JPH08338709A - Device and method for measuring thickness of thin film and manufacture of optical filter - Google Patents

Abstract

PURPOSE: To measure the thickness of a colored thin film by calculating order to interference corresponding to wave length for giving a maximum value or a minimum value of spectral intensity of interference light of white color light due to the thin film and correcting the order of interference on the basis of spectral transmission characteristics of thin film material. CONSTITUTION: Measuring white-color light from a light source 11 is collected by a reflection mirror 12 and a lens 13, projected and reflected on a color filter coated film for a color display to be measured as parallel light by a lens for projecting and receiving 23 and becomes interference light. The interference light is collected, made incident on a spectral part 3, collected by a light collecting lens 31, and passed through a pin hole 32 of a position separated by only a focus point, and thereafter is made parallel light by a lens 33. The parallel light is separated by a plane diffraction grating 34. A component in the range of specified wave length out of spectral interference light is image-formed on an image sensor 36 by an image formation lens 35. The spectral intensity is successively read by a sensor drive circuit 37 for every picture element, transmitted to an operation processing part 4 as a voltage signal and thin film measuring can be performed by operation processing.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a thin film thickness measuring method and measuring apparatus based on the principle of optical interference, and a method for manufacturing an optical filter. More specifically, the present invention relates to a color filter for a color display device. The present invention relates to a method for measuring a film thickness of a measurement target with high accuracy, in which spectral intensity of optical interference in a thin film is likely to be disturbed due to being colored, a measuring device, and a method for manufacturing an optical filter using the measuring method.

[0002]

2. Description of the Related Art Conventionally, as a method for measuring the thickness of a thin film, a method utilizing an optical interference phenomenon has been known (Japanese Patent Laid-Open No. 56-115905, etc.). In this method, white parallel light is made incident on the thin film at an incident angle θ, reflected light or transmitted light from the thin film is received, and the received interference light is spectrally obtained (spectral intensity obtained as a function of wavelength). It is based on the principle that the film thickness is calculated by obtaining the wavelength that gives the adjacent maximum value or minimum value.

This principle will be described below using mathematical expressions.
When white collimated light is incident on the thin film at an incident angle θ, (1) the component of the light that is immediately reflected on the thin film surface, and (2) the light component that is transmitted on the thin film surface, enters the inside, and then is reflected on the back surface of the thin film. The component of the light emitted further, (3) the component of the light transmitted from the thin film surface and incident on the inside and then emitted from the back surface of the thin film, and (4) the component of the light emitted from the thin film surface after multiple reflection inside the thin film. And (5) components of light emitted from the back surface of the thin film after multiple reflection inside the thin film.

The spectral intensity distribution of the light emitted to the light source side with respect to such a thin film surface is (1), (2),
It is obtained as a result of the interference due to the superposition of the light waves of the respective components of (4). However, the component (4) among them is smaller than the components (1) and (2) and can be ignored. Hereinafter, the case where the light emitted to the light source side with respect to the thin film surface is measured (so-called reflection type) will be mainly described.

In the reflection type measurement under the above conditions, in principle, the intensity (modulation) occurs in the spectral intensity distribution of the measurement light as shown in FIG. The intensity of this spectral intensity distribution is such that when the optical optical path difference ΔL generated between the light reflected on the front surface and the light reflected on the back surface coincides with an integral multiple of the wavelength, the phases of the two become opposite phases. When the wavelengths are (integer + 1/2) times the wavelengths, the phases of the both coincide with each other and reinforce each other. (This is because the phase of the light reflected on the back surface is inverted, and since the phase is not inverted in the case of transmitted light, this relationship is reversed.)

Such an optical path difference ΔL is expressed by the equation (2). ΔL = 2d√ (n ² −sin ² θ) (2) where d is the thickness of the thin film and n is the refractive index of the thin film.

The two adjacent maximum wavelengths of the spectral intensities (wavelengths giving the maximum values of the spectral intensity) or the two adjacent minimum wavelengths (wavelengths giving the minimum value of the spectral intensity) thus obtained are expressed by λ Assuming ₁ , λ ₂ (λ ₁ > λ ₂ ), the following equation (3) is established. ΔL = K · λ ₁ = (K + 1) · λ ₂ (3) where K is an integer (when λ ₁ and λ ₂ are minimum wavelengths) or an integer +1/2 (when λ ₁ and λ ₂ are maximum wavelengths) ) (Transmissive case interchanges the conditions of maximum wavelength and minimum wavelength).

The following equation (4) is obtained by rearranging the equations (2) and (3). d = λ ₁ · λ ₂ / {(λ ₁ −λ ₂ ) · [2√ (n ² −sin ² θ)]} (4) That is, two maximum wavelengths (or minimum wavelengths) adjacent to each other in spectral intensity If λ ₁ and λ ₂ are obtained, the film thickness can be calculated based on the equation (4).

However, the interference spectral intensity actually obtained by measuring the thin film is different from the ideal spectral intensity as shown in FIG. 4, and only the distorted spectral intensity as shown in FIG. 5 is obtained.
This is because the light source, the spectroscope, the spectral intensity detector, and the like each have unique spectral characteristics, and the spectral characteristics as a total characteristic thereof are measured by being superimposed on the modulation of the spectral intensity due to the interference in the thin film.

As a method of solving this, the thin film is measured by previously measuring the spectral characteristics of the entire measurement system including the light source, the spectroscope, and the spectral intensity detector in a state where the thin film is removed and the influence thereof is eliminated. It is known that only the intensity of the spectral intensity due to the interference in the thin film is extracted as a modulation signal by correcting the spectral intensity obtained by the above, and the film thickness is measured from the wavelength that gives the extreme value of the obtained modulation signal. (Japanese Patent Publication No. 6-3364).

That is, the spectral intensity of the interference light due to the thin film to be measured is F (λ), the spectral intensity of the measurement light when a reflecting plate or the like is placed instead of the thin film is B (λ), and nothing is placed. W (λ) is the spectral intensity of the measurement light of A, and modulation signal A obtained by A (λ) = [F (λ) −W (λ)] / [B (λ) −W (λ)] (5) The film thickness is calculated from the wavelength position that gives the extreme value of (λ), and high-precision measurement is possible for ordinary transparent thin films.

However, when the film thickness of a colored thin film such as a color filter for a color display device is measured by the conventional optical interference type film thickness measuring method as described above, the color filter itself absorbs light. Accurate measurement was impossible. That is, the spectral transmission characteristic of the color filter is superimposed on the modulation due to the interference in the thin film, and even if the above correction is performed, the modulation signal A (λ) is generated as shown in FIG.
Distortion remains at the position, making it difficult to accurately determine the position of the maximum wavelength or the minimum wavelength.

As a method for solving this problem, the present inventors (1) previously measure the average spectral intensity of interference light by a plurality of thin films of the same type as the thin film to be measured and different in film thickness, and Only the intensity of the spectral intensity due to interference in the thin film is extracted as a modulation signal by obtaining the spectral intensity ratio of the spectral intensity obtained by measurement to the average spectral intensity,
For measuring the film thickness from the wavelength that gives the extreme value of the obtained modulation signal, or (2) measuring system consisting of a light source, a spectroscope, and a spectral intensity detector with the thin film to be measured removed to eliminate its influence. Blank spectral intensity, and the average spectral intensity of the interference light by a plurality of thin films of the same type as the thin film to be measured and different in film thickness is previously measured, and the spectral intensity obtained by measuring the thin film and the blank spectral intensity Spectral intensity ratio of, and by extracting the spectral intensity difference between the average spectral intensity and the spectral intensity ratio of the blank spectral intensity by extracting only the intensity of the spectral intensity due to interference in the thin film as a modulation signal, the obtained modulation For measuring the film thickness from the wavelength that gives the extreme value of the signal, or (3) the irradiation optical system of the measurement system consisting of the light source, the spectroscope, and the spectral intensity detector with the thin film to be measured removed to eliminate its influence. Is obtained by placing a calibration thin film of the same type as the thin film to be measured in the optical path of the light receiving optical system at a position where the white light is diverging or converging, and measuring the reference spectral intensity in advance. By extracting the spectral intensity ratio of the spectral intensity to the reference spectral intensity, only the intensity of the spectral intensity due to the interference in the thin film is extracted as a modulation signal,
It proposes that the film thickness is measured from the wavelength that gives the extreme value of the obtained modulation signal (Japanese Patent Application No. 6-72238).

That is, first, a plurality of samples of the same type as the thin film to be measured and having different film thickness around a predetermined value are prepared. These samples are sequentially placed at the measurement position and the spectral intensity is measured. Next, the arithmetic mean of the measured spectral intensities is obtained, and this is defined as the average spectral intensity G (λ). Next, the sample to be measured is measured to obtain the measured spectral intensity F (λ). Next, normalization is performed. That is, the modulated signal X (λ) is obtained from the equation (6) using the measured spectral intensity F (λ) and the average spectral intensity G (λ). X (λ) = F (λ) / G (λ) (6) This modulation signal X (λ) has only the pure modulation component due to the interference in the thin film removed from the influence of the spectral transmission characteristic of the measurement object itself. Will remain.

In another method, first, the measurement is performed in a state where the influence of the thin film to be measured is excluded, and the blank spectral intensity B
Get (λ). Next, a plurality of samples having the same type as the thin film to be measured and having different film thicknesses around a predetermined value are prepared. These samples are sequentially placed at the measurement position and the spectral intensity is measured. Next, the arithmetic mean of the measured spectral intensities is obtained, and this is defined as the average spectral intensity G (λ). Next, the sample to be measured is measured, and the measured spectral intensity F (λ)
And Next, normalization is performed. That is, the measured spectral intensity F (λ) and the blank spectral intensity B (λ) are used to obtain the normalized measurement signal A (λ) by the following equation (7). A (λ) = F (λ) / B (λ) (7) This normalized measurement signal A (λ) is a signal in which the spectral transmission characteristics of the measurement object itself and a pure modulation component due to interference in the thin film are superimposed. That is, the influence of the spectral characteristics of the entire optical system is eliminated.

Similarly, for the average spectral intensity G (λ),
Normalization processing is performed by the following equation (8) to obtain a normalized average signal C (λ). C (λ) = G (λ) / B (λ) (8) This normalized average signal C (λ) shows the spectral transmission characteristics of the measurement target itself, which is not affected by the interference of the thin film, as the spectrum of the entire optical system. It has been corrected according to the characteristics. continue,
Base correction is performed by the following equation (9) to obtain a modulation signal Y (λ) which is a difference in spectral intensity between the normalized measurement signal A (λ) and the normalized average signal C (λ). Y (λ) = A (λ) −C (λ) (9) In this modulation signal Y (λ), only a pure interference component due to the thin film superimposed on the spectral transmission characteristic of the measurement object itself remains. It becomes a thing.

Another method is to remove the thin film to be measured and eliminate the influence of the thin film in the optical path of the irradiation optical system or the light receiving optical system of the measurement system including the light source, the spectroscope, and the spectral intensity detector. A calibration thin film of the same type as the thin film to be measured is placed at a position where white light is diverging or converging, and the spectral intensity is measured and set as the reference spectral intensity H (λ). Next, the sample to be measured is measured to obtain the measured spectral intensity F (λ). Next, normalization is performed. That is, the modulated signal Z (λ) is obtained from the equation (10) using the measured spectral intensity F (λ) and the reference spectral intensity H (λ). Z (λ) = F (λ) / H (λ) (10) This modulation signal Z (λ) has only the pure modulation component due to the interference in the thin film removed from the influence of the spectral transmission characteristic of the measurement object itself. Will remain.

When the film thickness of a colored thin film such as a color filter for a color display device is measured by these optical interference type film thickness measuring methods, the distortion due to the spectral transmission characteristic of the color filter itself is corrected, and FIG. The modulated signal X (λ) as shown was obtained, and the position of the maximum wavelength or the minimum wavelength could be accurately obtained.

However, the film thickness obtained by substituting the position of the maximum wavelength or the minimum wavelength obtained here into the formula (4) is used for other absolute thickness measurement methods (for example, a part of the thin film is sharpened by a sharp jig). Compared with the absolute thickness obtained by scratching with a stylus displacement meter to measure the step, the accuracy was improved compared to the conventional method, but there was still a few percent error. There was a problem in practical use.

There is also known a method of improving measurement accuracy which is different from the above-mentioned method. This is because the film thickness obtained by the equation (4) is used as a provisional film thickness d ₀ , this d ₀ is substituted into the equation (2) to obtain the optical optical path difference ΔL, and the optical optical path difference ΔL is obtained by the equation ( Substituting 3) into the modified equation (11), the provisional order K corresponding to the maximum wavelength or the minimum wavelength λ ₁ of the spectral intensity is obtained, and this K is the nearest integer (when λ ₁ is the minimum wavelength) or an integer. After correcting to +1/2 (when λ ₁ is the maximum wavelength) K ₀ , λ ₁ and K ₀ are substituted into equation (12) to obtain the film thickness (Japanese Patent Laid-Open No. 61-76905). . K = ΔL / λ ₁ (11) d = K ₀ λ ₁ / [2√ (n ² −sin ² θ)] (12)

Another method for improving measurement accuracy is also known. This is a plurality of maximum wavelengths of spectral intensity (wavelengths that give a maximum value of spectral intensity) or a plurality of minimum wavelengths (wavelengths that give a minimum value of spectral intensity) λ ₁ , λ ₂ , λ ₃ ...
After obtaining (λ ₁ > λ ₂ > λ ₃ ...), m is an arbitrary positive integer (when λ ₁ , λ ₂ , λ ₃ ... Is a minimum wavelength) or a positive integer +1/2 Using (13) as (when λ ₁ , λ ₂ , λ ₃ ... Is the maximum wavelength), the film thickness d ₁ , d ₂ , d ₃ ... Corresponding to each maximum wavelength or the minimum wavelength is obtained. . d ₁ = m · λ ₁ / [2√ (n ² −sin ² θ)] d ₂ = (m + 1) · λ ₂ / [2√ (n ² −sin ² θ)] d ₃ = (m + 2) · λ ₃ / [2√ (n ² −sin ² θ)] (13) Next, the maximum value and the minimum value of the film thickness d ₁ , d ₂ , d ₃ ... Corresponding to each maximum wavelength or minimum wavelength obtained here. The difference between the values is the bias value δ _m . m is varied in the vicinity of the expected film thickness seeking the deviation value [delta] _m, and requests the thickness by [delta] a m giving the minimum value of _m as a true order (13) (JP-61- No. 76904).

These two methods are effective when the accuracy of position measurement at the maximum wavelength or the minimum wavelength of the spectral intensity distribution based on the interference phenomenon of light has an error. However, when these methods are applied to the film thickness measurement in a thin film that is colored like a color filter for a color display device, the obtained film thickness is different from other absolute thickness measurement methods (for example, a part of the thin film is sharpened). There is also a case where it is possible to make a step by scraping it off with a different jig and measuring the step with a stylus displacement meter.
It could not be put to practical use because it may have an error of 0%.

As a result of extensive studies by the present inventors, the last remaining error when the thickness of a thin film colored like a color filter for a color display device is measured by an optical interference type film thickness measuring method is as follows. It was clarified that it is not an error caused by the accuracy of position measurement of the maximum wavelength or the minimum wavelength of the spectral intensity distribution based on the interference phenomenon of light.

That is, the light interference phenomenon in a colored thin film such as a color filter for a color display device has a spectral intensity distribution in the wavelength direction longer than the light interference phenomenon in a transparent thin film having the same thickness. It was found to be compressed with a dependency, and to appear as if it had become thicker. It was also found that this phenomenon is called the bathochromic effect or redshift.

Therefore, when the thickness of a colored thin film such as a color filter for a color display device is to be measured by the optical interference type film thickness measuring method, how accurately the maximum wavelength or the minimum wavelength of the spectral intensity distribution is minimized. Even if the position of the wavelength is set, it is difficult to accurately measure the film thickness, and in the manufacturing process of the color filter for the color display device, the film thickness is measured by the conventional optical interference film thickness measuring method described above. It was difficult to control the process.

Therefore, conventionally, in the manufacturing process of the color filter, a part of the color filter is scraped off with a sharp jig to form a step, and the step is measured by a stylus displacement meter to obtain the film thickness. However, this method has drawbacks such as destructive inspection and time-consuming measurement. For this reason, the adjustment operation may be delayed and defective products may occur.
It was also necessary to perform extra work such as putting in and taking out things that could not be shipped due to their quality. In addition, there is a problem that the final quality cannot be known unless the product is inspected because the history of the film thickness inspection is different from that of the film actually in process.

[0027]

Therefore, the first aspect of the present invention
The purpose of is, even if the thin film to be measured is colored,
An object of the present invention is to provide a method and an apparatus capable of measuring the thickness of a thin film with high accuracy.

A second object of the present invention is to control the process in a manufacturing process of an optical filter such as a color filter for a color display device by measuring the film thickness of the filter with high accuracy in a non-contact manner, thereby improving the yield. An object of the present invention is to provide a method for manufacturing an optical filter capable of improving the above.

[0029]

Means for Solving the Problems A thin film thickness measuring method of the present invention which achieves the above object is to irradiate a thin film with white light, and measure the spectral intensity of interference light of the white light by the thin film.
A temporary interference order corresponding to a wavelength that gives a maximum value or a minimum value of the spectral intensity is calculated based on the measured spectral intensity, and the calculation is performed by a correction formula separately determined based on the spectral transmission characteristic of the material of the thin film. The corrected interference order is corrected, and the thickness of the thin film is calculated based on the corrected interference order.

Further, the thin film thickness measuring method according to the present invention irradiates the thin film with white light, measures the spectral intensity of the interference light of the white light by the thin film, and determines the spectral transmission characteristic of the material of the thin film. Correcting the measured spectral intensity, calculating a temporary interference order corresponding to a wavelength that gives a maximum value or a minimum value of the spectral intensity based on the corrected spectral intensity,
A method of correcting the calculated interference order by a correction formula separately determined based on the spectral transmission characteristic of the material of the thin film, and calculating the thickness of the thin film based on the corrected interference order. Consists of.

In the latter method, as a method for correcting the measured spectral intensity by the spectral transmission characteristic of the material of the thin film, (1) interference light by a plurality of thin films of the same kind as the thin film but different in thickness (2) The average spectral intensity of the interference light due to a plurality of thin films of the same type as the thin film and different in thickness and the measured spectral intensity. (3) The spectral intensity ratio of the measured spectral intensity to the spectral intensity of the white light in the state in which the influence of the thin film is eliminated, (4) The state in which the influence of the thin film is eliminated In the optical path of the irradiation optical system for irradiating the thin film with the white light or the light receiving optical system for measuring the spectral intensity of the interference light of the white light by the thin film, the white light is diverged. Is used to determine the spectral intensity ratio of the measured spectral intensity of the interference light with respect to the reference spectral intensity measured by placing a calibration thin film of the same type as the thin film at the converging position. it can.

In such a method, the thickness of the thin film to be measured is not particularly limited, but is 0.4 to 4.0 μm.
It is particularly suitable to apply the method of the present invention to a thin film in the range of m.

The thin film thickness measuring apparatus according to the present invention further comprises an irradiation optical system for irradiating the thin film with white light, a light receiving optical system for measuring the spectral intensity of the interference light of the white light by the thin film, Interference order calculating means for calculating a temporary interference order corresponding to a wavelength that gives a maximum value or a minimum value of the spectral intensity based on the measured spectral intensity, and a correction separately determined based on the spectral transmission characteristic of the material of the thin film. It is characterized by comprising an interference order correction means for correcting the determined interference order by a formula, and a film thickness calculation means for calculating the film thickness of the thin film based on the corrected interference order.

Further, the thin film thickness measuring apparatus according to the present invention comprises an irradiation optical system for irradiating the thin film with white light, a light receiving optical system for measuring the spectral intensity of the interference light of the white light by the thin film, Measurement spectral intensity correction means for correcting the measured spectral intensity according to the spectral transmission characteristics of the material of the thin film, and temporary interference corresponding to a wavelength that gives a maximum value or a minimum value of the spectral intensity based on the corrected spectral intensity. An interference order calculating unit for calculating an order, an interference order correcting unit for correcting the calculated interference order by a correction formula separately determined based on a spectral transmission characteristic of the material of the thin film, and an interference order based on the corrected interference order. And a film thickness calculating means for calculating the film thickness of the thin film.

In a preferred embodiment of the thin film thickness measuring method and apparatus of the present invention, the thin film is either an optical filter or an optical filter coating film.

Further, in the method for producing an optical filter of the present invention, the thickness of the coating film of the optical filter is measured using the above-mentioned thin film measuring method, and the measured thickness is within a predetermined range. It is characterized in that the means for forming the coating film is controlled so as to enter.

In a preferred embodiment of the method for producing an optical filter, the thickness of the coating film of the optical filter before curing is measured, and when the measured thickness does not fall within a predetermined range, the optical In this method, the coating film of the filter is peeled off and the transparent substrate used for the optical filter is regenerated.

Further, in the preferred embodiment of the method for producing an optical filter of the present invention, the coating film forming means is any one of a slit die, a spin coater, a roll coater, a bar coater and a dipping / pulling device.

[0039]

According to the thin film thickness measuring method and the measuring device of the present invention, not only the influence of the distortion of the spectral intensity of the interference light due to the thin film to be measured but also the influence of the bathochromic effect of the thin film material to be measured is considered. Can also be corrected to measure the film thickness. Then, according to the method for producing an optical filter of the present invention, it is possible to early detect a defect in the process by correcting the influence of coloring of the optical filter coating film and accurately measuring the film thickness and controlling the process. .

The method and apparatus for measuring the thickness of a thin film according to the present invention will be described below with reference to the drawings by taking the case of measuring the thickness of a color filter for a color display device, which is a type of optical filter, as an example.

FIG. 1 shows a schematic configuration of an embodiment of an apparatus used in the thin film thickness measuring method of the present invention. The example of the present embodiment includes a measurement unit 100 having a light source unit 1, a light projecting / receiving unit 2, and a spectroscopic unit 3, and an arithmetic processing unit 4. The light projecting section and the light source section of the light projecting and receiving section 2 constitute an irradiation optical system, and the light receiving section of the light projecting and receiving section and the spectroscopic section 3 constitute a light receiving optical system.

The light source section 1 is composed of a light source 11, a reflecting mirror 12 and a lens 13. The light projecting / receiving unit 2 includes a light projecting optical fiber 21, a light receiving optical fiber 22, and a light projecting / receiving lens 23. Optical fiber for projection 2
1 and the light receiving optical fiber 22 are bundles of several hundred optical fibers, respectively, and are bundled near the light emitting and receiving lens 23. In this bundle, the light projecting optical fiber 21 and the light receiving optical fiber 22 are intermingled and bundled, and both parts of the light projecting and receiving surface facing the lens 23 for projecting and receiving light of this bundle are evenly distributed in both parts. An optical fiber is arranged.

The spectroscopic unit 3 includes a condenser lens 31 and a pinhole 3.
2 (thin plate having a pinhole), a lens 33, a plane diffraction grating 34, an imaging lens 35, an image sensor 36, an image sensor drive circuit 37, and a buffer amplifier 38.

The arithmetic processing unit 4 is composed of an A / D (analog / digital) converter 41, a microcomputer 42, a storage device 43, an input / output device (not shown), a power supply device and the like.

The operation of the above apparatus will be described below. The white measuring light emitted from the light source 11 is reflected by the reflecting mirror 12,
And the optical fiber 21 for projecting the light by the lens 13
Is incident on. The white measurement light that has passed through the light projecting optical fiber 21 is projected by the light projecting and receiving lens 23 as parallel light or weak convergent light onto the color filter coating film 5 for the color display device to be measured.

This white measuring light is applied to the color filter coating film 5
The reflected light becomes interference light, is condensed by the light projecting / receiving lens 23, and is guided to the spectroscopic unit 3 through the light receiving optical fiber 22.

The interference light incident on the spectroscopic unit 3 is collected by the condenser lens 31, and is placed at a pinhole 32 separated from the condenser lens 31 by the focal length of the lens, and the lens 33 behind the pinhole 32. The light is collimated by the lens 33 placed at a focal distance of.

The collimated light emitted from the lens 33 is incident on the plane diffraction grating 34 and is dispersed, and the components of a predetermined wavelength range of the dispersed interference light are imaged on the image sensor 36 by the imaging lens 35. It Image sensor 36
The spectral intensity of the interference light imaged above is sequentially read out for each pixel by the image sensor drive circuit 37, and is sent to the arithmetic processing unit 4 as a voltage signal via the buffer amplifier 38.

In the arithmetic processing section 4, the voltage signal sent from the buffer amplifier 38 is converted into a digital signal by the A / D converter 41, and then read into the microcomputer 42 for arithmetic processing.

Hereinafter, the arithmetic processing unit 4 which is the main part of the present invention.
The measurement procedure will be described. FIG. 2 is a flowchart showing a calculation process and a measurement procedure in one embodiment of the thin film thickness measuring method of the present invention. The measurement procedure is roughly divided into the following: (1) Measurement of blank spectral intensity (2) Measurement of spectral intensity of a plurality of samples having different film thicknesses,
And calculation of order correction coefficient based on it (3) measurement of sample (4) data processing. The process (3) may be performed before the processes (1) and (2).

The measurement procedure will be described below. First,
The blank spectral intensity B (λ) is measured. Here, the blank spectral intensity refers to the spectral intensity obtained in a case where the measurement is performed while the influence of the thin film to be measured is excluded. By measuring the blank spectral intensity, it is possible to obtain the spectral intensity characteristics of the irradiation optical system and the light receiving optical system that do not depend on the presence of the thin film. Further, in the case where the thin film is formed on a transparent or semi-transparent substrate as in the case of a color filter, it is preferable to measure the spectral intensity characteristic of the entire optical system including this substrate. By taking the spectral intensity ratio of the blank spectral intensity measured in this way and the spectral intensity when the sample is measured, only the modulated component of the spectral intensity due to the presence of the thin film can be extracted.

In this embodiment, the measurement object is the color filter coating film 5. Therefore, in this embodiment, when forming the color filter coating film 5, the spectral intensity characteristic of the entire optical system including the characteristic of the glass substrate or the plastic substrate as the base material is measured as the blank spectral intensity.
Therefore, the substrate before forming the color filter coating film to be measured is placed at the measurement position instead of the substrate with the color filter coating film 5, and the blank spectral intensity B (λ) obtained in this state is measured.

When the object to be measured is not a thin film formed on the above-mentioned substrate, a reflecting mirror or a glass plate having a flat spectral reflectance is placed at the measuring position instead of the thin film to be measured. You may. Further, in the case of a transmission type, the measurement may be performed without a thin film, or the measurement may be performed by placing a glass plate or the like as in the above.

It is not necessary to measure the blank spectral intensity B (λ) every time. However, it is preferable to periodically perform it in order to compensate for the deterioration with time of the light source and the change with time of the spectral sensitivity characteristic of the light receiving unit image sensor.

The procedure (1) blank spectral intensity B
Instead of measuring (λ), the spectral intensities of a plurality of samples having different film thicknesses may be measured and their average spectral intensities may be calculated.

That is, first, a plurality of samples having the same type as the thin film to be measured and having different film thicknesses around a predetermined value are prepared. These samples are sequentially placed at the measurement position and the spectral intensity is measured. Next, the spectral addition average of the measured spectral intensities is obtained, and this is calculated as the average spectral intensity G
(Λ). The steps up to here correspond to the procedure (1) above. The sample used at this time is the average spectral intensity G (λ)
In order to prevent the modulation of the spectral intensity due to the interference in the thin film of each sample from remaining, 3 to 256 are selected from the plurality of prepared samples that have slightly different wavelengths that give the extreme values of the spectral intensity due to the interference.

Average spectral intensity G measured and calculated in this way
(Λ) is only the spectral intensity characteristics of the light source section, the light emitting / receiving section and the spectroscopic section of the optical system used for measurement, and the effect of light absorption in the thin film of the color filter to be measured (excluding the effect of interference) It can be said that the average spectral intensity includes the spectral intensity characteristic of In other words, the average spectral intensity G (λ)
Represents the spectral characteristic obtained by averaging all the spectral intensity characteristics of the measurement optical system including the color filter, except for the modulation due to the interference at the thin film of the color filter.

It is not necessary to measure this average spectral intensity G (λ) every time. Target thickness, composition of color filter,
It may be performed only when the manufacturing conditions are changed. Also,
When the average spectral intensity G (λ) is measured instead of the blank spectral intensity B (λ) in the procedure (1), the blank spectral intensity B (λ) is calculated in the following procedure. It should be read as (λ).

Next, the order correction coefficient is calculated. Not only the color filter, but the thickness of a thin film which is an ordinary industrial product varies around a predetermined value, and it is rare to measure the thickness of a thin film whose thickness is unknown. Therefore, 3 to 256 samples of the same type as the thin film to be measured and having different film thicknesses around a predetermined value are prepared. One of these samples is placed at the measurement position and the spectral intensity is measured and designated as F (λ).

Next, normalization is performed. That is, the measured spectral intensity F (λ) and the blank spectral intensity B (λ) are used to obtain the normalized measurement signal A (λ) by the following equation (7). A (λ) = F (λ) / B (λ) (7) This normalized measurement signal A (λ) is a signal in which the spectral transmission characteristics of the measurement object itself and a pure modulation component due to interference in the thin film are superimposed. That is, the influence of the spectral characteristics of the entire optical system is eliminated.

Then, the obtained normalized measurement signal A (λ)
Of the maximum value of the normalized measurement signal A (λ) or a plurality of minimum wavelengths (wavelengths of the minimum value of the normalized measurement signal A (λ)) are detected. ) Λ ₁ , λ ₂ , λ ₃ ... (λ ₁ > λ ₂ > λ ₃ ...
・)

Normalized measurement signal A obtained as described above
Among a plurality of wavelengths (maximum wavelength and minimum wavelength) that give the extreme value of (λ), from the adjacent maximum wavelength or minimum wavelength,
The provisional film thickness d _i is calculated based on the equation (14). d _i = λ _i · λ _{i + 1} / {(λ _i −λ _{i + 1} ) · [2√ (n ² −sin ² θ)]} (14)

Then, the temporary film thickness d _i is substituted into the equation (15) to obtain the temporary order K _i corresponding to the maximum wavelength or the minimum wavelength λ _i of the normalized measurement signal A (λ). K _i = [2d _i √ (n ² −sin ² θ)] / λ _i (15)

Alternatively, instead of calculating the provisional film thickness d _i , the provisional order K _i corresponding to the wavelength λ _i may be directly obtained from the adjacent maximum wavelength or minimum wavelength by using the equation (16). K _i = λ _{i + 1} / (λ _i −λ _{i + 1} ) (16)

Furthermore, by using the formula (17), from three adjacent maximum wavelengths or minimum wavelengths, or three adjacent (maximum, minimum, maximum) wavelengths or (minimum, maximum, minimum) wavelengths. The temporary order K _i corresponding to the wavelength λ _i may be directly obtained by the combination. K _i = λ _i-1 · λ _{i + 1} / [(λ _i-1 −λ _{i + 1} ) · λ _i ] (17)

Next, the film thickness of the sample used for the above measurement is measured by another absolute thickness measuring method (for example, a part of the thin film is scraped off by a sharp jig to make a step, and the step is made by a stylus displacement meter. Is measured) to obtain the true film thickness D.

Substituting this true film thickness D into the equation (18), the maximum wavelength or the minimum wavelength λ _{i of the} normalized measurement signal A (λ)
The true order M _i corresponding to M _i = [2D√ (n ² −sin ² θ)] / λ _i (18)

This calculation is performed for a plurality of maximum wavelengths or minimum wavelengths of the normalized measurement signal A (λ) obtained by one sample, and the wavelength λ _i , the temporary order K _i, and the true order M _{i are obtained.}
A plurality of data sets (λ _i , K _i , M _i ) are obtained.

Further, the above calculation is performed on a plurality of samples 3 to 256, which are thin films of the same kind as the thin film to be measured and have different film thicknesses around a predetermined value, and the wavelength λ _i
And a provisional order K _i and a true order M _i of the data set (λ _i ,
Further, a plurality of K _i , M _i ) are obtained.

Next, the plurality of data sets (λ _i , K _i , M _i ) obtained here are arranged, and the coefficients a and b satisfying the equation (19) are obtained by the least square method. M = K + a + b · λ (19) Up to here corresponds to the calculation procedure of the order correction coefficient in (2) above.

The calculation of the order correction coefficient does not have to be performed every time. It may be performed only when the composition of the color filter, the manufacturing conditions, etc. are changed.

Next, the sample to be measured is measured. The color filter coating film 5 is placed at the measurement position, and the image sensor drive circuit 37 is instructed by the microcomputer 42.
Reads out the output of the image sensor 36 sequentially for each pixel.
This output is sent to the arithmetic processing unit 4 as a voltage signal via the buffer amplifier 38, converted into a digital signal by the A / D converter 41, and then read to obtain the measured spectral intensity F (λ). The steps up to here correspond to the procedure of (3) above.

The procedure for measuring a plurality of samples having different film thicknesses when obtaining the blank spectral intensity B (λ) and when obtaining the order correction coefficient is the same.

Next, the data processing will be described. Figure 2
As shown in, the data processing is performed through the steps of (a) smoothing, (b) normalization, (c) extreme value detection, (d) provisional order determination, (e) correction order determination, and (f) film thickness calculation.

The function of each block will be described in detail below. First, smoothing is performed. Image sensor 3 for smoothing
This is done to remove noise superimposed on the output signal of No. 6 and to compensate the variation in the gain of the odd / even bits of the image sensor 36 itself. Such a method may be used.

However, the moving average process is preferable in terms of speeding up the process and maintaining the phase characteristic after the smoothing process. The moving average processing is performed for the purpose of removing noise from time-series data or a continuous group of data, and applies an average value of preceding and following data as the i-th (i is an integer) data. . There is a method of simply averaging the data before and after, and a method of averaging with appropriate weights.

Further, it is preferable that the smoothing process is also performed on the blank spectral intensity B (λ). In addition,
This smoothing process is not always necessary, but in the present embodiment, each spectral intensity is assumed to have undergone a smoothing process.

Next, normalization is performed. That is, the measured spectral intensity F (λ) and the blank spectral intensity B (λ) are used to obtain the normalized measurement signal A (λ) by the following equation (7). A (λ) = F (λ) / B (λ) (7) Specifically, the measured spectral intensities F (λ) held inside the microcomputer 42 as data corresponding to each pixel of the image sensor 36 and Blank spectral intensity B
Data corresponding to the same pixel of (λ) is sequentially read and division processing is performed.

At this time, the data corresponding to each pixel of the measured spectral intensity F (λ) and the blank spectral intensity B (λ) are integer values of almost the same size, and if the division process is performed as it is, a digit loss occurs. The required accuracy may not be obtained. F
If the division processing is performed after converting the data of (λ) and B (λ) into real numbers, the problem of digit cancellation disappears, but the processing takes time and is not practical.

Therefore, if the data of the measured spectral intensity F (λ) is multiplied by 1024, for example, and then integer division processing is performed, there is no problem of digit cancellation, and high-speed calculation is possible. It should be noted that the reason that the magnification is 1024 times instead of 1000 times is
This is because the multiplication of 2 ⁿ can be performed by the bit shift process and can be processed at higher speed.

In the normalized measurement signal A (λ) obtained by this normalization processing, the influence of the spectral transmittance of the material itself of the color filter coating film 5 is removed and only the pure modulation component due to the interference in the thin film is removed. It remains.

Then, the obtained normalized measurement signal A (λ)
The wavelength that gives the extreme value of is detected. Since the normalized measurement signal A (λ) is almost equal to the theoretical interference signal with a small noise component by the above processing, the normalized measurement signal A (λ) is differentiated to change the slope. It is possible to easily and accurately know the positions of the maximum value and the minimum value by a method of examining, a method of obtaining a partial barycentric position of the signal, or the like.

The position of the extreme value obtained here is the pixel number of the image sensor 36. To know the wavelength corresponding to the pixel number giving this extreme value, the pixel number of the image sensor 36 corresponds to the wavelength. You need to know the relationship.

This correspondence is based on the color filter coating film 5
Instead, place a reflector with a relatively flat spectral reflection characteristic at the measurement position, and place an interference filter with a known spectral characteristic at the position where white light propagates in the irradiation optical system or light receiving optical system in a state close to parallel light. It can be obtained by installing and measuring, detecting the peak position and the like, and taking correspondence with the pixel number.

Since the peak position and the like are not strictly located at the center of a specific pixel, it is preferable to obtain the true peak position by interpolation using the data value of each pixel near the peak.

Subsequently, the provisional order is calculated. Among the plurality of wavelengths (maximum wavelength and minimum wavelength) that give the extreme values of the normalized measurement signal A (λ) obtained as described above, the formula (14) and the formula (15) are calculated from the adjacent maximum wavelength or minimum wavelength. ) Or Expression (16) or Expression (17), the provisional order K _i corresponding to the maximum wavelength or the minimum wavelength λ _i of the spectral intensity is obtained.

Then, the true order is calculated. The tentative order K _i obtained by the above procedure is substituted into the equation (19) to obtain the corrected order M _i corresponding to the maximum wavelength or the minimum wavelength λ _i of the spectral intensity.

Subsequently, the film thickness is calculated. The true film thickness d is obtained by substituting the corrected order M _i obtained by the above procedure into the equation (20). d = M _i λ _i / [2√ (n ² −sin ² θ)] (20) At this time, for a plurality of wavelengths (maximum wavelength or minimum wavelength) that give the extreme value of the normalized measurement signal A (λ) The film thickness may be calculated and the average value may be obtained.

In the method described in detail above, when light is projected and received directly through the optical window without using an optical fiber for projecting and receiving light, the spectral intensity when nothing is placed at the measurement position is W (λ ), It is more accurate if W (λ) is subtracted in advance from the spectral intensities used for the calculation of the measured spectral intensity F (λ), the blank spectral intensity B (λ), the average spectral intensity G (λ), etc. It is preferable because measurement becomes possible.

Further, in the present invention, when the white light is applied to the measurement object, it is not necessary that the light is perfectly parallel light. When the thin film is irradiated with the convergent light or the divergent light, the incident angle is not constant and the influence of interference does not appear on the spectral intensity in the thin film as shown in FIG. However, if the convergence angle or the divergence angle is sufficiently small and the distribution range of the incident angle of the light irradiating the thin film is narrow, the modulation due to the interference of the intensity sufficient for the spectral intensity in the thin film appears. Specifically, the width of the distribution range of the value of ΔL in Expression (2) is 0.2 which is the shortest wavelength of the effective spectral range of the light receiving optical system.
When the width is narrower than about twice, the modulation by the interference with the intensity of about 90% of the case where the perfect parallel light is incident can be obtained. Therefore, it is also preferable to use a lens having a long focal length to converge or diverge white light when irradiating the measurement object with white light.

An optical filter is suitable as the thin film to which the present invention is applied. In particular, the method for measuring the film thickness of the present invention is particularly suitable for measuring the film thickness of a film in which light is absorbed. As the optical filter, various optical filters for converting the spectral sensitivity characteristic of an optical device such as a camera into a desired one, and a color filter for a color display device are preferably used. As the color display device, there are a liquid crystal display device, a plasma display device, an electroluminescence display device, and the like.

Hereinafter, the case where the thin film manufacturing method according to the present invention is applied to a color filter thin film for a color liquid crystal display device will be described in detail with reference to the drawings. Generally, a color filter used for a display using liquid crystal or the like has three primary color patterns of red, green and blue arranged on a transparent substrate such as a glass substrate while maintaining a certain fixed positional relationship. As a method of manufacturing such a color filter, for example, the following method is known.

First, a light shielding layer having a fine pattern is formed on a transparent substrate. A thermosetting paint containing a colorant is applied onto the light-shielding layer and dried at a temperature of 50 to 150 ° C. (semi-cure) to form a thermosetting coating film. Subsequently, a positive photoresist is applied on the thermosetting coating film, mask exposure is performed using an ultra-high pressure mercury lamp or the like, then the positive photoresist is developed and etched to form a relief pattern, The thermosetting coating film is etched using the relief pattern as a mask.

Then, after removing the positive photoresist, it is heated and baked at about 200 to 350 ° C. (cure),
The thermosetting coating film is heat cured to form a color filter coating film. Repeat the above steps for red, green, and blue,
A predetermined fine pattern of three colors of red, green and blue is formed and used as a color filter.

Further, instead of the thermosetting paint containing the colorant, a photosensitive paint containing the colorant may be used.
In this case, a light-shielding layer having a fine pattern is formed on a transparent substrate, a photosensitive coating material containing a colorant is applied on the light-shielding layer, and the coating is dried at a temperature of 50 to 150 ° C. (semi-cure). Form a photosensitive coating. Next, mask exposure is performed using an ultra-high pressure mercury lamp or the like to photo-cur the photosensitive coating film.
After that, the uncured photosensitive coating film on the light shielding layer is developed, etched and peeled off. Then, it is heated and baked at about 200 to 350 ° C. (cure), and the photosensitive coating film is thermally cured to form a color filter coating film. The above steps are repeated for three colors of red, green and blue to form a predetermined fine pattern of three colors of red, green and blue to form a color filter.

As a transparent substrate, a plastic plate such as polycarbonate or polymethylmethacrylate, a plastic film or a glass plate having a total light transmittance of 70% is used.
The transparent or semitransparent flat plate described above is preferably used. In particular, a glass plate is preferable because it has high light transmittance.

As the type of paint, one or a mixture of two or more kinds of acrylic resin, polyester resin, polyvinyl alcohol resin, polyamide resin, polyimide resin and the like, and a coloring agent such as a dye or a pigment are added to these photosensitive resins and the like. Those dispersed and mixed are preferably used.

Various methods such as a spin coater method, a dip pull-up coating method, a roll coater method, a bar coater method, and a slit die method can be used as the color filter thin film coating method on the transparent substrate. FIG. 3 relates to the present invention. It is a schematic diagram showing an example of application of a slit die system of a color filter thin film.

The paint 70 is supplied from the pipe 68 to the inside of the die 66 by the pump 69, and as the transparent substrate 61 is conveyed (in the direction of arrow A), it is adsorbed on the adsorption stage 62 from the lip 65 at the tip of the die 66 in a curtain state. It is applied on the fixed transparent substrate 61. In addition, 63 is the applied paint and 67 is the slit gap adjusting bolt of the lip 65.

The thickness of the color filter thin film varies due to various causes in the manufacturing process of the color filter. When the final thickness of this color filter thin film is incorporated into a color liquid crystal display device, it appears as color unevenness in the color display, which impairs the commercial value. Therefore, in the manufacturing process of the color filter, it is necessary to constantly monitor the thickness unevenness of the color filter thin film and keep it within a certain range.

Further, the color filter thin film after curing is fixed to the transparent substrate, and if abnormal thickness unevenness is found at this stage, it must be discarded, which is a major cause of a decrease in yield. However, since the color filter thin film immediately after coating and after semi-cure is not fixed to the transparent substrate, if the thickness irregularity can be found at this stage, the coating film can be peeled off and the transparent substrate can be reused. .

Therefore, it is preferable to measure the thickness of the color filter thin film immediately after coating and after semi-curing. The thin film thickness measuring method according to the present invention described above is preferably applicable to the measurement of the thickness of such a color filter thin film, non-contact and high precision even in the portion of the fine pattern,
It can be measured at high speed. This can greatly contribute to the stabilization of the process, the improvement of the yield, and the cost reduction.

That is, immediately after coating or after semi-curing, the in-plane film thickness distribution of the color filter thin film was measured, and the result was used immediately for the lip 6 at the tip of the die.
5 to maintain the uniformity of the coating film by adjusting the gap between the transparent substrate 61 and the lip 65, that is, the clearance 64 to adjust the widthwise distribution of the paint discharged in a curtain shape. You can

Further, the in-plane average value of the coating film thickness may be kept constant by adjusting the discharge amount of the coating liquid 70 and / or the transport speed of the transparent substrate 61.

Further, the in-plane film thickness distribution of the semi-cured color filter is measured, and when the film thickness does not fall within a desired constant value, the process is adjusted and the coating film is peeled off to recycle the transparent substrate. Therefore, the yield can be improved and the cost can be greatly reduced.

[0106]

【Example】

Example 1 An example of applying the thin film thickness measuring method and measuring apparatus of the present invention to the film thickness measurement of a coating film of a color filter for a color display device will be described below. The structure of the device is that shown in FIG. A halogen lamp was used as a light source, and a quartz bundle fiber was used as the light projecting optical fiber 21 and the light receiving optical fiber 22. As the plane diffraction grating 34, one having a sufficient sensitivity to 600 to 1100 nm was used. Also, the image sensor 36 is 600 to 11
C composed of 1024 pixels with sufficient sensitivity to 00 nm
CD (Charge Coupled Device)
A device was used. As the A / D converter, a 12-bit full scale (4096 steps) was used. Further, smoothing was performed by using a simple moving average and a moving average of 8 points.

Using a slit die method shown in FIG. 3, an N-methylpyrrolidone solution, chlorobrominated phthalcyanine green (CI Pigment Green 36) and disazo yellow were placed on a glass substrate having a size of 300 mm × 400 mm and a thickness of 1.1 mm. A polyamic acid coating containing an HR (C.I. Pigment Yellow 83) pigment added and dispersed was applied at a solid content of 8% by weight and a viscosity of 0.15 Pa · s, and dried at 110 ° C for 10 minutes (semi-cure). At this time, adjust the discharge amount of paint,
A plurality of green surface-coated color filter coating film samples having different thicknesses were obtained.

First, the blank spectral intensity B (λ) was measured. That is, the glass substrate used as the base material when preparing the color filter coating film sample was placed at the measurement position, and the measurement was performed to obtain the blank spectral intensity B (λ).

Next, the correction coefficient was determined. That is, the spectral intensity F measured using the above-described measuring device for any 15 out of the plurality of green color whole-surface coating color filter coating film samples prepared above having different thicknesses.
₁ (λ), F ₂ (λ), ..., F ₁₅ (λ) were measured.

Further, the measured measured spectral intensity F
_The spectral intensity ratio of ₁ (λ), F ₂ (λ), ..., F ₁₅ (λ) to the blank spectral intensity B (λ) is calculated, and the normalized measurement signals A ₁ (λ) and A ₂ (λ) are obtained. , ..., A ₁₅ (λ).

Furthermore, the normalized measurement signals A ₁ (λ), A ₂
(Λ), ..., A ₁₅ (λ) which gives the extreme value of the wavelength is detected, and the provisional order K corresponding to the wavelength λ _i is detected by using the equation (16).
_i asked.

Next, a part of the 15 samples was scraped off with a sharp jig to form a step, and the step was measured by a stylus displacement meter to obtain the film thickness of each sample. Was substituted into the equation (18) to obtain the true order M _i corresponding to the maximum wavelength or the minimum wavelength λ _i of the normalized measurement signal A (λ).

Next, the 15 data sets (λ _i , K _i , M _i ) obtained here are used to calculate the difference M− between the true order and the tentative order.
When K is plotted on the ordinate and the wavelength λ giving the extreme value is plotted on the abscissa, the relationship of FIG. 8 was obtained.

When the coefficients a and b satisfying the equation (19) are obtained by the least square method, the equation (21) is obtained. M = K-4.9 + 0.0045 · λ (21)

Next, the sample was measured. Of the above-mentioned multiple green color coating color filter coating film samples,
Three samples different from those used to determine the correction factor were prepared. A part of these three samples was scraped with a sharp jig to create a step, and the step was measured by a stylus displacement meter. The film thicknesses were 1.43, 1.58, and 1.60 μm, respectively. It was

The spectral intensity F ₁ (λ), F ₂ (λ), and F ₃ (λ) of the interference light measured by the color filter coating films of the three samples were measured by using the above-described measuring device, and blanks were further obtained. Calculate the spectral intensity ratio to the spectral intensity B (λ),
Normalized measurement signals A ₁ (λ), A ₂ (λ) and A ₃ (λ) were obtained. The maximum wavelength and the minimum wavelength were obtained from this normalized measurement signal, and the provisional order K _i corresponding to the wavelength λ _i was obtained using equation (16).

Next, in equation (21), the wavelength λ _i and the temporary order K
by substituting the _i, to determine the true of order M with respect to the wavelength λ _i. Further, when the film thickness was calculated with n = 1.7 and θ = 0 in the equation (20), the obtained film thicknesses were 1.42, 1.
It was 57 and 1.60 μm, and accurate measurement could be performed. Table 1 shows the measurement and calculation results for these F ₁ (λ), F ₂ (λ), and F ₃ (λ).

[0118]

[Table 1]

Comparative Example 1 The same three samples as in Example 1 were measured with the apparatus having the same configuration as in Example 1, and the same blank spectral intensity B as in Example 1 was obtained.
The spectral intensity ratio with respect to (λ) is obtained, and the normalized measurement signal A ₁
(Λ), A ₂ (λ) and A ₃ (λ) were obtained. The maximum wavelength and the minimum wavelength were obtained from the normalized measurement signal, and the film thickness was calculated with θ = 0 and n = 1.7 in the equation (4). The obtained thickness was 1.57 and 1.71 respectively. 1.72
Since it was μm, it could not be said to be an accurate measurement.

The normalized measurement signals of the measurement results in Example 1 and Comparative Example 1 are shown in FIG.

[0121]

According to the thin film thickness measuring method and measuring apparatus of the present invention, since the thin film to be measured has a spectral transmission characteristic of light due to coloring, the spectral intensity of the interference light due to such thin film is Even when the maximum wavelength or the minimum wavelength changes due to distortion, the influence of such distortion can be eliminated and the film thickness can be accurately measured.

Further, according to the method for producing an optical filter of the present invention, the process control can be performed by correcting the influence of the coloring of the optical filter coating film and measuring the film thickness accurately and in a non-contact manner. it can. As a result, it is possible to detect a defect in a process at an early stage, improve the yield, stabilize the process, and improve the quality.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram of a thin film thickness measuring apparatus according to an embodiment of the present invention.

FIG. 2 is a flowchart showing an example of arithmetic processing according to the present invention.

FIG. 3 is a schematic configuration diagram showing an example of a method for manufacturing an optical filter according to an embodiment of the present invention.

FIG. 4 is a schematic diagram showing a spectrum of interference caused by a thin film.

FIG. 5 is a diagram showing a spectroscopic spectrum of interference when actually measuring a thin film.

FIG. 6 is a characteristic diagram showing a modulation signal as a measurement result according to a comparative example.

FIG. 7 is a characteristic diagram showing a modulation signal as a measurement result according to an example.

FIG. 8 is a relationship diagram between the order and the wavelength according to the embodiment.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 light source section 2 light emitting / receiving section 3 spectroscopic section 4 arithmetic processing section 5 color filter thin film for color liquid crystal display device 11 light source 12 reflecting mirror 13 lens 21 light projecting optical fiber 22 light receiving optical fiber 23 light projecting / receiving lens 31 condensing lens 32 pins Hall 33 Lens 34 Planar diffraction grating 35 Imaging lens 36 Image sensor 37 Image sensor drive circuit 38 Buffer amplifier 41 A / D converter 42 Microcomputer 43 Storage device 61 Transparent substrate 62 Adsorption stage 63 Applied paint 64 Clearance 65 Lip 66 Base 67 Adjusting bolt 68 Piping 69 Pump 70 Paint 100 Measuring part

Claims (16)

[Claims] 1. A thin film is irradiated with white light, the spectral intensity of the interference light of the white light by the thin film is measured, and a wavelength giving a maximum value or a minimum value of the spectral intensity is determined based on the measured spectral intensity. A corresponding provisional interference order is calculated, the calculated interference order is corrected by a correction formula separately determined based on the spectral transmission characteristics of the material of the thin film, and the thickness of the thin film is calculated based on the corrected interference order. A method for measuring the thickness of a thin film, characterized in that the thickness is calculated. 2. A thin film is irradiated with white light, the spectral intensity of interference light of the white light by the thin film is measured, and the measured spectral intensity is corrected by a spectral transmission characteristic of a material of the thin film, and the correction is performed. The temporary interference order corresponding to the wavelength that gives the maximum value or the minimum value of the spectral intensity is calculated based on the calculated spectral intensity, and is calculated by the separately determined correction formula based on the spectral transmission characteristic of the material of the thin film. A method of measuring the thickness of a thin film, which comprises correcting the interference order and calculating the thickness of the thin film based on the corrected interference order. 3. The method of measuring a thin film according to claim 1, wherein the thin film has a thickness of 0.4 to 4.0 μm. 4. A method of correcting the measured spectral intensity based on the spectral transmission characteristics of the material of the thin film includes (1) an average spectral intensity of interference light by a plurality of thin films of the same type as the thin film but different in thickness. (2) Spectral intensity difference between the average spectral intensity of the interference light due to a plurality of thin films of the same type as the thin film and different in thickness and the measured spectral intensity with respect to Ask,
(3) The spectral intensity ratio of the measured spectral intensity to the spectral intensity of the white light in the state of excluding the influence of the thin film is obtained. (4) The white light in the state of excluding the influence of the thin film, the thin film For the same kind of calibration as the thin film at the position where the white light is diverging or converging in the optical path of the irradiation optical system for irradiating the The thin film according to claim 2 or 3, wherein any one of a method of obtaining a spectral intensity ratio of the measured spectral intensity of the interference light to a reference spectral intensity measured with a thin film placed thereon is used. Thickness measurement method. 5. When the wavelength giving the maximum value or the minimum value of the spectral intensity is λ, the provisional interference order corresponding to λ is K, and the corrected interference order is M, the correction formula is expressed by equation (1). The method for measuring the thickness of a thin film according to claim 1, 2, 3, or 4. M = K + a + bλ (1) (a and b are coefficients) 6. The method for measuring the thickness of a thin film according to claim 5, wherein -10 ≦ a ≦ 0 and 0 ≦ b ≦ 0.01. 7. An irradiation optical system for irradiating a thin film with white light,
A light receiving optical system that measures the spectral intensity of the interference light of the thin film of the white light, and a temporary interference order corresponding to the wavelength that gives the maximum value or the minimum value of the spectral intensity is calculated based on the measured spectral intensity. Interference order calculation means, interference order correction means for correcting the calculated interference order by a correction formula separately determined based on the spectral transmission characteristics of the material of the thin film, and the thin film of the thin film based on the corrected interference order. An apparatus for measuring the thickness of a thin film, comprising: a film thickness calculating means for calculating the thickness. 8. An irradiation optical system for irradiating a thin film with white light,
A light receiving optical system that measures the spectral intensity of the interference light of the white light by the thin film, a measurement spectral intensity correction unit that corrects the measured spectral intensity by the spectral transmission characteristic of the material of the thin film, and the corrected spectrum. Interference order calculation means for calculating a temporary interference order corresponding to a wavelength that gives a maximum or minimum value of the spectral intensity based on the intensity, and the above-mentioned calculation by a correction formula separately determined based on the spectral transmission characteristic of the material of the thin film. Thin film thickness measuring apparatus, comprising: interference order correcting means for correcting the interference order thus formed; and film thickness calculating means for calculating the thickness of the thin film based on the corrected interference order. . 9. Means for correcting the measured spectral intensity according to the spectral transmission characteristics of the material of the thin film, (1) The average spectral intensity of interference light by a plurality of thin films of the same type as the thin film but different in thickness. Means for obtaining a spectral intensity ratio of the measured spectral intensity with respect to, (2) spectral intensity of the average spectral intensity of the interference light by the plurality of thin films of the same type as the thin film and different in thickness and the measured spectral intensity Means for obtaining a difference, (3) means for obtaining a spectral intensity ratio of the measured spectral intensity to the spectral intensity of the white light in a state in which the influence of the thin film is excluded, (4) in a state in which the influence of the thin film is excluded For the reference spectral intensity measured by placing a calibration thin film of the same type as the thin film at the position where the white light is diverging or converging in the optical path of the irradiation optical system or the light receiving optical system. 8. The thin film thickness measuring apparatus according to claim 7, wherein any one of means for obtaining a spectral intensity ratio of the measured spectral intensity of the interference light is used. 10. When the wavelength giving the maximum value or the minimum value of the spectral intensity is λ, the provisional interference order corresponding to λ is K, and the corrected interference order is M, the correction formula is expressed by equation (1). 10. The thin film thickness measuring device according to claim 7, 8 or 9. M = K + a + bλ (1) (a and b are coefficients) 11. The thin film thickness measuring apparatus according to claim 10, wherein -10 ≦ a ≦ 0 and 0 ≦ b ≦ 0.01. 12. The method of claim 1, 2, 3, wherein the thin film is either an optical filter or an optical filter coating film.
The method for measuring the thickness of a thin film as described in 4, 5, or 6. 13. The method according to claim 7, wherein the thin film is either an optical filter or an optical filter coating film.
10. The thin film thickness measuring device according to 10 or 11. 14. The thickness of an optical filter coating film is measured by the thin film thickness measuring method according to claim 1, 2, 3, 4, 5 or 6, and the measured thickness is within a predetermined range. A method for producing an optical filter, characterized in that the means for forming the coating film is controlled so as to enter. 15. The thickness of the optical filter coating film before curing is measured, and when the measured thickness does not fall within a predetermined range, the coating film of the optical filter is peeled off to obtain the optical filter. 15. The method for manufacturing an optical filter according to claim 14, wherein the transparent substrate used in step 1 is reclaimed. 16. A slit die is used for forming the coating film,
The method for producing an optical filter according to claim 14 or 15, which is one of a spin coater, a roll coater, a bar coater, and a dipping / pulling device.