JP2933666B2 - Infrared thickness measuring device for multilayer structure - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial application field) The present invention relates to a multilayer film composed of two or more materials, a multilayer coating film on a metal, a specific layer of a multilayer structure such as a coating film on paper, or the like. The present invention relates to an optical thickness measuring device for measuring a thickness.

(Prior Art) Optical infrared film thickness measurement of a single-layer film or the like is performed based on the principle that the thickness is proportional to the infrared absorbance.

Conventionally, an infrared ray having a wavelength in a characteristic absorption band (hereinafter referred to as measurement light) of a substance forming an object to be measured, which is a target of thickness measurement, and a reference light having a wavelength appropriately separated from the characteristic absorption band. (Hereinafter simply referred to as reference light) are respectively incident on the object to be measured from a direction perpendicular thereto, and the transmission intensity of the measurement light transmitted through the object to be measured and the transmission intensity of the reference light are measured. Determine the absorbance of the measured object,
An infrared thickness gauge that calculates the thickness of the object to be measured based on the fact that the absorbance is proportional to the thickness is well known.

In recent years, various multilayer structures such as multilayer films have been used, and it is necessary to accurately control the thickness of each layer of the multilayer structure in order to exhibit its function. Therefore, a film thickness measuring device for measuring and managing these is required.

There are various multilayer structures as shown in FIGS. 8 (a) to 8 (d), for example. As shown in FIG. 8 (a), there are a film in which two kinds of films 101 and 102 are laminated, and a film in which a coating film 112 is applied on a film 111 as shown in FIG. 8 (b). Further, as shown in FIG. 8 (c), a coating film 123 is applied on a metal base 121 with a base coating film 122 interposed therebetween, or as shown in FIG. 132 is applied.

The measurement of the thickness of these multilayer structures has a specific problem different from that of a single-layer film or the like.

The principle that the absorption is proportional to the thickness of each layer constituting the multilayer structure also holds for the multilayer structure.
However, it is a problem how to apply this principle for the whole multilayer structure.

In the infrared multilayer film thickness measuring method described in JP-B-58-9362, for example, a film A and a film B are used.
In the case of a two-layer film consisting of:

Here, λ ₁ and λ ₂ are characteristic absorption wavelengths of each film, and λ _R is a wavelength (reference wavelength) that is hardly absorbed by both films. Also, I ₀ (λ ₁ ), I ₀ (λ ₂ ),
I ₀ (λ _R ) is the incident light intensity of the infrared light incident on the film at the wavelengths λ ₁ , λ ₂ , and λ _R , respectively.
(Λ ₁ ), I (λ ₂ ), I (λ _R ) are the wavelengths λ ₁ ,
are the transmitted light intensities of the infrared light transmitted through the film at λ ₂ and λ _R , and α (λ ₁ ), α (λ ₂ ) and α (λ _R ) are the wavelengths λ ₁ , λ ₂ and λ _R , respectively. Are the intrinsic absorption coefficients of the film A at β, λ (λ ₁ ), β (λ ₂ ), β (λ _R )
Respectively, the wavelength lambda _1, lambda _2, a specific absorption coefficient of the film B in the lambda _R, d _A is the thickness of the film A, d _B is the thickness of the film B. Therefore, if the intrinsic absorption coefficients α (λ) and β (λ) of each film are known, the amount of transmitted light at each wavelength is measured, and the ratio I (λ ₁ ) / I (λ
_R), by obtaining the _{I (λ 2) / I (} λ R), by solving the simultaneous equations, the thickness d _A of the film can be calculated d _B and the entire film thickness d _A + d _B.

(Problems to be Solved by the Invention) In the case of using the conventional film thickness measuring method described above, even if the thickness of only a part of the layers constituting the multilayer structure is measured, the absorption of the layer which does not need to be measured is measured. Coefficients must also be determined.

If the absorption coefficient of a layer other than the layer to be measured cannot be determined, measurement cannot be performed. For example, the film thickness of a coating film (layer 123 in FIG. 8C) on an underlayer on metal In the measurement, when the material and thickness of the base layer 122 are unknown, or in the thickness measurement of the coating film (layer 132 in FIG. 8D) on the paper as a reflector, the absorption of the surface layer 133 of the paper is not sufficient. This is a fixed case.

In the measurement of light absorption of the multilayer structure, the effects of surface reflection and internal interface reflection and the effects of light interference caused by multiple internal reflection cannot be ignored, and the amount of light absorption proportional to the thickness of each layer is accurately measured. There is a problem that it cannot be detected.
Especially when the influence of surface reflection or light interference is large, for example,
When the total thickness of the multilayer structure is as thin as 20 μm or less, when it is a coating film on metal, when it contains a material with a high refractive index (high reflectivity), when it has high transparency, or when it has a smooth surface In such a case, the SN ratio decreases, and practical thickness accuracy cannot be obtained.

SUMMARY OF THE INVENTION An object of the present invention is to provide an infrared thickness measuring device capable of accurately and easily measuring the thickness of a specific layer or each layer even in a multilayer structure which has conventionally been difficult to measure.

(Means for Solving the Problems) An infrared film thickness measuring apparatus for a multilayer structure according to the present invention converts P-polarized light of measurement light of a plurality of wavelengths belonging to a characteristic absorption band of the multilayer structure sample into the multilayer structure sample. Light incidence means for entering at an angle near the polarization angle of the measured layer; light detecting means for detecting the intensity of light of each of the plurality of wavelengths transmitted at least once through the measured layer; and An absorbance measuring means for determining the absorbance of the sample for each wavelength from each of the received light intensities output from the detecting means, and a polynomial including at least a first-order term of the absorbance for each wavelength in a plurality of wavelengths, the thickness of the layer to be measured. And a thickness calculating means for calculating the thickness of the layer to be measured from the measured value of the absorbance using the calibration curve formula represented by

(Operation) In optics, it is known that reflection does not occur when P-polarized light is incident at the polarization angle of the substance (in other words, this gives the definition of the polarization angle). P parallel to the plane of incidence
Polarized light is incident on the DUT at an angle equal to or near the intrinsic polarization angle (Brewster angle) of the DUT, and the multiple reflection in the DUT is greatly reduced, resulting in surface reflection and interface reflection. And the influence of interference due to multiple reflected light are suppressed to such an extent that measurement accuracy is not adversely affected.

Further, as a calibration curve equation for calculating the thickness t _k of the k th measured layer of the multilayer structure from the absorbance, the form including at least first-order terms of the absorbance of the total sample at a plurality of wavelengths belonging to the characteristic absorption band of the sample polynomial Is used. For example, the following polynomial including at least the first-order term is used.

t _k = ΣC _i A _i + ΣC _ii A _i ² + C ₀ where A _i is the absorbance of the entire sample at the ith wavelength and C ₀ , C _i and C _ii are the calibration curve coefficients . Although the absorbance A _i of each wavelength includes the absorption of layers other than the layer to be measured, the absorbance of multiple wavelengths with different absorption characteristics of each layer should satisfy the above formula for samples with different layer thicknesses. If the calibration curve coefficient is selected, the absorption of layers other than the layer to be measured can be canceled.

The calibration curve formula of this form includes the thickness t _j (j ≠
k) and absorption coefficient are not included. Therefore, when measuring the thickness t _k only part of the layers of the multilayer structure, there is no need to determine the thickness and absorption coefficient of the measurement unnecessary layers. For example, when measuring the thickness of a coating film on a film, it is not necessary to determine the film thickness and the absorption coefficient of the film that do not need to be measured.

In addition, when the absorption coefficient or thickness of the undercoat layer is unknown by measuring the thickness of the coating film on the undercoat layer on metal, or by measuring the thickness of the coating film on paper as a reflector, the absorption of the surface layer of the paper A calibration curve can be created even when the absorption coefficient of a layer other than the measurement layer cannot be determined, for example, when it is uncertain.

The form of the polynomial may be appropriately selected according to the properties of the sample. The second order term of the absorbance in the case of the above formula, the relationship between the thickness t _k and the absorbance A _i according absorbance increases Considering that coming out from the linear relationship is obtained by adding to the correction. Therefore, it is not necessary to add a quadratic term for all of the used wavelengths, and only a wavelength having a large absorbance is required.

The ratio of the absorbance A A _i, the incident light intensity at the i th wavelength of the multilayer structure is to be measured (i.e., the received light intensity when no multilayer structure) I ₀ and the transmitted light intensity I _i Logarithm of lo
g (I ₀ / I _i ).

The measurement wavelength is generally selected to be a wavelength characteristic of each layer of the multilayer structure and having a relatively large absorption in the measurement thickness range. The number of measurement wavelengths is equal to or more than the number of layers of the multilayer structure (the number of constituent materials). This is because the absorbance A _i is a function of the absorbance of each layer, and is considered to include an unknown number of constituent materials.

When performing baseline correction at a reference wavelength that does not belong to the characteristic absorption band of the multilayer structure, the incident light intensity I _ro at the reference wavelength,
It corrected from the transmitted light intensity I _r and absorbance A _r absorbance A _i '
May be used.

A _i ′ = A _i −A _r = log (I ₀ / I _i ) −log (I _r0 / I _r ) = log (I _r / I _i ) −log (I _r0 / I ₀ ) May be selected from one or two wavelengths having low absorption in both the layer to be measured and the other layers.

〔Example〕

Hereinafter, embodiments of the present invention will be specifically described with reference to the accompanying drawings.

As shown in FIG. 1, the infrared thickness measuring device includes a measuring unit A and a calculating unit B, and the detected thickness is displayed and recorded by a display recording device C.

The measurement unit A makes the multilayer film 11 as an object to be measured enter a plurality of P-polarized infrared rays at an angle equal to or near a polarization angle determined by the refractive index of the layer to be measured. it is for measuring the light Ra _i of a plurality of measurement wavelengths lambda _i including a reference wavelength lambda _R the absorbance due.

It should be noted that there is not much difference in the degree of reduction of multiple reflections near the polarization angle. The refractive index of a film or the like to be measured is usually about 1.50 to 1.65, and its polarization angle is about 56 to 59 degrees. Therefore, the incident angle may be set to 56 to 59 degrees.
Further, even if the incident angle is slightly shifted, the accuracy is slightly reduced, and the object of the present invention is not spoiled. Furthermore, it is possible to measure at one incident angle even if the materials laminated by the multilayer film are different.

The measuring section A includes an infrared light source 12 composed of, for example, a nichrome line light source.
A wide band of infrared rays including the characteristic absorption band is generated.

The infrared light emitted from the infrared light source 12 is focused on the rotating disc 14
Projected above. The rotating disk 14 is made of a light-shielding material and is driven to rotate by an electric motor 15 at a rotation speed of about 30 Hz. As shown in FIG. 6 to 8 band-pass filters 16 that allow only the light of each measurement wavelength to pass therethrough.
a,…, 16h are installed. The infrared light source 12, the electric motor 15, and the band-pass filters 16a,..., 16h constitute a chopper for the above-mentioned infrared rays, and sequentially pass infrared rays of each wavelength. In order to detect the passage timing of the infrared light of each of these wavelengths, first and second photoelectric switches 18 and 19 formed of a combination of a light emitting diode and a phototransistor facing the outer periphery of the rotating disk 14 are disposed therebetween. Is installed, and on the outer diameter side of the first bandpass filter 16a,
Two light passing holes 20 and 21 for turning on both the first and second photoelectric switches 18 and 19 are provided in the other band-pass filter 1.
On the outer diameter side of 6b,..., 16h, one light passage hole 21 for turning on only the first photoelectric switch 18 is provided.

On the light passing side of the rotating disk 14, a pair of opposing reflecting mirrors 22, 23 are installed. The reflecting mirror 22 corresponding to the spot position of the rotating disk 14 reflects the passing light toward another opposing reflecting mirror 23, and the reflecting mirror 23 is equal to the polarization angle determined by the refractive index of the layer to be measured. Light is incident on the multilayer film 11 at an inclination angle near the polarization angle. In the optical path from the reflection mirror 23 to the film 11, a polarizer 24 is provided, which polarizes incident infrared light, that is, passes only a polarized light component parallel to the incident surface, and passes through the polarizer 24. By doing so, the infrared light becomes a P-polarized wave.

The P-polarized infrared light is reflected by the reflector 25 so as to pass through the multilayer film 11 twice, and
Infrared rays degrees passed him Shirubebi the infrared detector 27 by the concave reflecting mirror 26, an infrared detector 27 outputs a detection signal proportional to the received light intensity of the infrared rays R _ai of each wavelength to the detection circuit 28. As the infrared detector 27, for example, a PbSe detector or a PbS detector can be advantageously used, but is not limited thereto. Further, in order to increase the absorbance, the infrared light is transmitted twice through the multilayer film 11. However, when the thickness of the multilayer film 11 is large, the infrared light may be transmitted only once, that is, the transmitted light may be measured.

The detection circuit 28 amplifies the output of the infrared detector 27 and shapes the waveform to form a dark current cancel circuit 29 in the next stage.
Output to

As shown in FIG. 3 (a), the output signal of the detection circuit 28 is a signal V _ai (hereinafter simply referred to as a measurement light transmission signal) proportional to the received light intensity of the measurement light R _{ai, as} a dark current component V _d . It is a superimposed signal. Dark current component V _d is the noise generated by the measuring unit A and out of the heat radiation and the like.

The dark current canceling circuit 29 includes a timing generation circuit 3 that uses each output of the first and second photoelectric switches 18 and 19 as an input signal.
The first timing signal T ₁ of the 0 at the timing when both are not projected infrared R _ai, reads the output signal of the detection circuit 28, i.e. reads the dark current component V _d, the dark current component V _d read Subtracting from the output of the detection circuit 28, the third
As shown in FIG. 7B, only the measurement light transmission signal V _ai is extracted and separated.

Next, the configuration of the arithmetic unit B will be described. The arithmetic unit B transmits a plurality of measurement light transmission signals V _ai output from the dark current canceling circuit 29 to the second
Respectively individually sample hold timing of the timing signal T ₂ (FIG. 3 (c), (d) refer) and sample-and-hold circuit 31, and the wavelength measurement light transmission signal V _ai which is sampled and held by the sample-and-hold circuit 31 From each of the previously obtained sample-free wavelength measurement light passing signals V _bi and
og (V _bi / V _ai) i.e. the log ratio circuit (absorbance calculation section) 32 for calculating the respective wavelength absorbance A _i, the absorbance A _i at each wavelength in the form and the S reference sample of the set calibration curve equation and from the thickness data of the measured layer (known), and a calibration curve calculation unit 33 for calculating a calibration curve factor C _i, the thickness t _k of the measurement layer and a respective wavelength absorbance a _i and calibration equation of the measured multi-layer body And a thickness calculation circuit 34 for calculating The calculations in the calibration curve calculation unit 33 and the thickness calculation unit 34 are performed using known means.

The procedure of calculating the calibration curve coefficient and measuring the thickness of the unknown sample is performed according to the flow shown in FIG.

The preparation of the calibration curve, that is, the determination of the calibration curve coefficient is performed as follows.

(A) First, to specify the shape of the calibration curve equation used for thickness k _k of the measurement layer (k-th layer) (step S1). In this embodiment, the following polynomial is used.

Where A _i is the absorbance at the ith wavelength,
C ₀ , C _i , and C _ii are calibration curve coefficients.

 Here, the number of wavelengths is four here.

(B) Next, a plurality of reference samples are prepared in which the thickness of each layer constituting the multilayer structure is variously changed in advance so as to include the thickness range to be actually used (step S2).
A part of the reference sample may include a monolayer. The number S of reference samples is set to be larger than the number of calibration curve coefficients that are unknown.

(C) Next, the used wavelength λ _i and the reference wavelength λ _R are determined from the infrared absorption spectrum of the material of each constituent layer of the multilayer structure (Step S3). Then, for example, in the apparatus of FIG. 1, the band-pass filters 16a,.
Attach 6h.

(D) On the other hand, a t _k and the thickness of the measured layer of the reference sample, accurately measured by other methods as a reference (step S
Four). This data is input to the calibration curve calculation unit 33.

(E) Next, with respect to each reference sample, the absorbances A _ji and A _jr at each used wavelength and the reference wavelength are measured (step S).
Five). Here, j represents a sample number (j = 1,...,
S), i represents a wavelength number (i = 1,..., 4), and r represents a reference wavelength. That is, each reference sample is set, and the rotating disk 14 is rotated to obtain the output (absorbance) of the log ratio circuit 32.

(F) Next, the absorbance A _ji of each reference sample and the thickness t _kj of the k-th layer are substituted into a predetermined calibration curve equation, and S simultaneous equations are obtained. To obtain calibration curve coefficients C _i , C _ii , and C ₀ and store them (step S6). This calculation is performed by the calibration curve calculator 33 by, for example, the least squares method. Thus, the calibration curve equation (1) is determined, and the obtained calibration curve coefficients are coefficients that minimize the influence of the layers other than the k-th layer.

In the above, the calibration curve equation corresponding to the k-th measured layer was obtained, but the calibration curve equation corresponding to the k-th measured layer is similarly obtained.

When determining a plurality of calibration curve equations, the calculations in the calibration curve calculation unit 33 can be performed simultaneously.

The measurement of the unknown sample is performed using the calibration curve equation (1) determined as described above.

First, the absorbances A _ji and A _jr of the unknown sample are measured at the above wavelengths (step S7).

Then, it calculates the thickness t _k by applying the measurement data to the calibration curve equation (1), and outputs (step S8). That is, the thickness calculating circuit 34 calculates the thickness t _k from the absorbance output from the calibration curve factor and the log ratio circuit 32 which is output from the calibration curve calculation unit 33.

In order to simultaneously output the thicknesses of a plurality of layers, the absorbance data is applied to the calibration curve equation (1) corresponding to each measurement layer.
In this case, the thickness calculation circuit 34 can simultaneously calculate the thicknesses of the plurality of measurement target layers from the measured values of the absorbance for each wavelength using the plurality of calibration curve equations (1).

Next, an example of measuring the thickness of each layer of a two-layer film of nylon and an ethylene vinyl alcohol copolymer (trade name: EVAL) will be described.

The samples used for the measurement were a single-layer film A of nylon (15 μm thickness) and a single-layer film B of Eval (15 μm thickness)
And nylon / EVAL (10μm / 5μm thick and 10μm / 7μm
(Thickness), each of which is a total of four of the two-layer films C and D. Further, two types of combinations (A + D, B + C) in which two single-layer films and two-layer films are overlapped were also used as samples. (See table). That is, a total of six types of thickness combinations of the respective materials were selected as samples.

The measurement wavelengths are 3605, 3762, 4062, and 4883 nm.

The absorbance A _i at each wavelength was measured by the above thickness measuring device.

The thickness of the nylon layer of each of the six reference samples was measured using an optical microscope and an electronic micrometer together, and a calibration curve for measuring the nylon layer was determined from the thickness and the absorbance.

 Similarly, a calibration curve for eval layer measurement was determined.

The equation of the calibration curve, the thickness t ₁ of the nylon layer, with thickness t ₂ are both of the form of the EVAL layer.

_{t k = C k0 + C k1} (A 1 -A 4) + C k2 (A 2 -A 4) + C k3 (A 3 -A 4) (k = 1,2) (3) Here, t _k is the sample layer a thickness, a ₁ to a ₄ is an absorbance measured by the band-pass filter having a center wavelength of _{λ 1 = 3605,3762,4064,4883nm, C k0 ~C} k4 are calibration coefficients.

The obtained calibration curve factor for nylon _{layer, C 10 = -1.7959, C 11} = 140.53, C 12 = -262.15, C 13 =
200.95, C ₂₀ = 0.3393, C ₂₁
= -49.517, C ₂₂ = 170.54, was C ₂₃ = -208.88. After the preparation of the calibration curve, measurement was performed on the other six kinds of samples using the above calibration curve, and at the same time, the measured values (reference thickness) by an optical microscope and an electron micrometer were obtained.

The table shows the results. 5 and 6, and FIG.
The figures show correlations for nylon thickness measurement, Eval thickness measurement, and total thickness measurement of nylon and Eval, respectively.

The maximum deviation of the thickness of the nylon layer, the EVAL layer and the total layer was 0.37 μm (3.6%) and 0.51 μm (10.5%, respectively).
%) And 0.37 μm (2.2%), and the measurement could be performed accurately.

(Effects of the Invention) The thickness of only a part of the layers constituting the multilayer structure can be measured without measuring the extinction coefficient of the layer whose measurement is unnecessary.

[Brief description of the drawings]

FIG. 1 is a diagram showing a schematic configuration of an infrared thickness measuring device. FIG. 2 is a plan view of the rotating disk. FIG. 3 is a timing chart of infrared absorption measurement. FIG. 4 is a flowchart of a procedure for measuring the thickness of the multilayer structure. FIG. 5, FIG. 6 and FIG. 7 are graphs showing the results of measurement of the thickness of the nylon layer, the eval layer, and the total thickness of the nylon layer and the eval layer, respectively, of the multilayer structure. 8 (a), (b), (c) and (d) respectively show
It is sectional drawing of the example of a multilayer structure. 12 ... Light source, 11 ... Multilayer film, 16,17 ... Band pass filter, 24 ... Polarizer, 27 ... Infrared detector, 33 ... Calibration curve calculator, 34 ... Thickness calculator.

Claims (1)

(57) [Claims] 1. A light incident means for injecting P-polarized light of measurement light of a plurality of wavelengths belonging to a characteristic absorption band of a multilayer structure sample at an angle near a polarization angle of a layer to be measured in the multilayer structure sample, A light detecting means for detecting the intensity of light at each wavelength of the plurality of wavelengths transmitted at least once in the layer; and an absorbance of the sample at each wavelength based on the received light intensity output from the light detecting means. The absorbance measurement means to be determined and the thickness of the layer to be measured are determined from the measured absorbance at each wavelength by using a calibration curve formula expressed by a polynomial including at least the first-order term of the absorbance at each wavelength among a plurality of wavelengths. An infrared thickness measuring apparatus for a multilayer structure, comprising: a thickness calculating means for calculating a thickness of a measuring layer.