JPH0431522B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a film thickness measuring device that is applied, for example, to a continuous oil amount meter for a galvanized steel plate flowing at high speed in a line.

[Conventional technology]

Some devices of this type apply the elliplymetry technique to measure the thickness of thin films of several 1000 mm or less. This method is based on changes in the polarization state when light is reflected on a thin film sample surface, that is, the reflectance r _p of the component parallel to the plane of incidence of the electric vector (p component), and the reflectance r p of the component perpendicular to the plane of incidence (s component). The ratio ρ with _s was measured using the following formula: ρ=r _p /r _s =tanφe ^j 〓 ...(1), and was made into a constant function of the already established polarization reflectance ratio ρ and film thickness d. Therefore, the film thickness d is determined. By the way, since the polarization reflectance ratio ρ is generally a complex number, it is necessary to obtain two ellipso parameters, that is, the amplitude ratio tanφ and the phase Δ. Conventionally, these ellipso parameters φ and Δ are determined by irradiating a sample 5 coated with a thin film with monochromatic light 2 emitted from a monochromatic light source 1 through a polarizer 3 and a quarter-wave plate 4, as shown in FIG. This reflected light 6 from the sample surface 5 passes through an analyzer 7 and is received by a photoelectric detector 8 . And this photoelectric detector 8
An electrical signal corresponding to the amount of light received is sent to the feedback control circuit 9, whereby the servo motors 10 and 11 of the polarizer 3 and the analyzer 7 are servo-controlled, and the polarizer angle of the polarizer 3 is calculated from the polarizer angle of the polarizer 3 at this time. The ellipsoparameter Δ is determined, and the same parameter φ is determined from the analyzer angle of the analyzer 7.

However, the above measurement method cannot determine the film thickness unless the refractive index of the base on which the thin film of the sample 5 is applied is known or considered constant.

In addition, a method for measuring film thickness online is disclosed in Japanese Patent Application Laid-Open No. 55-26410, and in this measurement method, when the film thickness is less than several 100 mm, the phase angle Δ of the ellipso parameter is proportional to the film thickness. It takes advantage of the fact that In other words, the film thickness is proportional to the difference between the phase angle Δs (or cosΔs) at the base and the phase angle Δ (or cosΔ) between the base and the thin film. It can be seen that highly accurate measurements cannot be made if the surface roughness changes or the physical properties change. FIG. 7 shows the relationship between the measured phase angle cosΔ and the film thickness, that is, the amount of oil applied, when the phase angle of the base layer is not corrected. The figure shows an example in which a tin-plated steel plate is coated with mineral oil at a concentration of 30 to 60%. The amount is ±
Exceeds the range A-A' of 1 mg/m ² .

On the other hand, in online measurements, since the object to be measured is not specially polished or otherwise pretreated, variations in the roughness of the base surface or minute variations in the physical properties of the base itself cannot be avoided. For example, when measuring an oil film of the order of 100〓 applied on a moving steel plate, the change in phase difference due to the film thickness is δΔ
= about 10°, whereas the phase variation δΔ due to variations in the base reaches at least ±2° to 3° within one coil, and the error is as much as ±20 to 30%. Therefore,
Removing the influence of the substrate is an essential condition. Moreover, this kind of variation in the base changes continuously. There is no point in correcting the base using the refractive index of the base at a distant point, and it is necessary to perform the correction using the refractive index of the base at the same position as the film thickness measurement point. However, in all conventional methods, the refractive index, etc. of the substrate is measured in advance, or the refractive index, etc. of the substrate in front of the paint film is measured at a distance, and the coating is moved at high speed.
Furthermore, it is difficult to apply this method to online film thickness measurement where the refractive index of the underlying layer changes continuously.

[Problem that the invention seeks to solve]

As described above, the conventional apparatus can measure only when the refractive index of the base is considered to be constant, and it is difficult to measure when the refractive index of the base changes continuously.

Therefore, in order to solve the above problems, the present invention has the following features:
It is an object of the present invention to provide a highly accurate film thickness measuring device that can accurately measure film thickness even if the refractive index, physical properties, etc. of the underlying layer vary continuously.

[Means for solving the problem] An irradiation means that irradiates light of two different wavelengths at the same angle of incidence and the same linear polarization onto the object to be measured on which a thin film is applied, and that receives light reflected from the thin film and the base. and a film thickness calculation means for calculating the film thickness of the thin film from the relationship between the phase angle cosine of each reflected light from the thin film and the base for each wavelength of light. .

[Effect]

By providing such means, at least two different wavelengths of light are irradiated from the irradiation means to the object to be measured at the same angle of incidence and with the same linear polarization, and each wavelength of the reflected light is irradiated to the thin film on the object to be measured. For example, the difference in the phase angle cosine of each reflected light from the base is determined by the film thickness calculation means to calculate the film thickness.

〔Example〕

An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 and FIG. 2 are configuration diagrams when the film thickness measuring device of the present invention is applied to a high-speed ellipmeter. By the way, the above-mentioned apparatus was constructed based on the following experimental items, and the experimental items will be explained first, and then the apparatus configuration will be explained.

Now, the experimental items are as follows.

When light with two different wavelengths λ ₁ and λ ₂ are incident on the substrate with the same linear polarization and the same angle of incidence, the cosine of the phase angle of the reflected light that changes to each elliptically polarized light is
cosΔλ ₁ ,s and cosΔλ ₂ ,s change linearly with slight changes in the roughness of the base or the refractive index.

In other words, the following equation holds true.

cosΔλ ₂ , s=lcosΔλ ₁ , s ^-m ...(2) Here, l and m are the object to be measured and the wavelength λ ₁ ,
It is a constant determined by λ ₂ . Therefore _{, Fig. 3 shows that the phase angle cosine cosΔλ 1 , s} and
An example of measuring cosΔλ ₂ ,s is shown below. From the same figure, two cosΔλ ₁ , s,
It is clear that cosΔλ ₂ and s have a linear relationship.

The difference between the cosine of the phase angle for a transparent thin film of several hundred 〓 or less and the cosine of the phase angle of the underlying layer is proportional to the film thickness. In other words, d=k ₁ (cosΔλ ₁ −cosΔλ ₁ , s) ……(3) d=k ₂ (cosΔλ ₂ −cosΔλ ₂ , s) ……(4) Here, d is the film thickness, k ₁ , k ₂ is a proportionality coefficient for wavelengths λ ₁ and λ ₂ , and these proportionality coefficients k ₁ and k ₂
is a constant determined by the angle of incidence φ ₀ on the object to be measured and the object to be measured. In addition, the proportional coefficient
k ₁ and k ₂ are ±10% in absolute value, centered around the average value −−−−n, for fluctuations in the complex refractive index n~ ₂ of the base.
It can be considered constant if it is within the range of , that is, |n~ ₂ --- ~ _{n 2} | 0.1 | n --- ~n ₂ | ...(5). Note that the above equations (3) and (4) are confirmed by calculating the above equation (1). Therefore, Fig. 4 shows the phase angle change cosΔλ using equations (3) and (4) above.
-cosΔλ,s shows the relationship between the film thickness d and the case where the wavelength λ ₁ is 633 nm and λ ₂ is 514 nm. As can be seen from the figure, the difference between the phase angle cosine for the thin film and the phase angle cosine for the underlying layer is proportional to the film thickness d.

Therefore, from the above, the film thickness d can be determined by deriving the above equations (2), (3), and (4) to (6), and it is possible to eliminate the influence of the underlying layer. In other words, d=k ₁ k ₂ /k ₁ −k ₂ l(cosΔλ ₂ −lcosΔλ ₁ +m) =AcosΔλ ₂ +BcosΔλ ₁ +C (6) The coefficients A, B, and C are used to determine the incident angle φ ₀ For example, it is a constant determined by the object to be measured, and can be determined separately experimentally.

In addition, when the fluctuation range of the phase angles Δλ ₁ , s and Δλ ₂ , s in the base is small and the film thickness is small, such as several 100 〓 or less, Δλ ₁ and Δλ ₁ , s and Δλ ₂ and Δλ ₂ , s
Since the difference between is small, the following equation also holds true corresponding to equations (2) to (4).

Δλ ₂ , s=l′Δλ ₁ , s−m′ ……(7) d=k ₁ ′(Δλ ₁ −Δλ ₁ , s)……(8) d=k ₂ ′(Δλ ₂ −Δλ ₂ , s) ...(9) Here, l', m', and _k1 ', _k2 ' are constants. Therefore, for the film thickness d, it is also possible to use the relationship of the following equation corresponding to equation (6).

d=A'Δλ ₂ +B'Δλ ₁ +C'...(10) A', B', and C' are constants.

Now, the configuration of the apparatus shown in FIGS. 1 and 2 will be explained. Figure 1 shows the optical system.
Collimated monochromatic light sources 20 and 21 having different wavelengths λ ₁ and λ ₂ are provided. Then, the light emitted from the monochromatic light source 20 passes through the semi-mirror 22 and the polarizer 2.
3 and is irradiated onto the surface of the sample 24 coated with a thin film, which is the object to be measured, at an angle φ ₀ , and
The light emitted from the monochromatic light source 21 is transmitted through a reflecting mirror 25, a semi-mirror 22, and a polarizer 23, and is irradiated onto the surface of a sample 24 at an angle φ ₀ . Shutters 26 and 27 are arranged on the optical paths of the wavelength lights λ ₁ and λ ₂ , respectively, and are opened and closed by a shutter drive circuit 28. ST is the shutter open/close signal. Note that in the sample 24, the incident surface is parallel to the paper surface, and the light traveling direction is the Z direction. Then, the light traveling direction Z and
It is assumed that the coordinate axis forming a 90° angle is the P axis, and the coordinate axis orthogonal to the P direction and the Z direction is the S axis, and the P, S, and Z directions form a right-handed orthogonal coordinate system. In addition, the polarizer angle and the analyzer angle are all such that the P axis is 0° and the S axis is 90°.

On the other hand, on the receiving side of the reflected light, there is an optical flat (beam splitter part) 2 of the same material and shape.
9a, 29b, 29c are provided, and the reflected light passes through an aperture 30 for limiting the beam diameter and is divided into 3 by each optical flat 29a, 29b, 29c.
The light is now split into a book beam.

These optical flats 29a to 29c
are optically isotropic and transparent, and are fixed parallel to each other, and their thickness and spacing are set so that multiple reflected light is not detected.

Here, for the three beams mentioned above, the beam light ch1 is set to transmit the two optical flats 29a and 29b, and the beam light transmitted through the optical flat 29a and reflected at the optical flat 29b is set. The beam light reflected by the optical flat 29a and transmitted through the optical flat 29c is set to ch3.
It is assumed that the beam light of ch2 and the light beam of ch3 are parallel to each other. Above channel ch
Each beam light of channels 1 to ch3 passes through each analyzer 31a to 31c having a fixed transmission azimuth angle α ₁ to α ₃ , respectively, and passes through each condenser lens 3 having the same focal length.
The light is focused by 2a to 32c, passes through each pinhole 33a to 33c arranged at the focal position, and is further removed by each interference filter 34a to 34c, and then is sent to each photoelectric converter 35a to 35c. The photoelectric detectors 35a to 35c convert the light intensity ( ₁ ) _λ1 , _λ2 to ( ₂ ) _λ1 , _λ2 into electrical signals corresponding to the light intensity. The electrical signals are subjected to certain arithmetic processing by the signal processing circuit shown in FIG. 2 to calculate ellipsoscopic parameters φ and Δ.

In FIG. 2, the electrical signals of ch1, ch2, and ch3 output from the photoelectric converters 35a to 35c, respectively, are passed through the respective sample-and-hold circuits through the respective amplifiers 36a to 36c and the low-pass filters 37a to 37c for removing noise components. (hereinafter abbreviated as S/H circuits) 38a to 38c. These S/H circuits 38a to 38c simultaneously sample the output signals of channels ch1, ch2, and _ch3 using respective gate signals _G1 to G3 output from a microcomputer (hereinafter abbreviated as microcomputer) 39, and then hold them. channels ch1 to ch3 sampled at the same time.
The output is given to the microcomputer 39. The microcomputer 39 receives the shutter opening/closing signal ST from the shutter driver circuit 28, takes in the signals from the SH circuits 38a to 38c for each wavelength λ ₁ and λ ₂ , and from these signals, the phase angle is determined for each wavelength λ ₁ and λ ₂ . It has the function of calculating and determining the cosine cosΔ, and further determining the film thickness from this phase angle cosine cosΔ. It should be noted that the measurement results are outputted as digital outputs D and also as analog outputs through the D/A converter 40. Note that 41 is a preset circuit for presetting each channel gain, eigenvalues ₀ , |σ ₁ σ ₂ |, and the minimum light level I _s min, which will be described later.

Next, the operation of the apparatus configured as described above will be explained. One of the shutters 26 and 27, for example 26, is opened by a shutter driver circuit 28, and the other shutter 27 is closed. Therefore, the light emitted from the monochromatic light source 20 passes through the polarizer 23 and the like and is irradiated onto the sample 24, and the reflected light is reflected from each optical flat 29a, 29.
b, 29c branch into beam light of each channel ch1 to ch3, and each photoelectric converter 35a to 35c
sent to. Here, these photoelectric converters 35a~
Amount of light incident on 35c (I ₁ ) λ ₁ ~ (I ₃ ) λ ₁ , (I ₁ )
λ ₂ ~ (I ₂ ) λ ₂ is the polarizer azimuth θ of 45°, the analyzer azimuth α1, α2, α3 of α1=0°, α2=45°,
When α3=−45° is set, (I ₁ )λ ₁ , λ ₂ = 1/2I _G tan ² φ ……(7) (I ₂ )λ ₁ , λ ₂ = 1/4I _G {tan ² φ+2tanφ ｜σ1・
^σ2 _{| _ _ _ _} ^_
σ2｜×cos(Δ− ₀ )+｜σ1・σ2｜ ² } ...(9). Here, σ1 is the optical flat 2
σ2 is the amplitude reflectance ratio between the P-polarized electric vector and the S-polarized electric vector of the optical flats 29a-29c.

Then, these light amounts (I ₁ ) λ ₁ , λ ₂ ~ (I ₃ )
Electric signals corresponding to λ ₁ and λ ₂ are sent to the amplifier 36a, respectively.
36c and low pass filters 37a to 37c, the signals are sent to S/H circuits 38a to 38c, and then taken into the microcomputer 39. Thus, microcontroller 3
9 calculates the phase measurement value cosΔ by calculating the following equation for each wavelength λ ₁ and λ ₂ .

In addition, the fixed phase difference in equations (10) and 〓
₀ λ ₁ and ₀ λ ₂ can be corrected and obtained as cosΔλ ₁ and cosΔλ ₂ .

Here, the measurement when a tin-plated steel plate is used as the sample 24 will be explained. The amount of oil applied M has the following relationship: M = ρ × d (ρ = oil density) ... (13) Therefore, the amount of oil applied M and the film thickness d are in a proportional relationship and are equivalent to each other. . Then, the _coefficients A ( _82.8 ), B ( _-57.7 ), C (1.61 ) is obtained through a separate calibration experiment, the oil application amount M is M=82.8 (cosΔλ ₂ −0.697cosΔλ ₁ +0.0194)
...(14) is obtained as. Therefore, the microcomputer 39 calculates the oil application amount M by substituting cosΔλ ₂ and cosΔλ ₁ into the above equation and performing calculation processing. Figure 5 shows the correspondence between the measurement results according to the present invention and the measurement results using the hydrofiltration balance method, which has been established as an online method for measuring the amount of oil applied. ²
It shows that the match is within the accuracy. Therefore, the effect of improved accuracy compared to the conventional method is obvious.

In this way, in the above-mentioned embodiment, the sample 24 is irradiated with light of different wavelengths λ ₁ and λ ₂ at the same incident angle and the same linear polarization, and the phase angle cosine of each reflected light from the thin film of the sample 24 and the base is determined. Since the film thickness is determined from the difference in cosΔ, the phase angle cosine of the reflected light
cosΔλ ₁ , s, cosΔλ ₂ , s change linearly with changes in the roughness and refractive index of the substrate, and the difference in each phase angle cosine between the thin film and the substrate is proportional to the film thickness. The film thickness can be accurately measured without being affected by variations in roughness or physical properties. Therefore, it is suitable for online film thickness measurement and can be applied to a high-speed ellipmeter.

Note that the present invention is not limited to the above embodiment. For example, it is applicable even if the relationship between the film thickness d and the phase angle cosine cosΔλ ₁ and cosΔλ ₂ with respect to the coating film thickness is as follows. In other words, for each wavelength λ ₁ , λ ₂ there is a constant relationship cosΔλ ₂ , s= ₀ (cos Δλ ₁ , s) (15) within the variation range of the base, and for each wavelength λ ₁ , λ ₂ If the following relationships hold between the film thickness d and the phase angle: cosΔλ ₁ = ₁ (d) + cosΔλ ₁ , s (16) cosΔλ ₂ = ₂ (d) + cosΔλ ₂ , s (17), then these Eliminate the phase angle of the base from the formulas 1 and 2 and get cosΔλ ₂ − ₂ (d)= ₀ {cosΔλ ₁ − ₁ (d)}……(1
8) It can also be applied when the following relationship is established.

〔Effect of the invention〕

As described in detail above, according to the present invention, it is possible to provide a highly accurate film thickness measuring device that can accurately measure film thickness even if the refractive index, physical properties, etc. of the underlying layer vary continuously.

[Brief explanation of drawings]

Figures 1 and 2 are block diagrams when the film thickness measuring device according to the present invention is applied to a high-speed ellipmeter, Figure 3 is a diagram showing the correspondence of the lower phase angle cosine to each wavelength, and Figure 4 is a diagram showing the correspondence of the lower phase angle cosine to each wavelength. FIG. 5 is a diagram showing the relationship between film thickness and phase angle change, FIG. 5 is a diagram showing the measurement results of the apparatus of the present invention, and FIGS. 6 and 7 are diagrams for explaining the conventional apparatus. 20, 21... Monochromatic light source, 22... Half mirror, 2
3...Polarizer, 24...Sample, 25...Reflector,
26, 27...Shutter, 28...Shutter driver circuit, 29a-29c...Optical flat, 31a-31c...Analyzer, 32a-3
2c...Condensing lens, 33a-33c...Pinhole, 34a-34c...Interference filter, 35a
~35c...Photoelectric converter, 36a-36c...Amplifier, 37a-37c...Low pass filter, 3
8a to 38c...sample and hold circuit,
39...Microcomputer.

Claims (1)

[Claims] 1. In a film thickness measuring device for measuring the thickness of a thin film applied to a base, an irradiation means for irradiating light of two different wavelengths at the same angle of incidence and with the same linear polarization onto an object to be measured on which the thin film is applied; , a film thickness calculating means for receiving the reflected light from the thin film and the base, and calculating the film thickness of the thin film from the relationship of the phase angle cosine of each reflected light from the thin film and the base for each of the wavelengths of light; A film thickness measuring device comprising: eliminating the influence of the underlying change.