁ and the film thickness d ₁ may be determined for all the wavelength direction pixel numbers i. In this case, the result of statistical processing such as averaging the values of the set of the refractive index n ₁ and the film thickness d ₁ may be finally output.

As will be described in refractive index measurement method based on the position direction information will be described later (Part 3), when considering the wavelength dependence of the refractive index n ₁ is directed to a plurality of wavelengths direction pixel number i There is a need.

  In the above equation (23), it is described that all of the position direction pixel numbers j are used when calculating the residual sum of squares Q, but it is not always necessary to use all of them, and is required. A predetermined number of pixel columns may be used depending on the accuracy. In this case, the measurement points at which the sample S is arranged may be arranged at coarse intervals as compared with the resolution in the film thickness measurement method.

  The processing procedure of the refractive index measurement method (part 1) based on the position direction information is the same as the processing procedure of the refractive index measurement method (part 1) based on the wavelength direction information shown in FIG. Detailed description will not be repeated.

(F7: Refractive Index Measurement Method Based on Position Direction Information (Part 2))
In the refractive index measurement method (part 1) based on the information on the position direction described above, the refractive index n ₁ and the film thickness d ₁ are determined for each wavelength direction pixel number i by comparing the measured value with the theoretical value. .

In the refractive index measurement method (part 2) based on the information on the position direction, the refractive index n ₁ is determined with higher accuracy by applying prior information with the same film thickness d ₁ . More specifically, the flatness of the film thickness trend may be adopted as a cost function, and the refractive index n ₁ that minimizes the value of this cost function may be determined. The film thickness trend curve d ₁ (i) when the pixel number i in the wavelength direction is changed is approximated to a constant function f (j) = μ (μ is a constant value). At this time, the residual sum of squares S can be defined as the following equation (24). The value of the constant value μ in the equation (24) is determined by the least square method. More specifically, when a condition that minimizes the residual sum of squares S, that is, a constant value μ when ∂S / ∂μ = 0 holds, the following equation (25) is obtained.

The constant value μ calculated according to the equation (25) corresponds to the average value of the film thickness d ₁ (i). Then, the residual sum of squares S is, since the residual square sum to the average value of the film thickness d _{1 (i),} the dispersion of the film thickness d _{1 (i)} (hereinafter, also referred to as "film thickness dispersion".) To Equivalent to.

Figure 27 is a diagram for explaining a method of determining a more probable value of the film thickness d ₁ of the refractive index measurement method based on the position direction of the information according to the present embodiment (Part 2). The film thickness d ₁ can be calculated as many as the number of wavelength direction pixel numbers i, but the film thickness d ₁ calculated for each wavelength direction pixel number i should be the same value. Therefore, as shown in FIG. 27, the film thickness trend curve d ₁ (i) when the wavelength direction pixel number i is changed is approximated to a constant function f (i) = μ (μ is a constant value). Then, the constant μ is determined so that the residual sum of squares S of d ₁ (i) and f (i) takes a minimum value (minimum value). So determined constant μ has a thickness d ₁ more probable.

Next, a more probable film thickness d ₁ = μ is given to the term indicating the theoretical value of transmittance in the equation (25), and the residual square between the theoretical value of transmittance and the measured value of transmittance is calculated. The sum Q is defined as the following equation (26).

By solving the condition that the residual sum of squares Q defined in the above equation (26) is the minimum, that is, 式 Q / ∂n ₁ (i) = 0, the refractive index n for the focused wavelength direction pixel number i ₁ (i) and the corresponding film thickness d ₁ can be determined.

As described above, the film thickness d ₁ are the constant value independent of the wavelength direction pixel number i and the position-direction pixel number j, without considering the wavelength dependence of the refractive index n ₁ (i.e., the refractive index n ₁ May be a constant value), the refractive index n ₁ (i) calculated for one wavelength direction pixel number i and the corresponding film thickness d ₁ may be output as final values.

In order to further improve the measurement accuracy, a set of the refractive index n ₁ and the film thickness d ₁ may be determined for all the wavelength direction pixel numbers i. In this case, the result of statistical processing such as averaging the values of the set of the refractive index n ₁ and the film thickness d ₁ may be finally output.

As will be described in refractive index measurement method based on the position direction information will be described later (Part 3), when considering the wavelength dependence of the refractive index n ₁ is directed to a plurality of wavelengths direction pixel number i There is a need.

In the above equation (26), when calculating the residual sum of squares Q, it is described that all of the position direction pixel numbers j are used, but it is not always necessary to use all of them.
A predetermined number of pixel columns may be used according to the required accuracy. In this case, the measurement points at which the sample S is arranged may be arranged at coarse intervals as compared with the resolution in the film thickness measurement method.

  The processing procedure of the refractive index measurement method (part 1) based on the information on the position direction is the same as the processing procedure of the refractive index measurement method (part 1) based on the information on the wavelength direction shown in FIG. Do not repeat.

(F8: Refractive index measurement method based on position direction information (part 3))
In the description of the refractive index measurement methods (No. 1) and (No. 2) based on the information on the position direction, the case of refractive index n ₁ (λ) = n ₁ (constant value) is assumed. However, in practice, the refractive index n ₁ (λ) has wavelength dependency. In this case, the refractive index n ₁ (λ) is defined using a polynomial, and the refractive index n ₁ (λ) in consideration of wavelength dependence can be determined by using each coefficient of the polynomial as a fitting target.

As an example, a Cauchy dispersion formula as shown in the above formula (18) may be used. In this case, the coefficient (E, F, G) of each term of the equation (18) can be determined by calculating the refractive index n ₁ (i) for at least three points of the wavelength direction pixel number i.

In addition, when the Cauchy dispersion formula is used, a plurality of data strings in the position direction in each of the wavelength direction pixel numbers i can be handled collectively. That is, for a plurality of different wavelength direction pixel numbers i, the film thickness d ₁ and the coefficients (E, F, G) are handled as common parameters independent of the wavelength direction pixel numbers i, and these four parameters are changed. Further, in addition to the position direction pixel number j, the coefficient (E, F, G) can be determined so that the residual square sum Q including the sum related to the plurality of wavelength direction pixel numbers i is minimized.

More specifically, a set of film thickness d ₁ and coefficients (E, F, G) satisfying ∂Q / ∂d ₁ = ∂Q / ∂E = ∂Q / ∂F = ∂Q / ∂G = 0. You may ask for. As an algorithm at this time, Gauss-Newton method, steepest descent method, Levenberg-Marquardt method, or the like can be used.

As another method, the least square method is applied to the above equation (26) to calculate the refractive index n ₁ (i) for each wavelength direction pixel number i, and then the calculated wavelength direction pixel number i is calculated. Refractive index wavelength dependence n ₁ (λ) is determined by aggregating the refractive indices n ₁ (i) (i = 1, 2, 3,..., C _x / B _x ) for each. You can also. In this method, it is not necessary to specify the function form (model formula) of the wavelength dependency of the refractive index.

The refractive index n ₁ (λ) considering the wavelength dependency can be obtained by the procedure as described above.

<G. Application example>
Next, an application example of the optical measurement device according to the present embodiment will be described.

For example, the optical measurement device according to the present embodiment can be measured in-line by being disposed on a film production line or the like. The optical measurement apparatus according to the present embodiment can output an in-plane distribution of sample film thickness (that is, a two-dimensional distribution of film thickness). For example, from the change in film thickness trend in the sample transport direction, that is, the MD direction (Machine Direction),
It is also possible to specify a defective portion that may occur in the film production line.

More specifically, for example, the film production line has a plurality of conveyance rollers, and any one of the conveyance rollers has a defective portion such as formation of a convex portion on the film or contamination of the roller surface. Suppose you were. In this case, it is considered that the film thickness changes in a cycle depending on the roller radius (or circumferential length) of the transport roller and the film winding length of the transport roller. Identify defects on the film production line from the periodicity of changes that occur in the film thickness trend in the MD direction (thickness fluctuations, streaks and unevenness, local unevenness, etc.) Can do.

  In this way, defect inspection or the like can be performed by utilizing the periodicity of the film thickness trend output by the optical measurement device according to the present embodiment.

  The optical measurement device according to the present embodiment is not limited to the application as described above, and can be used for any purpose.

  For example, in-line film thickness measurement targets include semiconductors, functional films, plastics, and various filters.

<H. Other Embodiments>
(H1: Determination of angle of view and center position from measured film thickness value)
In the above description, the angle of view φ (= Atan (b / 2f)) of the measurement optical system 10 is theoretically determined from the length b of the image sensor 160 on the catalog specification and the focal length f of the objective lens 12. It was assumed that.

However, depending on the type of the objective lens 12 that is actually used, the effective focal length f ′ may slightly deviate from the catalog value of the focal length f due to lens distortion or a change in the focus degree. In addition, when optical adjustment is performed, it may be a little difficult to make the pixel center position (j = C _y / 2B _y ) coincide with the imaging center of the imaging unit 16.

In such a case, the effective value of the angle of view and the center position may be determined from the actually measured value of the film thickness. For example, with respect to the same sample S, the transmittance spectrum or reflectance spectrum in the wavelength direction when the position direction pixel number j (that is, the incident angle θ ₀ ) is varied is measured. Then, the film thickness is respectively calculated according to the processing procedure (part 1) of the above-described film thickness measurement method without correcting the incident angle θ ₀ for the measured transmittance spectrum or reflectance spectrum. By such a procedure, a film thickness trend indicating a change in film thickness corresponding to each measurement point can be obtained.

Then, the film thickness value at the center position of the pixel is normalized to 1 with respect to the obtained film thickness trend, and then fitting with a function such as y = cosA (x−x0) Based on the actual measurement value, the effective angle of view φ ′ (= Atan (b / 2f ′)) and the center position x ₀ can be calculated.

(H2: Parallel arrangement of a plurality of optical measuring devices)
When the optical measuring device according to the present embodiment is arranged on a film production line or the like, it is assumed that a plurality of optical measuring devices according to the present embodiment are arranged in parallel according to the line width of the film. In such a case, a portion overlapping the measurement range of another measurement optical system 10 arranged adjacent to the measurement optical system 10 may be formed near the end of the measurement range of the measurement optical system 10. That is, it is assumed that the same point of the sample S is included in the measurement range of the plurality of measurement optical systems 10. In such a case, the measurement results for the same point of the sample S output from the respective optical measurement devices may be different from each other. Since such inconsistency is not preferable in line management, the following correction method may be employed to match the measurement results.

  In the optical measurement apparatus according to the present embodiment, the influence of the incident angle of the measurement interference light can be eliminated, so that the respective film thicknesses calculated for the same point of the sample are constant regardless of the incident angle. For example, each optical measurement device measures the film thickness of a small piece (for example, 1 mm square) of the same sample S, and measures each film thickness (for example, at a measurement point where the incident angle becomes zero). What is necessary is just to set offset correction and / or a coefficient, etc. to each optical measuring device so that consistency may be taken between measurement values.

<I. Advantage>
As described above, according to the present embodiment, the in-plane distribution of film thicknesses of various samples can be measured at higher speed and with higher accuracy. Further, according to the present embodiment, it is possible to measure the optical characteristics of the sample such as the refractive index without using a dedicated measuring device.

  The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

  1, 2 Optical measurement device, 4 Base member, 6 Support member, 10 Measurement optical system, 12 Objective lens, 14 Imaging spectroscope, 16 Imaging unit, 20, 174 Light source, 22 Line light guide, 24 Measurement line, 28 Vertical direction , 100 processor, 102 processor, 104 main memory, 106 input unit, 108 display unit, 110 storage, 112 operating system, 114 measurement program, 116 two-dimensional image data, 118 measurement result, 120 communication interface, 122 network interface, 124 Media drive, 126 recording medium, 142 slit, 144 first lens, 146 diffraction grating, 148 second lens, 150 two-dimensional image, 152,156 buffer, 154 modeling module, 158 I Tsu computing module, 160 imaging device, 170 positioning mechanism, 172 a shutter.

Claims (11)

An irradiation optical system that linearly irradiates measurement light having a predetermined wavelength range with respect to the measurement target, and linear measurement interference light that is transmitted light or reflected light generated from the measurement target by irradiation of the measurement light. An optical measurement method using an optical measurement device including a measurement optical system that expands a wavelength in a direction orthogonal to a longitudinal direction of measurement interference light and outputs a two-dimensional image,
Obtaining a measured value distribution when the incident angle is varied for the same sample;
Calculating a correction element according to an incident angle from each measurement point to the measurement optical system in association with an area on the two-dimensional image corresponding to each measurement point of the measurement target irradiated with the measurement light; ,
Calculating an optical characteristic including a refractive index of the sample based on a pixel value group for one or a plurality of columns along any one direction of the measured value distribution and a corresponding correction element; Optical measurement method. Calculating the optical characteristic comprises:
Calculating a film thickness of each of the plurality of positions of the measured value distribution based on a set refractive index, a correction element corresponding to each position, and a pixel value group in the wavelength direction at each position; ,
Calculating a film thickness dispersion that is a dispersion for each of the calculated film thicknesses;
Setting the refractive index of the sample to a plurality of different values, respectively, repeating the steps of calculating the film thickness and calculating the film thickness dispersion;
The optical measurement method according to claim 1, further comprising: determining a refractive index of the sample based on the calculated film thickness dispersion.   The optical measurement method according to claim 2, wherein the step of determining the refractive index of the sample includes a step of determining a refractive index that reduces the calculated film thickness dispersion as the refractive index of the sample. Determining the refractive index of the sample comprises:
Fitting a polynomial indicating a predetermined film thickness dispersion with respect to the relationship between the refractive index and the film thickness dispersion;
The optical measurement method according to claim 2, further comprising: determining a refractive index of the sample based on a point at which the film thickness dispersion represented by the polynomial determined by the fitting takes an extreme value. Determining the refractive index of the sample comprises:
Fitting a polynomial representing a predetermined residual square value with respect to the relationship between the refractive index and the residual square value for each calculated film thickness;
The optical measurement method according to claim 2, further comprising: determining a refractive index of the sample based on a point where a residual square value represented by a polynomial determined by fitting takes an extreme value. The refractive index of the sample is calculated according to a predetermined wavelength dispersion formula,
Determining the refractive index of the sample comprises:
The least square method is applied to either the relationship between each coefficient that defines the chromatic dispersion formula and the film thickness dispersion or the relationship between each coefficient that defines the chromatic dispersion formula and the residual square value. Steps,
The optical measurement method according to claim 2, further comprising: determining a refractive index of the sample based on a set of coefficients when the film thickness dispersion or the residual square value takes an extreme value. Calculating the optical characteristic comprises:
Calculating an actual value distribution indicated by a pixel value group in a position direction for an arbitrary wavelength of the actual value distribution;
Calculating a theoretical value distribution for the arbitrary wavelength based on a preset film thickness and refractive index of the sample and a correction factor corresponding to each position;
The optical measurement method according to claim 1, further comprising: determining a film thickness and a refractive index of the sample so as to reduce an error between the theoretical value distribution and the actual value distribution.   The optical measurement method according to claim 7, wherein the step of calculating the optical characteristic includes a step of determining a refractive index of the sample for each of a plurality of wavelengths of the measured value distribution. Calculating the optical characteristic comprises:
Calculating each film thickness of the sample for a plurality of wavelengths of the actual value distribution based on an error between the theoretical value distribution and the actual value distribution;
The optical measurement method according to claim 7, further comprising: determining a more probable film thickness based on the calculated film thicknesses. The refractive index of the sample used for calculation of the theoretical value distribution is calculated according to a predetermined wavelength dispersion formula,
Fitting each coefficient and film thickness defining the predetermined wavelength dispersion formula so as to reduce the error between the theoretical value distribution and the actual value distribution for a plurality of wavelengths of the actual value distribution; The optical measurement method according to claim 7, further comprising: An irradiation optical system for linearly irradiating measurement light having a predetermined wavelength range with respect to the measurement object;
Measurement optics that outputs a two-dimensional image by expanding the wavelength of linear measurement interference light, which is transmitted light or reflected light generated from the measurement object by irradiation of the measurement light, in a direction perpendicular to the longitudinal direction of the measurement interference light. The system,
A processing device, the processing device,
An acquisition means for acquiring an actual value distribution when the incident angle is varied for the same sample;
A correction element corresponding to an incident angle from each measurement point to the measurement optical system is calculated in association with a region on the two-dimensional image corresponding to each measurement point of the measurement target irradiated with the measurement light. Means for calculating
Second calculation means for calculating an optical characteristic including a refractive index of the sample based on a pixel value group for one or a plurality of columns along any one direction of the measured value distribution and a corresponding correction element. An optical measuring device comprising: