JPH09218133A - Method for inspecting anisotropic thin film - Google Patents

Abstract

PROBLEM TO BE SOLVED: To measure the molecular orientation of a thin film formed on a glass base by obliquely emitting a light to a thin film sample surface from a plurality of directions, and measuring the anisotropy of the polarized state of the reflected lights. SOLUTION: The light of a light source 1 is made into a linear polarized light with a refractive index ϕ=π/4 and a film thickness Δ=0 by a polarizer 2, and incident on the surface of a sample 6 with a fixed incident angle from the normal of the sample 6 surface. A phase plate 3 determines the quadrant of the film thickness Δ by its insertion and removal. The reflected light by the sample 6 is entered to an analyzer 7, and the intensity of the transmitted light is measured by a light receiving tube 8. At this time, the analyzer 7 is rotated to determine the refractive index Δ and the film thickness Δ of the transmitted light intensity from the Fourier sum, but the film thickness Δ has two values because of the periodic function. Thus, the polarized state of the incident light is changed by inserting the phase plate 3 to perform the measurement, and the film thickness Δ of the same value as that determined when no phase plate 3 is present is finally adapted. The anisotropy of the sample 6 is measured by rotating rotating stage 9. Thus, the molecular orientation and the thickness of the oriented part can be measured.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an organic thin film used for sensors and wiring materials, and a method and apparatus for evaluating an organic thin film for controlling the orientation of liquid crystal molecules in a liquid crystal display device.

[0002]

2. Description of the Related Art The orientation of molecules in an organic thin film has a great effect on the function of a device using the same. In particular, in an organic thin film used to give an initial alignment to liquid crystal molecules in a liquid crystal display device for controlling the alignment of liquid crystal molecules in a liquid crystal display device, it is known that there is a close relationship between the alignment of the organic thin film molecules and the alignment of the liquid crystal molecules. (Ishihara et al., Liquid Crystals, Vol. 4, No. 6, page 669, 1989), the higher the molecular orientation of the organic thin film, the greater the alignment regulating force of the liquid crystal molecules. For this reason, quantitative measurement of the molecular orientation of the organic thin film is important for evaluating the function of the device.

The evaluation of organic thin films (particularly liquid crystal alignment films) is centered on methods of observing the state of molecules from molecular vibrations such as infrared absorption spectroscopy and Raman scattering spectroscopy. The knowledge about the degree of molecular orientation and the orientation direction in the thin film is evaluated by measuring the optical anisotropy of the film due to the molecular orientation by measuring the dichroic ratio using the polarization of light (JP-A-6- No. 160862, Ezawa et al. "Method for evaluating characteristics of liquid crystal display device and its alignment film").

In addition to the vibrational spectroscopy, the optical anisotropy is evaluated by the anisotropy of the birefringence phase difference of the light transmitted through the sample (JP-A-6-102512, Kurai et al., "Liquid Crystal Display"). Device orientation evaluation device and liquid crystal display device manufacturing method "). In addition, a method has been proposed in which the in-plane refractive index anisotropy of the film caused by the molecular orientation is observed by injecting linearly polarized light whose polarization direction is horizontal to the film surface or orthogonal thereto. (JP-A-4-9584
No. 5 publication Ishihara "Method for evaluating liquid crystal alignment ability of alignment film").

As a method other than these, the shape of the thin film surface is two-dimensionally measured by an atomic force microscope or a scanning tunneling microscope (Isono et al. Japan Society for the Promotion of Science 142 Committee A Section Special Research Group) Sample 34 Page 1
994).

[0006]

The method based on vibrational spectroscopy using light such as infrared absorption is used to measure a liquid crystal display device having a transparent electrode film formed on a glass substrate and a liquid crystal alignment film formed thereon. In addition, the influence of the glass substrate and the transparent electrode film cannot be avoided. In particular, infrared rays having a wave number lower than 1500 cm ^-1 do not pass through the glass, so that the absorption spectrum cannot be measured. The observation of the molecular orientation of the liquid crystal alignment film by infrared spectroscopy has been conducted by paying attention to the absorption at 1240 cm ^-1 (Sawa et al., Japanese Journal of Applied Physics, Japanese Journal).
of Applied Physics Volume 33 627
Page 3, 1994), it is impossible to inspect the alignment film of the liquid crystal display element actually used.

When the birefringence phase difference is measured, it is difficult to measure the birefringence phase difference of the alignment film itself because the glass substrate itself usually has birefringence due to strain or the like. Therefore,
It is not possible to accurately evaluate the alignment state of molecules in the alignment film by a method known in the related art. In addition to this, the birefringence phase difference φ has a relationship of φ = 2π (Δn · d) / λ (1) among the birefringence Δn, the film thickness d, and the wavelength λ of light. The amount obtained from the measurement of the phase difference φ is the birefringence Δn caused by the film thickness d and the molecular orientation.
It is shown to be the product of It is known that the rubbing-treated polyimide film, which is widely used as an alignment film in liquid crystal display devices, is aligned not on the entire film but on the surface (Sawa et al., Japanese Journal of Applied Physics Japanes Jourse.
nal of Applied Physics, Vol. 33, p. 6273, 1994), molecular orientation cannot be quantitatively known unless the thickness of the oriented portion can be measured.

A method has been proposed in which the in-plane refractive index anisotropy of the film is measured from the dependence of the intensity of reflected light from the film on the polarization state of the incident light and the in-plane incident direction of the film to measure the molecular orientation. However, it is not possible to measure the film thickness of the aligned portion similarly to the birefringence retardation measurement, since molecular weight can be measured in the same manner as the birefringence phase difference measurement (Japanese Patent Laid-Open No. 4-95845, Ishihara “Method for evaluating liquid crystal alignment ability of alignment film”). Can not. Further, the technical difficulty in carrying out the measurement includes the effect of the anisotropy of the surface shape. It is known that a polyimide film, which is widely used as a liquid crystal alignment film, not only has the molecules of the film aligned by rubbing, but also has a fine groove-like shape that runs along the rubbing direction on the surface (Isono et al. Japan. 142 Committee A, Japan Society for the Promotion of Science
Special report of the working group, 34 pages, many reports such as 1994). Because of the presence of this groove, the light incident on the surface of the film exhibits in-plane anisotropy in the amount of light scattered in directions other than the specular reflection direction. The amount does not accurately reflect the optical anisotropy caused by the molecular orientation of.

Observation by an atomic force microscope only shows surface morphology such as surface roughness, and there is no example observed with an atomic level resolution in an alignment film. In the case of a liquid crystal alignment film, it has been reported that the surface shape observed by this method has almost no effect on the alignment state of liquid crystal molecules and has no correlation with the molecular alignment in the film (Isono et al., Japan Society for the Promotion of Science). 142 Committee A Subcommittee Special Study Group Sample 34 pages 199
4 years). Further, it is known that a film obtained by treating the surface of a rubbed film with an organic solvent such as acetone has a force for regulating the alignment of liquid crystal molecules, but does not have a groove-like form on the surface caused by the rubbing. Thus, the surface morphology observation is not an appropriate evaluation method because it does not give direct information to the alignment control force of the liquid crystal molecules of the liquid crystal alignment film.

On the other hand, as an optical evaluation method for a thin film, the film thickness, the refractive index, the absorptance, etc. are measured from the polarization state of the reflected light generated when the light is incident on the surface of the sample in a certain direction, the incident angle and the polarization state. Ellipsometry is widely practiced. This method uses two of the three quantities of film thickness, refractive index, and absorptance on a isotropic substrate whose refractive index and absorptivity are known.
The following amount is calculated from the phase difference Δ and intensity ratio (tan ψ) 2 of the S-polarized component of the reflected light (polarized component parallel to the sample surface) and the P-polarized component perpendicular to it, the incident angle, and the wavelength of the light. (Azzam and others "Ellipsometry and Polalized light" North-Holland 1987 and Zaghloul and others Applied Opts (Applied Opt)
ics) Vol. 21, No. 4, p. 739, 1987). In this calculation, both the substrate and the thin film are treated as optically isotropic substances.

On the other hand, the liquid crystal alignment film has a portion having optical anisotropy on an isotropic portion on the substrate, and is substantially 2
Since the film of layers is on the substrate and one is optically anisotropic, conventional ellipsometry cannot measure the thickness and refractive index of each of the isotropic and anisotropic portions.

Therefore, a first object of the present invention is to provide a method and apparatus capable of evaluating the state of molecular orientation in an alignment film formed on a glass substrate.

A second object of the present invention is to provide the thickness of the molecularly oriented portion of the outermost surface of the liquid crystal alignment film formed on the glass substrate, the main coordinate system, and the main dielectric constant (dielectric constant tensor as a diagonal matrix). The orientation of the coordinate system to be expressed and the value of its diagonal component (Yoshihara “Physical Optics” Kyoritsu Shuppan, 1986, p. 186), and the refractive index and thickness of the non-aligned part were measured, and the liquid crystal alignment film It is an object of the present invention to provide a method and a device capable of evaluating the molecular orientation of a surface.

[0014]

In order to solve the above-mentioned first problem, the present invention measures the polarization state of reflected light generated by injecting monochromatic light having a constant polarization state onto the surface of a thin film sample. By doing so, the molecular orientation of the thin film is measured. In the present invention, since the reflected light is observed, it is possible to measure the molecular orientation of the thin film formed on the strained glass substrate.

Since the present invention measures not the intensity of the reflected light but the polarization state, the optical anisotropy of the film is not affected by the change in the scattered light intensity due to the anisotropy of the surface shape such as roughness and grooves. Can be measured.

Usually, the electric field vector intensity of light of angular frequency ω observed at a fixed position in space at time t is two directions XY orthogonal to each other defined on a plane perpendicular to the propagation direction of light. X component A1 × exp (i (ωt + δ1)) y component A2 × exp (i (ωt + δ2)) (2) δ1 and δ2 are the initial positions in each direction,
A1 and A2 are amplitudes in the respective directions. The polarization state excluding the absolute value of the intensity is the ratio of these two components (A1 / A2) × exp (i (δ1-δ2)) (3), and tan ψ = (A1 / A2) Δ = (δ1-δ2) ( 4) It is represented by two quantities ψ and Δ defined in. It is the refractive index, absorption coefficient and film thickness of the film that affect these two parameters, but the absorption coefficient of a polyimide film, which is usually used for liquid crystal alignment film, is less than 1/1000 in the visible light region. Since the wavelength dependence is small, it can be treated as a substance that does not absorb. Therefore, two parameters (refractive index, film thickness) representing the state of the film can be determined from the two measured parameters (ψ, Δ).

According to this method, it is possible to independently obtain the film thickness and the refractive index of the portion where the molecules are oriented, which cannot be known by the birefringence phase difference measurement or the reflected light intensity anisotropy measurement.
Quantitative measurement of molecular orientation is possible.

Since the known quantity is increased by measuring the incident light wavelength dependence and the incident angle dependence of the polarization state of the reflected light,
The distribution of the refractive index of the film in the depth direction can be obtained according to the number of measured conditions.

Incidentally, in the birefringence phase difference measurement for measuring the polarization state of the transmitted light, there was a problem that the strain of the glass substrate supporting the film had a great influence, but in the case of the reflected light, it was different like the distorted glass substrate. Since the phase change that occurs when the material is anisotropic is reflected by the surface, it is determined only by the magnitude relationship between the reflective medium and the refractive index of the material through which incident light and reflected light propagate. The phase of the reflected light hardly changes due to the anisotropy of the ratio. Under such circumstances, by measuring the anisotropy of the phase change of the reflected light, the anisotropy of the film near the surface can be measured without being affected by the optical anisotropy of the underlying substrate such as the distortion of the glass substrate. .

To measure the anisotropy of the reflected light, first, the light from the light source is made to pass through a spectroscope and a polarizer to form incident light having a constant polarization state and wavelength, and the incident light is made incident on the sample surface. The polarization state of the reflected light generated in 1 is measured using an analyzer. Anisotropy is measured by rotating the sample in-plane to change the incident direction of the light and measuring the polarization state dependence of the reflected light. Anisotropy is measured at the same time by measuring the polarization state of the reflected light of a plurality of light rays with different directions. The polarization states of incident light and reflected light are represented by a component parallel to the sample surface and a component orthogonal thereto. In the case where the incident light is ψ = π / 4 in a state where the value of the component that is horizontal to the surface and the component that is orthogonal to it are the same, it is most suitable for simultaneously obtaining the film thickness and the refractive index.

In order to solve the above-mentioned second problem, the present invention makes monochromatic light having a constant polarization state incident on the surface of a liquid crystal alignment film sample, and causes the polarization state of reflected light (S polarization component and P (Phase difference and intensity ratio of polarization components) are observed from multiple directions in the sample plane, and the principal permittivity and principal coordinates of the orientation part that directly reflects the molecular orientation state of the film from the incident direction dependence of the polarization state of the reflected light. The molecular orientation of the liquid crystal alignment film surface is evaluated by obtaining the angle and thickness of the system with respect to the film surface, and the refractive index and thickness of the non-aligned portion that does not cause molecular alignment even by rubbing treatment.

Even if there is strain, the phase difference between the S-polarized component and the P-polarized component does not change when reflected on the substrate surface where light absorption is extremely small. Therefore, the film was formed on a strained glass substrate. Liquid crystal alignment film The in-plane incident direction dependence of the phase difference between the S-polarized light component and the P-polarized light component of the reflected light from the sample is the main dielectric constant of the alignment portion corresponding to the molecular alignment state of the liquid crystal alignment film, the film surface of the main coordinate system It accurately reflects the angle, the thickness, and the refractive index of the non-oriented portion that does not cause molecular orientation even by the rubbing treatment. For this reason, the present invention, which observes the in-plane incident direction dependence of the polarization state of reflected light, can evaluate the molecular orientation of a liquid crystal orientation film on a glass substrate having distortion.

In the present invention, the dependence of the intensity of the reflected light on the in-plane incident direction of the polarization state is measured, and therefore the in-plane incident direction dependency of the polarization state is measured. The optical anisotropy corresponding to the molecular orientation of the film can be measured without being affected by the change in the reflected light intensity.

In order to measure the in-plane incident direction dependence of the S-polarized component and the P-polarized component, light from a light source is made to pass through a spectroscope and a polarizer to produce light having a constant polarization state and wavelength. The polarization state of the reflected light (phase difference and amplitude ratio of S wave and P wave) generated when a plurality of in-plane directions of this light are made incident on the same point on the surface of the liquid crystal alignment film sample is determined by the respective in-plane incident directions. Is measured using an analyzer. The measurement at a plurality of in-plane incident angles is performed by rotating the sample in-plane about an axis passing through the point on the sample surface where the light is incident, or simultaneously applying a plurality of light sources to the sample surface. When the incident light has a polarization state in which the amplitude ratios of the S-polarized component and the P-polarized component are equal, the analysis becomes a little easier.

The liquid crystal alignment film formed by the rubbing treatment is considered to be a film having two layers of a substance having an anisotropic outermost surface phase and a non-alignment phase below it. When light is reflected on the film surface of such a structure, the polarization state of the reflected light is the main dielectric constant of the anisotropic layer, the angle of the main coordinate system with respect to the film surface, the thickness, and the refractive index and the thickness of the non-oriented portion. And the refractive index of the substrate. The polarization state of this reflected light is determined by the 4 × 4 matrix method (DW Berrman Journal of the Optical Soc
Society of America) Vol. 62, No. 4, p. 502, 1972).

The normal vector of the sample surface is the Z-axis, the direction in which light travels is parallel to both the sample surface and the incident surface, and X is the positive direction.
A coordinate system in which the Y axis is perpendicular to the Z axis and the X axis is defined. The angular frequency of light is ω, and the electric field vector of light is (Ex, Ey, E
z), the magnetic field vector (Hx, Hy, Hz) is Ψ = t (Ex, Hy, Ey, -Hz) (2 ') Introducing a column vector of the following expression (1) 3 ') is satisfied.

[0027]

[Equation 1] In the formula (3 ′), M is a matrix with 4 rows and 4 columns,
It reflects the optical properties of the medium through which the light propagates.

When M is constant in the range of Z = Z0 to Z0 + d, there is a relation of Ψ (Z0 + d) = exp (iωdM) Ψ (Z0) (4 ').

Z = 0 on the surface of the alignment film, light is incident from the Z <0 side at an incident angle φ0 and is reflected on the Z <0 side, and the thickness of the anisotropic layer on the surface is d1. When the thickness is d2, the light state Ψ (d1 + at the interface between the alignment film and the glass substrate)
d2) is a matrix M1 corresponding to the anisotropic layer and a matrix M of the isotropic layer
By using 2, Ψ (d1 + d2) = exp (iωd2 M2) exp (iωd1 M1) Ψ (0) (5 ').

Here, Ψ (0) and Ψ (d1 + d2) are the S polarization component Eis of the incident light and the E polarization of the P polarization component, the S polarization component Ers of the reflected light, the P polarization component Eps, and the S polarization of the transmitted light. When expressed by the component Ets and the P-polarized component Etp, the refractive index of the atmosphere is 1,
Letting the refractive index of the glass substrate be N2, Ψ (0) = ((Eip−Erp) cosφ0, Eip + Erp, Eis + Ers, (Eip−Erp) cosφ0) (6 ′) Ψ (d1 + d2) = (Etpcosφ2, N2Etp, Ets , N2 Erpcos φ2) (7 '). Here, φ2 satisfies: sinφ0 = N2 sinφ2 (8 ').

Substituting equations (6 ') and (7') into equation (5 ') and erasing Etp and Ets of Ψ (d1 + d2), t (Erp, Ers) = Rt (Eip, Eis) A matrix R of 2 rows and 2 columns (9 ′) is obtained. S component of incident light and P
If the magnitudes of the components are equal, the i-row, j-column component of the matrix R is
In terms of Ri, j, the reflected light is Erp = R1,1 + R1,2 (10 ') Ers = R2,1 + R2,2 (11'), and the phase difference Δ and amplitude ratio tan ψ between S-polarized light and P-polarized light are , Tan ψ · exp (iΔ) = (R1,1 + R1,2) / (R2,1 + R2,2) (13 ′).

The above was calculated for the case where the liquid crystal alignment film was formed directly on the glass substrate, but when there is a transparent electrode film or a protective film between the liquid crystal alignment film and the glass substrate (5 ').
The formula is transformed as follows. The position of the surface of the glass substrate is Z = Zg, there are h layers other than the alignment film, and the matrix showing the respective optical characteristics is H1 and H from the surface toward the inside.
2 ,. . , Hh, and film thicknesses f1, f2 ,. . , Fh, a 4 × 4 matrix L of exp (iωd2 M2) exp (iωd1 M1) = L (14 ′) and a 4 × 4 column Kj of exp (iωfj Hj) = Kj (15 ′) Introducing, Ψ (Zg) = Kh.Kh-1 .. . .・ K2 ・ K1 ・ LΨ (0) (16 ')

Since the state of the reflected light from the liquid crystal alignment film can be calculated as described above, the principal permittivity of the anisotropic layer and the principal coordinate system of the anisotropic layer that reproduce the polarization state of the measured reflected light can be calculated. The optimum values of the angle and thickness with respect to the film surface, and the refractive index and thickness of the non-oriented portion are obtained. It is efficient to optimize this value using the nonlinear least squares method.

When the underlying structure and the refractive index of the liquid crystal alignment film are known, the main dielectric constant of the anisotropic layer, the angle of the main coordinate system with respect to the film surface, the thickness, and the refractive index and the thickness of the non-alignment portion are more significant. Accurately required, even if the underlying structure and refractive index are unknown, those values are treated as parameters to be optimized, and the main dielectric constant of the anisotropic layer, the angle with respect to the film surface of the main coordinate system, By obtaining the optimum values at the same time as the thickness, the refractive index and the thickness of the non-aligned portion, the state of the liquid crystal alignment film can be evaluated although the accuracy is poor.

[0035]

Embodiments of the present invention will be described below with reference to the drawings.

Embodiment 1 As an embodiment of the present invention, an apparatus using a light beam of a single wavelength as incident light will be described with reference to FIG. The light emitted from the incident light source 1 is ψ = π / 4, Δ by the polarizer 2.
= 0 (ψ and Δ are as defined in the equation (4)). The linearly polarized incident light is put in and taken out by the phase plate 3 upstream of the sample which is incident on the surface of the sample 6 at a constant incident angle from the normal to the surface of the sample in order to determine the phenotype of Δ. In this measurement, He-Ne laser 633
A 1/4 wavelength plate was used as a phase plate with a light of nm as a light source. If necessary, the measuring range of light on the surface can be limited by using the slit 4 and the lens 5. However, if light is collected using a lens, the blurring of the incident angle with respect to the sample becomes large, and the measurement accuracy is sacrificed. The light reflected by the sample 6 enters the analyzer 7, and the intensity of the light passing therethrough is measured by the light receiving tube 8. In this measurement, a photodiode was used as the light receiving tube.

The analyzer is rotated, the light intensity is measured by the angle of the analyzer, and the intensity of the light passing through the analyzer is determined from ψ and Δ from the Fourier sum for the analyzer angle. Since it is a periodic function, Δ has two values. Therefore, the measurement is performed with the phase plate inserted and the polarization state of the incident light is changed, and Δ, which is the same value as obtained without the phase plate, is adopted as the final value.

The sample 6 is placed on the rotary stage 9, and the anisotropy is measured by rotating the stage.

In this measurement example, the rotation analyzer method is used to determine the polarization state of the reflected light. However, in the analyzer and the polarizer in which the reflected light intensity detected by the same arrangement of the optical elements is zero. It is considered that the measurement can also be performed by the extinction point method (Otsuka The Japan Institute of Metals, Vol. 20, No. 7, page 614, 1981), which determines the polarization state from the angle.

The following three types of samples were measured using this apparatus.

(Sample A) A polyimide raw material liquid Hitachi Chemical LQ120 (trade name) was applied to the surface of 7059 glass manufactured by Corning Co., Ltd. using a spin coater, and then 250 ° C.
Calcination was performed by heating for 2 hours.

(Sample B) After being fired in the same manner as in Sample A, the indenting length of the cloth was 0.
Rubbing was performed at 4 mm, a rotation speed of 800 rpm, and a moving speed of 20 mm / s.

(Sample C) The rubbing conditions were the same as those of Sample B except that the roller rotation speed was 80 rpm.

FIG. 2 is a diagram showing the relationship between Δ of sample A, B and C and the sample rotation angle measured while rotating the rotary stage 9 with the incident angle of 70 °. The vertical axis represents Δ of 356.
The amount of change from 5 ° is shown.

As can be seen from FIG. 2, sample A without rubbing treatment has no molecular orientation, so that optical anisotropy is not observed, but sample B has an angle dependence and shows optical anisotropy due to molecular orientation. Anisotropy is especially observed for Δ.

Regarding the polarization state of the reflected light in FIG. 2, a significant difference is recognized between the normal direction incidence and the reverse direction incidence of rubbing. This difference is due to the fact that the molecules in the film have an orientation direction with respect to the film surface that is not parallel to the film surface, and the difference indicates the tilt angle of the molecule with respect to the surface.

From the measurement result of the sample B, the oriented partial film thickness of the polyimide molecule and the tilt angle of the molecule were determined as follows. In conducting the analysis of the film structure, it is assumed that the structure of the polyimide film of Sample B is composed of two parts, that is, a layer in the vicinity of the surface in which the polyimide molecules are oriented and a region in which the molecules in the depth are not oriented (random), The film thickness of each was determined. The portion where the molecules near the surface are oriented is treated as a uniaxial crystal with extremely small optical anisotropy, and the principal axis direction of the dielectric constant matrix is determined and used as the orientation direction of the polyimide molecules. The polarization state of the reflected light from the sample having such a structure is calculated as follows.

The propagation of electromagnetic waves including light is described by solving Maxwell's equation. The equation for monochromatic light of angular frequency ω is rotE = iωB rotH = iωD (5). In solving this, a vector G = (Ex, Ey, Ez, Hx, Hy, Hz) (6) and C = (Dx, Dy, Dz, Bx, which have the orthonormal coordinate components of the electric field E and the magnetic field H as components, By, Bz) (7) is introduced. If a 6-row 6-column tensor M having diagonally the permittivity tensor and the permeability tensor showing the properties of the medium through which light propagates is used, C = MG (8). On the other hand, the above equation (5) also uses the matrix P consisting only of the non-diagonal elements corresponding to the rotation operation on the left side, and PG = iωMG G = iω (P ⁻¹ M) G = iωLG L = (P ⁻¹ M) (9) In this equation, since the measured light state is reflected in G and P is known, determining the state of the polyimide film corresponds to solving the M component. This sample B
In a structure with different optical characteristics such as, L = L ₁ L ₂ L ₃ (10), and L ₁ , L ₂ and L ₃ correspond to the glass substrate, the non-aligned portion and the aligned portion of the polyimide film, respectively. . Of these, known values can be used for L ₁ and L ₂ . Since all tensors have almost no magnetic effect, the magnetic permeability was set to 1 isotropically. Specifically, line 4-6
Of the nine elements in columns ˜6, the diagonal elements were 1 and the others were 0. The coordinate system was defined so that the polyimide surface and the polyimide-glass interface were perpendicular to the Z axis, and the rubbing direction was the X axis.

First, in order to know the refractive index of the glass substrate, after the measurement of all the polyimide films was completed, the back surfaces of the samples A, B, and C were measured with a photodevice ellipsometer MAXY-102 to obtain the glass substrate itself. The refractive index of 1.5270 ± 0.001
I got the value. From this, tensor M1 lines 1-3
Of the nine elements in rows 1 to 3, the diagonal element was 2.331 and the other elements were 0.

Then, the polyimide film of Sample A was photo-coated.
When the result of measurement with a device Ellipsometer MAXY-102 was calculated as a refractive index of the glass substrate of 1.5270, a refractive index of 1.61 and a film thickness of 1068A were obtained. Note that this measurement was performed before the measurement of the back surface of the glass substrate. In the tensor M2, the diagonal element is set to 2.592 and the other elements are set to 0 out of the 9 elements in 1 to 3 rows and 1 to 3.

The tensor of the region where the molecules are oriented is expressed as follows. Let θ be the tilt angle of the molecule and ψ be the in-plane angle of the incident light with respect to the rubbing direction. The dielectric constant tensor of polyimide is displayed on the principal axis, and the 1st row and 1st column component u, the 2nd row and 2nd column component, and the 3rd row and 3rd column component are v, and the other non-diagonal elements are set to 0. These u and v will be obtained. The component Mij at the i-th row and the j-th column of the tensor M3 in the coordinate system in which the normal to the sample surface is the Z axis is as follows.

M11 = u cos ² θ cos ² φ + v cos θ sin ² φ−v sin ² θ cos φ M12 = −u cos ² θ sin φ cos φ−v sin φ cos φ M13 = −u sin θ cos θ cos ² φ + v sin θ sin ² θ + v sin θ cos θ cos φ M21 = u cos ² θ sinφ cosφ + v cosθ sinφ cosφ + v sin ² θ sinφ M22 = u cosθ sin ² φ + v cos ² φ M23 = −u sinθ cosθ sinφ cosφ + v sinθ sinφ cosφ + v sinθ cosθ sinφ M31 = u sinθ cosinθ cosθ cosθ sinφ cos θ M32 = u sin θ sin φ M33 = −u sin ² θ cos φ + v cos ² θ (11)

The operator P on the left side is essentially Z in this case.
It is a derivative with respect to the component, and obtaining the value of θ in the matrix element on the right side from the measured value corresponds to determining the molecular orientation. However, calculation is complicated and inefficient to obtain analytically, so the parameters are set to constant values and the state of the reflected light is calculated according to this formula. I searched for the value. It is considered more efficient to perform this parameter optimization process by the least squares method, but this time it was done by trial and error.

0th approximation When the tilt angle is 0, M11 = u cos ² φ + v sin ² φ M12 = u sinφ cosφ−v sinφ cosφ M13 = 0 M21 = −u sinφ cosφ + v sinφ cosφ M22 = u sin ² φ + v cos ² φ M23 = 0 M31 = 0 M32 = 0 M33 = v (12) Further, when φ = 0 in the rubbing direction, M11 = u M12 = 0 M13 = 0 M21 = 0 M22 = v M23 = 0 M31 = 0 M32 = 0 M33 = v (13) When φ = 90 ° M11 = v M12 = 0 M13 = 0 M21 = 0 M22 = u M23 = 0 M31 = 0 M32 = 0 M33 = v (14), which is a diagonal matrix and facilitates comparison with measured values.
However, since the approximation with the inclination angle θ = 0 has a two-fold symmetry angle dependence and does not represent an actual state, the average value of θ = 0 ° and φ = 180 °, and φ = 90 ° φ
Solve u, v using the average value of = 180 °. Then, in this case, the relationship between Ez and Dz and Hz and Bz becomes an identity, so that the matrix elements Mi3, M3j, and Mi related thereto are Mi.
6, M6j can be ignored. Then, when φ = 0 ° and 180 °, the tensor M has only diagonal elements, which is the same as the equation that holds for a sample of a uniform film whose outermost layer has a refractive index (u) ^1/2 . Similarly, in the case of φ = 90 ° and 270 °, it is the same as when the film having the refractive index (u) ^1/2 is on the outermost surface. That is, the 0th approximation value of u and v is φ = 0 °, 180
Derived from the measured values of °, φ = 90 °, 270 °. The film thickness of the outermost layer (alignment portion) and the film thickness of the non-oriented layer thereunder are unknown parameters, but since the thickness of the entire polyimide film is determined to be about 1070A from the measurement of Sample A. Assuming that the thickness of the oriented portion is d and the thickness of the unoriented portion below is 1070-d, d, u, and v can be uniquely determined. In the case of sample B, u = (1.635) ² = 2.673 v = (1.534) ² = 2.353 d = 150 A (15) θ = 0 °, u = (1.635) ² = 2.67
3, v = (1.534) ² = 2.353, d = 150A
The change in Δ in the case of is shown by a circle in FIG.

First approximation With u and v as initial values, θ,
The values of u, v, and d were optimized. Φ =
At 0 ° M11 = u cos ² θ−v sin ² θ M12 = 0 M13 = 0 M21 = 0 M22 = v M23 = 0 M31 = 0 M32 = 0 M33 = v (16) At φ = 180 ° M11 = u cos ² θ + v sin ² θ M12 = 0 M13 = 0 M21 = 0 M22 = v M23 = 0 M31 = 0 M32 = 0 M33 = v (17) The apparent difference in refractive index between 0 ° and 180 ° develops with θ as a minute amount, and approximately becomes vθ ² . Solving by approximation considering up to the second order of θ, θ to 6 ° is obtained.

Second approximation u = (1.635) ² = 2.673 v = (1.53
4) ² = 2.353 d = 150 A With θ = 6 ° as an initial value, the values of u, v, d and θ are optimized so as to be close to the values at all measurement points. By this trial and error, this process finally obtained u = (1.648) ² = 2.716 v = (1.501) ² = 2.253 d = 120 A θ = 7 ° (18). Calculated values obtained from these values are indicated by ⋄ in FIG.

In the same manner as for sample C, u = 2.658 v = 2.440 d = 120 A θ = 7 ° (19) was obtained.

(Comparative Example) FIG. 4 shows an example of measurement of the birefringence phase difference between Sample A and Sample B shown for comparison with the prior art.
Also in this measurement, the light source is a He-Ne laser 633.
nm light was used. The amount of change in the polarization state of the birefringence retardation measured by transmission shown in FIG. 4 is ± 0.15 ° at the maximum for sample B.
However, in the case of the reflected light shown in FIG. 2, the maximum is ± 0.
It reaches 9 °, which is more sensitive. In FIG. 4, an angle of 300 ° is an arrangement in which light is incident in parallel to the rubbing direction, and 120 ° corresponds to a case of incident in the opposite direction. Further, although there is no significant difference in the birefringence phase difference by the transmission measurement shown in FIG. 4 corresponding to the positive or negative in the rubbing direction, the polarization state of the reflected light in FIG. There is a significant difference. This difference is due to the fact that the molecules in the film are not oriented parallel to the film surface, and the difference indicates the tilt angle of the molecule with respect to the surface.

Embodiment 2 The influence of substrate strain on measurement was investigated.

(Sample D) Hitachi Chemical LQ12 of polyimide raw material liquid on the surface of soda lime glass which seems to have strain
After applying 0 (trade name) using a spin coater,
Firing was performed by heating at 250 ° for 2 hours.

(Sample E) After firing in the same manner, a buff cloth roller having a radius of 40 mm was used to press the cloth in a length of 0.4 mm.
Rubbing was performed at a rotation speed of 800 rpm and a moving speed of 20 mm / s.

(Comparative Example) FIG. 5 shows the results of birefringence phase difference measurement by transmission of Samples D and E. The difference between the rubbing sample E (●) and the non-rubbing sample D (⋄) is extremely small, which indicates that it is difficult to measure the molecular orientation of the polyimide due to the strain of the substrate.

The incident direction dependence of the polarization state of the reflected light generated when light of 633 nm was incident on the surface of the same samples D and E at an incident angle of 70 ° was measured. FIG. 6 shows the measurement result of the phase difference of the reflected light. In the sample D not rubbed, the phase difference of the reflected light has no anisotropy (◯ in FIG. 6), while
In the rubbed sample, anisotropy due to the molecular orientation as observed in Samples B and C of Example 1 was observed. In this way, the phase difference measurement of reflected light is not affected by the distortion of the glass substrate.

Embodiment 3 As an embodiment of the present invention, an apparatus using two light beams intersecting on a sample will be described with reference to FIG. FIG. 7 shows the arrangement of the device as seen from the normal direction of the sample, and the traveling directions of the two light beams are orthogonal to each other. The two lights emitted from the incident light source 1 are linearly polarized by ψ = π / 4 Δ = 0 by the polarizer 2. The phase plate 3 placed upstream of the sample is taken in and out to determine the delta phenotype. In this measurement, a mercury lamp was used as a light source, and monochromatization was performed by monochromators 36 and 46 in which a diffraction grating and a slit were combined. A quarter wave plate was used as the phase plate. The light measures the measuring range on the surface,
The slit 4 and the lens 5 were used to limit the distance, and the portions where the two light rays hit the sample were matched. The light reflected by the sample 6 enters the analyzer 7, and the intensity of the light passing therethrough is measured by the light receiving tube 8. In this measurement, a photodiode was used as the light receiving tube. With this arrangement, the polarization state can be measured by either the rotation analyzer method or the extinction point method. The sample 6 is placed on a parallel-moving XY stage 39 placed on a rotary stage 9. Using this stage, the film thickness of the oriented portion and the in-plane distribution of the refractive index in an arbitrary direction are measured.

Using this device, a Corning 705
9 Hitachi Kasei LQ120 was applied to the surface of glass using a spin coater and baked by heating at 250 ° C. for 2 hours, and then a buff cloth roller with a radius of 40 mm was used to push the cloth in a length of 4 mm, the rotation speed was 800 rpm, and moved. Speed 10 mm /
The sample subjected to rubbing of s was measured. 40 for measuring wavelength
0nm and 633nm are selected and the incident angle is 50 °, 55 °,
We chose 60 ° and 70 °.

The directions of the two light rays are the X direction and the Y direction, respectively. X, Y at each wavelength λ and incident angle measured in an arrangement in which light rays in the X direction are parallel to the rubbing direction of the sample
Tables 1 and 2 below show Δ and ψ observed in each direction.
It writes in.

[0067]

[Table 1]

[0068]

[Table 2] This sample shows that the polarization state, particularly the anisotropy of Δ, is the largest when the incident angle is 60 °, and the measurement can be performed with high accuracy and high sensitivity.

Attempts were made to determine the film thickness and refractive index at each wavelength. From the result of the sample B of Example 1, it is considered that the region of about 120 A near the surface is oriented. Therefore, it was tried to divide into 4 layers and obtain the film thickness and the refractive index of each layer. Since it is considered that the oriented portion is about 120 A, in order to know the distribution of molecular orientation in the depth direction, the refractive index (fixed film thickness) at every 160 A near the surface and the film thickness and refractive index of the deeper portion I asked. In this analysis, there are only two incident directions, and there are four parameters regarding the refractive index, so the tilt angle was not obtained. Since the tilt angle obtained from Example 1 is about 7 °, the influence on the refractive index is small (since sin θ is about 0.1, about 10%). Therefore, each layer is treated as an isotropic film having a different refractive index. It was First layer, second layer in order from the surface
As the layers, the third layer, the fourth layer, and the fifth layer (bulk portion), the refractive indexes of the respective layers are N1, N2, N3, N4, N5 =
As 1.61, the element in the i-th row and the j-th column of the k-th layer tensor M (k) is [M (k)] 11 = [M (k)] 22 = Nk ² (k = 1, 5) [M (K)] i, j = 0 (i ≠) (20) and L = L (glass) L (5) L (4) L
(3) L (2) L (1) is obtained. Nk was decided by trial and error so as to match the measured value. The initial value of Nk is 12 at the outermost surface.
0A uniform film, 1050A underneath, refractive index 1.61
The uniform film and the refractive index of the substrate were 1.527, and the refractive index of the outermost surface was determined. Since the approximation of this model is low, there is a difference due to the difference in the incident angle. X-axis direction 1.71 ± 0.6 Y-axis direction 1.5
It became 4 ± 0.5. As trial and error optimize the values of each layer,
First, an appropriate value of the refractive index of the fourth layer was obtained, and the outermost surface had a larger value in the X direction and a smaller value in the Y axis direction than the initial value. Also, the refractive index of the deeper layer was made closer to the bulk value than the refractive index of the layers above it. The results are shown in Tables 3 and 4 below.

[0070]

[Table 3]

[0071]

[Table 4] Thus, at 120A near the surface, the molecular orientation gradually changes from the surface to the inside of the film.
It can be seen from the change in refractive index that the molecular orientation is discontinuously lost near the depth of.

As shown in Tables 5 and 6 below, the measurement results of Δ and ψ when the sample is rotated by 45 ° from this state using the rotary stage are as follows.

[0073]

[Table 5]

[0074]

[Table 6] Therefore, there is almost no difference between the X-axis direction and the Y-axis direction. The analysis result also shows that the refractive index of each layer is 1.6 at a wavelength of 633 nm.
The range is 15 to 1.621, and at 400 nm, 1.
It became 630 to 1.633.

Embodiment 4 As an example of the present invention, an apparatus using two light beams intersecting on a sample will be described with reference to FIG. FIG. 8 shows the arrangement of the device viewed from the normal direction of the sample, and the traveling directions of the two light rays are orthogonal to each other. The two lights emitted from the incident light source 1 are linearly polarized by ψ = π / 4 Δ = 0 by the polarizer 2. The phase plate 3 placed upstream of the sample is taken in and out to determine the delta phenotype. In this measurement, He-Ne
Laser light of 633 nm is used as a light source, and 1 /
Four wavelengths were used. The light limits the measuring range on the surface by using the slit 4 and the lens 5, and the two light beams are made to coincide with each other at the portion where they hit the sample. However, if light is collected using a lens, the blurring of the incident angle with respect to the sample becomes large, and the measurement accuracy is sacrificed. The light reflected by the sample 6 enters the analyzer 7, and the intensity of the light passing therethrough is measured by the light receiving tube 8. In this measurement, a photodiode was used as the light receiving tube. With this arrangement, the polarization state can be measured by either the rotation analyzer method or the extinction point method. The sample is placed on the XY stage 19 which moves in parallel. Using this stage, the in-plane distribution of the film thickness and the refractive index of the oriented portion is measured.

Using this apparatus, samples F and G produced under the following conditions were measured.

(Sample F) NISSAN CHEMICAL SE7311, which is a polyimide raw material, was applied to the surface of 7059 glass manufactured by Corning Co., Ltd. using a spin coater, and baked at 250 ° C. for 3 hours.

(Sample G) After firing in the same manner, rubbing was performed with a buff cloth roller having a radius of 40 mm at a pressing length of the cloth of 4 mm, a rotation speed of 600 rpm, and a moving speed of 40 mm / s.

In the measurement, a square area of 100 mm × 100 mm was measured at intervals of 20 mm in each length and width. The measurement results of Samples F and G are shown in Table 7, Table 8, Table 9, and Table 10 below. The values of the measurement position XY are coordinates that the device itself has. The sample G was arranged so that the forward direction of rubbing coincided with the traveling direction of one light. Table 7 is a measurement result when a light ray is incident in the X direction of the sample F, Table 8 is a measurement result when a light ray is incident in the Y direction of the sample F, and Table 9 is a light ray in the X direction of the sample G. Table 10 shows the measurement results when the light rays were incident on the sample G, and Table 10 shows the measurement results when the light rays were incident on the sample G in the Y direction.

[0080]

[Table 7]

[0081]

[Table 8]

[0082]

[Table 9]

[0083]

[Table 10] As shown by the measurement results (Tables 7, 8, 9, and 10), the anisotropy of the refractive index of the sample G subjected to the rubbing treatment is almost the same at any point, and the molecular orientation state is almost uniform in the measurement range. I understand. On the other hand, there is a difference of about 1 nm in the film thickness between the samples F and G due to the difference in the incident direction, which means that the measurement accuracy regarding the film thickness is about 1 nm in this device. Since this measurement has many measurement points and the detailed analysis performed in the first embodiment takes a lot of time, it is limited to a simple analysis. For viewing the in-plane distribution of the alignment state of the alignment film, for the following reason, the sample structure is treated as an isotropic multilayer film having different refractive indices in each incident direction, as in Embodiment 3. Analysis seems to be sufficient. First, the tilt angle of the molecule is the result of the first embodiment and the result of infrared absorption of the same polyimide (SE7311 (trade name) manufactured by Nissan Kagaku) (Sawa et al. Japanese Journal of Applied Physics Japanese Journal).
l of Applied Physics Volume 33 6
It is expected to be as small as about 10 ° from page 273 (1994), and the difference with the tensor component when the oriented portion is considered to be an isotropic material is less than 20% at the maximum. Therefore, the refractive index of the substrate was set to 1.527 as a model of the sample structure of the sample F, and the film thickness and the refractive index of the polyimide were determined as an isotropic film.

The tensor M is a diagonal tensor showing a glass substrate (the values of M11 to M33 are all 2.332) and a diagonal tensor showing polyimide (the values of M11 to M33 are all the same value ε). = 2.567-2.588
I got Table 7 shows Δ, ψ film thickness, and refractive index of sample F. As the result shows, the refractive index of the polyimide SE7311 in the isotropic state was 1.604 ± 0.03, and the film thickness when manufactured under the conditions of this example was determined to be 750A. On the other hand, sample G is infrared absorption and (Sawa et al. Japanese Journal of Applied Physics Japan Jo
urnal of Applied Physics
Vol. 33, page 6273, 1994) Since the thickness of the oriented portion is expected to be about 150 A from the first embodiment,
As a structural model, a glass substrate having a refractive index of 1.527, a refractive index of 1.610 indicating an isotropic portion of polyimide, and a film thickness of 6
A film of 00A and a three-layer structure in which the refractive index and the film thickness are unknown are used. The orientation part was treated as an isotropic film with unknown refractive index and film thickness. The results are shown in Tables 9 and 10. As shown in Tables 9 and 10, there are systematic differences in refractive index and film thickness depending on the incident direction. Further, it is shown that in the same incident direction, a film having a uniform refractive index and film thickness at each point is produced. It should be noted that there is a systematic difference in not only the refractive index but also the film thickness of the alignment portion depending on the incident direction, which is considered to be due to the fact that the anisotropy of the alignment layer was not considered in the analysis.

Embodiment 5 A sample having a structure similar to that of a liquid crystal display element actually manufactured was prepared, and the incident angle dependence of the polarization state of reflected light was measured.

(Sample H) A silicon oxide film having a thickness of 400 nm was formed by sputtering on the surface of 7059 glass manufactured by Corning Co., Ltd. having a thickness of 1.1 mm, and IT having a thickness of 32 nm was formed.
An O (indium oxide / tin oxide) film was formed by vacuum vapor deposition. On top of this, the polyimide raw material Nissan Chemical SE731
1 was applied using a spin coater and baked by heating at 250 ° C. for 3 hours, and a buff cloth roller having a radius of 40 mm was used to press the cloth 0.4 mm in length, 600 rpm in rotation speed,
Rubbing treatment was performed at a moving speed of 40 mm / s.

(Sample I) A silicon oxide film having a thickness of 200 nm was formed by sputtering on the surface of 7059 glass manufactured by Corning Co., Ltd. having a thickness of 1.1 mm, and IT having a thickness of 32 nm was formed.
An O (indium oxide / tin oxide) film was formed by vacuum vapor deposition. On top of this, the polyimide raw material Nissan Chemical SE731
1 was applied using a spin coater and baked by heating at 250 ° C. for 3 hours, and a buff cloth roller having a radius of 40 mm was used to press the cloth 0.4 mm in length, 600 rpm in rotation speed,
Rubbing treatment was performed at a moving speed of 40 mm / s.

The measurement is performed by the device used in the third embodiment,
Light having a wavelength of 633 nm was used. Sample E has an incident angle of 70
The polarization state of the reflected light was measured at °, 65 °, 60 °, 55 °, and 50 °, and the sample F was measured at incident angles of 70 °, 55 °, and 40 °. Both samples were arranged so that the rubbing direction was parallel to the X axis. The measurement results are shown in Tables 11 and 12 below.

[0089]

[Table 11]

[0090]

[Table 12] As shown by the measurement results, it was revealed that in Sample H, the anisotropy of the polarization state of the reflected light was maximum near the incident angle of 60 ° as in Example 3. On the other hand, in Sample I, the anisotropy of the polarization state is maximized when the incident angle is around 40 °, reflecting the difference in the structure of the base of the alignment film. As a result of calculating the polarization state of the reflected light based on these measurement results, in the case of a liquid crystal display device that has a glass substrate, an insulating film, a transparent electrode film, and a polyimide alignment film as a part of the components, the incident angle with the highest sensitivity Was found to be between 35 ° and 65 °.

Sixth Embodiment As a sixth embodiment of the present invention, FIG. 9 showing an apparatus configuration as seen from the direction normal to the sample surface of an apparatus using a light beam of a single wavelength as incident light, and FIG. This will be described with reference to FIG. 10 showing the device configuration for one of the light rays of the book (in the X direction). The light emitted from the incident light source 1 in the X and Y directions is made into parallel rays having a wide diameter by the lens group 101. After that, by the polarizer 2, ψ = π / 4 Δ =
It is changed to 0 linearly polarized light. The phase plate 3 placed upstream of the sample is taken in and out to determine the delta phenotype. In this measurement, 633 nm light from a He-Ne laser was used as a light source, and a quarter wavelength plate was used as a phase plate. The slit 4 was used to shape the light beam incident on the sample so that the two light beams measured the same area on the sample surface. In this measurement, the incident angle of light on the sample 6 was selected to be 55 °, and the two light rays were shaped to have a length of 0.6 mm and a width of 1 mm so that a square of 1 × 1 mm was formed on the sample surface. The light reflected by the sample 6 enters the analyzer 7 after passing through the slit 4, and the light passing therethrough is magnified 10 times using the lens groups 102 and 103, and a total of 9 optical fibers in 3 rows and 3 columns are rectangular. Light receiving surface 10
Image at 4. The angle of the analyzer is controlled by the computer.

FIG. 11 shows the arrangement of optical fibers for detecting the reflected light in the X and Y directions. The light receiving parts of the optical fibers were arranged at intervals of 4.0 mm in the horizontal direction and 2.3 mm in the vertical direction for both the two reflected lights. That is, 51 and 52, 52 and 5 in the X direction
3, 54 and 55, 55 and 56, 57 and 58, 58 and 59
With a 4 mm spacing of 51 and 54, 54 and 57, 52 and 5
The intervals of 5, 55 and 58, 53 and 56, 56 and 59 are 2.
3 mm. The other optical fiber group 61 to 69
The same is true for The intensity of the light guided by the optical fiber was measured with a photodiode. The dependence of the reflected light intensity measured by the optical fiber on the analyzer angle is shown in FIGS.
The polarization states ψ and Δ of the reflected light at each position recorded in the computer 106 are obtained by calculating the Fourier sum for the analyzer angle of the light intensity. In the case of determining the quadrangle of Δ, the calculation is performed by using the result of the measurement in which the polarization state of the incident light is changed by inserting the phase plate. The optical anisotropy of the sample is reflected in the difference in the polarization state of the reflected light of two lights having different incident directions. 5 in the bidirectional optical fiber for this measurement
7 and 61, 54 and 62, 51 and 63, 58 and 64, 65
And 55, 52 and 66, 59 and 67, 56 and 68, and 53 and 65 correspond to each other, so that the same position on the sample surface is measured. In each of the above combinations, it is possible to know the molecular orientation state of the sample at each position by comparing the polarization states of reflected light. Rather than comparing polarization states
References (Marked Applied Optics (U. Ma
rkt Applied Optics Vol. 2
pp. 307 (1981)), the refractive index of the sample in each direction is calculated and the difference between the directions is compared, and it is considered that the amount indicating the molecular orientation state is more direct. For example, it is considered that the amount defined by | refractive index in X direction−refractive index in Y direction | / | refractive index in X direction + refractive index in Y direction | is the amount corresponding to the strength of molecular orientation.

The following samples were measured using this device.
Nissan Chemical's SE7311 (brand name) of polyimide raw material liquid
Was coated on the surface of 7059 glass manufactured by Corning Co., Ltd. using a spin coater and baked by heating at 250 ° C. for 2 hours, and then a resist was printed and coated on half of the sample surface. In this state, using a buff cloth roller with a radius of 40 mm, the pressing length of the cloth is 0.4 mm, the rotation speed is 200 rpm, and 20 mm / s.
After rubbing five times, the resist was peeled off.
In this sample, the result of measuring the vicinity of the boundary between the resist-coated portion and the exposed portion is shown in Table 13 below. The X-direction position and the Y-direction position below are the numbers in FIG. 11 of the optical fiber that captures the reflected light in each of the X and Y directions. In addition, the rubbing direction and the Y-axis direction were matched.

[0094]

[Table 13] When the refractive index of the polyimide film is calculated with the refractive index of the glass substrate being 1.52, as shown in Table 14 below,

[0095]

[Table 14] Then, the anisotropy of refractive index due to the molecular orientation caused by rubbing from 61 to 65 is observed. In this analysis, the polyimide film was treated as if it was uniformly oriented, but from the infrared absorption measurement (Sawa et al., Japanese Journal of Applied Physics Japans).
e Journal of Applied Physs
ics 33: 6273 (1994)) It has been revealed that only the surface is oriented. In consideration of this, the refractive index of the deep portion of the film is 1.604, and the result of another analysis conducted assuming that it is isotropic is shown in Table 15 below. The degree of orientation is the amount defined by the above formula.

[0096]

[Table 15] The image output device 107 expresses the in-plane distribution of the orientation degree when viewed from the Y direction by expressing the orientation degree with a black and white eight-step contrast.
It is shown in FIG.

As described above, in this embodiment, the in-plane distribution of the molecular orientation generated by rubbing can be observed at the same time. In this embodiment, an array of optical fibers is used to detect the reflected light. However, a CCD image sensor which is an example of a position-sensitive two-dimensional detector, a high-sensitivity image pickup tube SIT tube (a vidicon camera manufactured by Hamamatsu Photonics) C27
By using 41), higher resolution measurement can be performed.

By measuring the refractive index anisotropy of the minute portion according to this embodiment, the in-plane distribution of the molecular orientation in the minute region of the liquid crystal alignment film can be observed,
It can be visualized and observed.

Embodiment 7 Referring again to the apparatus of FIG. 1 which uses a light beam of a single wavelength as incident light, the light emitted from the incident light source 1 is caused by the polarizer 2 to have amplitudes of S-polarized component and P-polarized component. Are equal to each other and the phase difference between them is 0. This incident light is incident on the surface of the sample 6 at a constant incident angle from the normal to the surface of the sample. The phase plate 3 placed between the sample and the light source is moved in and out to determine the Δ of the reflected light. In this measurement, 633 nm light from a He-Ne laser was used as a light source, and a quarter was used as a phase plate.
A wave plate was used. By passing through this phase plate, the incident light changes from linearly polarized light to circularly polarized light. If necessary, the measurement range of incident light on the surface can be limited by using the slit 4 and the lens 5. However, when light is condensed using a lens, the blur of the incident angle with respect to the sample becomes large, and the measurement accuracy of the polarization state is sacrificed. The light reflected by the sample 6 is the analyzer 7
The intensity of the light entering and passing through it is measured by the light receiving tube 8. A photomultiplier tube was used in this measurement.

The analyzer was rotated to measure the intensity of the light transmitted through the analyzer for each angle of the analyzer, and was defined by the equation (13 ') from the Fourier sum for the analyzer angle of the measured intensity. Obtain ψ and Δ of the reflected light. In order to determine the quadrature of Δ, the measured Δ when linearly polarized light is input by operating the retardation plate and the Δ obtained when circularly polarized light is incident are equal to the final Δ. And

The sample 6 is placed on the rotary stage 9, and the in-plane anisotropy is measured by rotating the stage.

In this measurement example, the rotation analyzer method was used to determine the polarization state of the reflected light. However, from the angles of the polarizer and the analyzer such that the light passing through the analyzer becomes zero. It is considered that the extinction point method for determining the polarization state (Otsuka The Japan Institute of Metals, Vol. 20, No. 7, pp. 614, 1981) can also be used for measurement.

The following samples were measured using this apparatus.

(Sample A) A polyimide raw material solution, Hitachi Chemical LQ120 (trade name), was applied to the surface of 1.1 mm thick soda lime glass using a spin coater, and then 250
Firing was performed by heating at 0 ° C. for 2 hours. Then radius 4
Rubbing was performed with a 0 mm buff cloth roller at a cloth pressing length of 0.4 mm, a roller rotation speed of 300 rpm, and a substrate moving speed of 20 mm / s.

FIG. 13 shows the value of Δ with respect to the sample rotation angle obtained by measuring the sample A at 15 ° intervals while rotating the sample stage at the incident angle of 70 °. The state where the light is incident on the sample in parallel with the rubbing direction is defined as a sample angle of 0 °.

FIG. 14 shows the value of ψ with respect to the sample rotation angle obtained by measuring the sample A at 15 ° intervals while rotating the sample stage at the incident angle of 70 °. The state where the light is incident on the sample in parallel with the rubbing direction is defined as a sample angle of 0 °.

The principal dielectric constant of the anisotropic layer of the alignment film of this sample,
The angle of the main coordinate system with respect to the film surface, the thickness, and the refractive index and the thickness of the non-oriented portion were obtained according to the procedure of the formulas (1), (2 ′) to (13 ′). The analysis process is shown. (0th approximation) On the glass substrate, a phase difference of about ± 1 ° was observed in the transmitted light by measuring the birefringence of the transmitted light before film formation, but the difference in refractive index between ordinary light and extraordinary light was 3 × on average. The reflectance at the glass-polyimide interface is as small as 10 −6, and is almost the same as the reflectance at the glass-polyimide interface without distortion. That is, the distortion of the glass substrate can be ignored.

In order to obtain the thickness of the entire alignment film and the average dielectric constant, it is considered that there is a uniform film on the glass substrate.
Assuming that the average dielectric constant is <ε>, the components of the tensor M in the formula (3 ′) are as follows: M11 = 0 M12 = 1−η2 / <ε> M13 = 0 M14 = 0 M21 = <ε> M22 = 0 M23 = 0 M24 = 0 M31 = 0 M32 = 0 M33 = 0 M34 = 1 M41 = 0 M42 = 0 M43 = <ε> −η2 M44 = 0 (17 ′). Note that η = sin φ0 (18 ′).

As a result of the measurement, as shown in FIGS. 13 and 14, the minimum value of Δ is 354.950 °, the maximum value is 356.494 °,
The minimum value of ψ is 16.278 ° and the maximum value is 16.390 °.
Find a film thickness close to this value. (First approximation) Since many liquid crystal alignment films have a dielectric constant in the range of 1.8 to 3.4, the dielectric constant was assumed to be 2.6. The refractive index of the glass substrate was set to 1.525. Table 16 below shows the film thickness and Δ, ψ calculated under these conditions. Film thickness is 90-10
Since it is expected to be 0 nm, it is set to 95 nm.

[0110]

[Table 16] (Second approximation) Infrared absorption measurement of the thickness of the oriented portion near the surface of the total film thickness of 95 nm (Sawa et al. Japanese Journal of Applied Physics Japanese
Journal of Applied Physi
cs 33 vol. 6273 p. 1994) 15nm
Assume that Assuming that the oriented part is optically uniaxial, the dielectric constants εp and εn are related to the dielectric constant <ε> of the non-oriented part and <ε> = εp / 3 + 2εn / 3 (19 '). Suppose

In this case, the components of the matrix M reflecting the optical properties are εp −εn = Δε (where θ is the inclination angle of εp with respect to the sample surface and φ is the angle formed by the in-plane component of the incident direction of light. 20 ') εp ・ cos2θ + εn ・ sin2θ = ε3 (21') M11 =-(Δcos θsin θsin φ) η / ε3 M12 = 1-η2 / ε3 M13 = (Δcos θsin θcos φ) η / ε3 M21 = εn ・ (εp -Δεsin2θcos2φ) / ε3 M22 = −εn ・ (Δεsin2θcos φ) / ε3 M23 = −εn ・ Δεsin2θsin φcos φ / ε3 M31 = 0 M32 = 0 M33 = 0 M34 = 1 M41 = 0 M42 = 0 M43 = εn ・ (εp -Δεsin2θsin2φ) / ε3 M44 = 0 (22 ')

Table 17 below shows the range (minimum value, maximum value) of the value of Δ calculated from the values of εp and εn that satisfy the relationship of the equation (19 '). At this time, ε p of the oriented portion was assumed to be parallel to the surface (tilt angle 0 °). The closest to the measured results are εp = 2.66 and εn = 2.57.

[0113]

[Table 17] In the above calculation, since the tilt angle of the molecule is 0 °, Δ has twofold symmetry with respect to the in-plane incident direction of light. On the other hand, in the actual measurement, there is no 2-fold symmetry, and it can be seen that the direction (principal coordinate axis) of εp is inclined with respect to the plane. FIG. 15 shows the incident direction dependence of Δ calculated with the inclination of εp set to 0 °, 20 °, 40 °, and 60 °. The value of Δ largely depends on the inclination angle of εp with respect to the surface, but 20 ° is the closest to the measurement result, and therefore this is adopted. (Optimized) Glass refractive index of 1.525, non-oriented dielectric constant <
ε> = 2.6, thickness of non-oriented part 80nm, thickness of oriented part
15nm, dielectric constant of oriented part εp = 2.66, εn = 2.57
, With an inclination angle of 20 ° as an initial value, Δ obtained by measurement,
The non-linear least squares method is used to reproduce the refractive index of the glass substrate, the dielectric constant of the non-oriented portion <ε>, the thickness of the non-oriented portion, the thickness of the oriented portion, the permittivity of the oriented portion εp, and the inclination angle of εn to reproduce Optimized with. The least squares method is the Marcat method (Nakagawa, Koyanagi
"Experimental data analysis by least-squares method" University of Tokyo Press 19
1982) was adopted and the weight of each measurement data was assumed to be equal.

As a result of the optimization, the refractive index of glass was 1.527, the dielectric constant of the non-oriented portion <ε> = 2.64, and the thickness of the non-oriented portion was 80 nm.
, The thickness of the alignment portion is 13 nm, the dielectric constant of the alignment portion εp =
2.74, ε n = 2.59, and inclination angle 38 ° were obtained. 16 and 17 show measured values and calculated values by the parameters obtained by the optimization. The ψ of the measured data has a large fluctuation in the data and is considered to be due to the distortion of the glass substrate.

Embodiment 8 The molecular orientation was discussed from the angle dependence of ψ using unstrained glass.

(Sample B) A 1.1 mm-thick Corning 7059 glass surface was coated with Hitachi Chemical LQ120, a polyimide raw material liquid.
After applying (trade name) using a spin coater, 2
Firing was performed by heating at 50 ° C. for 2 hours. Then, with a buff cloth roller with a radius of 40 mm, press the cloth in 0.4 mm,
Roller rotation speed 300 rpm, substrate moving speed 20 mm /
The rubbing was performed at s.

(Sample C) The conditions are the same as those of sample B except that the rubbing process is not performed.

Samples B and C were measured by the same device as in the first embodiment. However, the in-plane incident angle was measured at every 10 °. FIG. 18 shows the measurement result of ψ. Sample C was isotropic, but sample B exhibited anisotropy due to molecular orientation. When the same analysis as in Embodiment 7 was performed on Sample B, the film thickness of the non-oriented portion was 82 nm, and the other parameters were the same as those of Sample A. In addition, sample C is a non-oriented layer
It became 94 nm. When these samples are measured under the above conditions, Δ is more sensitive than ψ.

Ninth Embodiment The ninth embodiment will be described with reference to the apparatus of FIG. 7 which uses two light beams intersecting on the sample surface. As described above, FIG. 7 shows the arrangement of the device viewed from the normal direction of the sample, and the traveling directions of the two light beams are orthogonal to each other. The two lights emitted from the incident light source 1 are linearly polarized by the polarizer 2 so that the S and P components have the same amplitude. The phase plate 3 placed at a position closer to the light source than the sample is put in and out for determining the phenotype of Δ. In this measurement, a mercury lamp was used as a light source and a quarter wave plate was used as a phase plate. Light is monochromator 3
9 and 46 are used to obtain a single color. Light of 630 nm was extracted for comparison with the seventh embodiment. The light limits the measurement range on the surface by using the slit 4 and the lens 5, and the two light beams are made to coincide with each other at the portion where they hit the sample. However, when the light is condensed using the lens, the blurring of the incident angle of the light on the sample becomes large and the measurement accuracy is sacrificed. The light reflected by the sample 6 enters the analyzer 7, and the intensity of the light passing therethrough is measured by the light receiving tube 8. In this measurement, a photodiode was used as the light receiving tube. With this arrangement, the polarization state of the reflected light can be measured by either the rotation analyzer method or the extinction point method. Sample can be translated X
It is placed on the Y stage 19. Using this apparatus, the samples B and C described in the eighth embodiment were measured at 25 points in a square area of 20 mm × 20 mm at 5 mm intervals in both the X and Y directions.

For the sample B, the in-plane distribution of the difference in the polarization state of the reflected light was compared between when the incident light was incident parallel to the rubbing direction and when it was incident perpendicularly. The results obtained when the incident light is incident perpendicularly to the rubbing direction and when the incident light is incident parallel to the rubbing direction are shown in Table 18 and Table 19 below. Table 18 and Table 19
As can be seen from the above, it was observed that the value of Δ was systematically larger when light was incident from the direction perpendicular to the rubbing direction, reflecting the optical specific anisotropy due to the molecular orientation.

[0121]

[Table 18]

[0122]

[Table 19] The same in-plane distribution measurement was also performed for sample C. The results obtained when the incident light is incident perpendicularly to the rubbing direction and when the incident light is incident parallel to the rubbing direction are shown in Table 20 and Table 21 below. Table 2
0 and as shown in Table 21, there was no significant difference between Δ and ψ.

[0123]

[Table 20]

[0124]

[Table 21] As described above, the presence or absence of molecular orientation due to rubbing can be observed from the difference in Δ obtained by simultaneous measurement from two directions.
Since the sample is not rotated, the time required for measurement is shorter than that of the device of the seventh embodiment, and the in-plane distribution measurement of the orientation of the sample can be efficiently performed.

Since this apparatus measures anisotropy in only two directions, the state of the orientation film of the sample whose molecular orientation is completely unknown (main dielectric constant of the orientation part, relative to the film surface of the main coordinate system) It is difficult to know the angle, thickness, and the refractive index and thickness of the non-oriented portion. When the state of the film is to be obtained in detail, it is necessary to rotate the sample as in the case of the seventh embodiment and further increase the directions of the light that is simultaneously incident.

However, if the structure is the same as that of the sample once measured in the seventh embodiment (the same liquid crystal alignment film material and the same substrate), the anisotropy of the film thickness and the dielectric constant can be estimated to some extent by the measurement from two directions. .

[0127]

As described above, according to the present invention, from the measurement of the anisotropy of the polarization state of obliquely incident light, it is possible to measure the degree of orientation of molecules and the portion that is oriented, which cannot be measured by the conventional method. Can be quantitatively measured. Further, according to the present invention, it is possible to obtain information on the angle of molecular orientation with respect to the film surface. Further, in the present invention, the distribution of molecular orientation in the depth direction can also be measured from the dependence on incident light and wavelength.

Further, according to the present invention, from the measurement of the in-plane incident direction dependence of the polarization state of obliquely incident light, the main dielectric constant of the oriented portion, which cannot be measured conventionally, which reflects the molecular orientation of the film, It is possible to quantitatively measure the angle of the coordinate system with respect to the film surface, the thickness, and the refractive index and the thickness of the non-oriented portion that does not cause molecular orientation even by the rubbing treatment. Further, according to the present invention, the state of the alignment film can be measured without being affected by the strain of the glass substrate, which cannot be avoided by the conventional method.

[Brief description of drawings]

FIG. 1 is a schematic perspective view of an inspection device of the present invention.

FIG. 2 is a diagram showing the relationship between Δ and the sample rotation angle of samples A, B, and C that were subjected to the inspection method of the present invention with an incident angle of 70 °.

FIG. 3 is a diagram showing measured values and calculated values of Δ of sample B.

FIG. 4 is a diagram showing a relationship between Δ and a sample rotation angle, which are obtained by performing a conventional inspection method on samples A, B, and C.

FIG. 5 is a diagram showing the relationship between Δ and the sample rotation angle obtained by performing a conventional inspection method on samples D and E.

FIG. 6 is a diagram showing the relationship between Δ and the sample rotation angle for samples D and E that have been subjected to the inspection method of the present invention.

FIG. 7 is a schematic plan view of the inspection device of the present invention.

FIG. 8 is a schematic plan view of the inspection device of the present invention.

FIG. 9 is a schematic plan view of the inspection device of the present invention.

FIG. 10 is a schematic perspective view of the inspection device of the present invention.

FIG. 11 is a layout view of an optical fiber of the present invention.

FIG. 12 is a diagram showing an in-plane distribution of the orientation degree obtained by the inspection method of the present invention.

FIG. 13 is a diagram showing the relationship between Δ and the sample rotation angle of the sample A subjected to the inspection method of the present invention with the incident angle of 70 °.

FIG. 14 is a diagram showing the relationship between ψ and the sample rotation angle of the sample A subjected to the inspection method of the present invention with an incident angle of 70 °.

[Fig. 15] Non-oriented portion film thickness 80 nm, dielectric constant 2.6, oriented portion film thickness
15nm, dielectric constant 2.66, 2.57, refractive index of glass substrate 1.525, tilt angle with respect to the film surface of the main coordinate system is 0 °, 20 °,
It is a figure which shows the sample rotation angle dependence of (DELTA) calculated as 40 degrees and 60 degrees.

16 is a diagram showing the sample angle dependence of Δ calculated from the measured values of Sample A and the optimized parameters. FIG.

17 is a diagram showing the sample angle dependence of ψ calculated from the measured value of sample A and the optimized parameters. FIG.

FIG. 18 is a diagram showing sample angle dependence of ψ of reflected light of samples B and C.

[Explanation of reference numerals] 1 light source 2 polarizer 3 phase plate 4 slit 5 lens 6 sample 7 analyzer 8 light receiving tube 9 rotating stage 19, 39 XY stage 36, 46 monochromator 101 lens group 104 optical fiber array 105 photodetector 106 computer 107 image output device

Claims (14)

[Claims] 1. The polarization state of the reflected light is obtained by measuring the polarization state of the reflected light generated when a monochromatic light beam having a predetermined polarization state is incident on the surface of the thin film sample at a predetermined angle, in a plurality of directions within the sample plane. Anisotropic thin film inspection method characterized by determining the molecular orientation of a sample from the anisotropy of the sample. 2. A method for measuring the polarization state of reflected light generated when a monochromatic light beam having a predetermined polarization state is incident on a surface of a thin film sample at a predetermined angle in a plurality of directions within a sample plane. An anisotropic thin film inspection method characterized in that the molecular orientation of a sample and the molecular orientation distribution in the sample are determined from the dependence of the wavelength of light and the direction of incidence on the surface of the sample. 3. A monochromatic light beam having a predetermined polarization state is incident on a predetermined area portion, reflected light measured in a plurality of directions within the sample plane is enlarged, and the difference in polarization state of the reflected light causes The anisotropic thin film inspection method according to claim 1, wherein the in-plane distribution of the molecular orientation in the predetermined area portion is measured. 4. The anisotropic thin film inspection method according to claim 1, wherein the incident angle is 35 ° or more and 65 ° or less. 5. A sample stage on which a membrane sample is placed and which can be rotated in-plane, and means for injecting a monochromatic light beam of a predetermined polarization state at a predetermined angle on the sample surface on the sample stage. And a means for measuring the polarization state of the reflected light from the sample, and determining the molecular orientation of the sample from the anisotropy of the polarization state of the reflected light. 6. A sample stage on which a membrane film sample is placed, and a plurality of means for injecting a monochromatic light beam of a predetermined polarization state at a predetermined angle to one point of the sample surface on the sample stage are provided. Means for measuring the respective polarization states of the reflected light, and determining the molecular orientation of the sample from the difference in the polarization states of the plurality of reflected lights. 7. The means for injecting the monochromatic light beam having a predetermined polarization state at a predetermined angle can change the wavelength and the incident angle of the incident light, and the refractive index of the sample can be determined from the incident angle dependence of the polarization state of the reflected light. Obtain the distribution for each incident direction in the sample plane,
The anisotropic thin film inspection apparatus according to claim 5 or 6, wherein the molecular orientation of the sample and the molecular orientation distribution in the sample are determined. 8. A sample stage on which a membrane film sample is placed, and a plurality of means for injecting monochromatic light beams having a predetermined polarization state at the same position on the sample surface on the sample stage so as to intersect each other at a predetermined angle. A means for enlarging a plurality of reflected light from the sample, a means for measuring the in-plane distribution of the polarization state of the reflected light, and a molecule of the sample from the difference of the corresponding encounter components for each of the plurality of reflected light An anisotropic thin film inspection apparatus characterized by determining an in-plane distribution of orientation. 9. The anisotropic thin film inspection apparatus according to claim 8, wherein the means for obtaining the in-plane distribution of the polarization state of the reflected light is a position-sensitive two-dimensional detector. 10. The polarization state of reflected light generated when a monochromatic light beam having a predetermined polarization state is incident on the surface of the liquid crystal alignment film at a predetermined angle is measured from a plurality of directions in the same region on the sample surface. A method for inspecting a liquid crystal alignment film, which comprises measuring the anisotropy of the phase difference between the S-polarized component of the reflected light (polarized component parallel to the sample surface) and the P-polarized component perpendicular thereto. 11. The polarization state of reflected light generated when a monochromatic light beam having a predetermined polarization state is incident on the surface of the liquid crystal alignment film at a predetermined angle is measured from a plurality of directions in the same region on the sample surface. A method for inspecting a liquid crystal alignment film, which comprises measuring the anisotropy of the intensity ratio of the S-polarized component of the reflected light (polarized component parallel to the sample surface) and the P-polarized component perpendicular to it. 12. The polarization state of reflected light produced when a monochromatic light beam having a predetermined polarization state is incident on the surface of a liquid crystal alignment film at a predetermined angle is measured from a plurality of directions in the same region on the sample surface. , Anisotropy of the thickness of the molecular alignment layer in the liquid crystal alignment film due to the anisotropy of the phase difference and the intensity ratio of the S-polarized component of the reflected light (polarized component parallel to the sample surface) and the P-polarized component perpendicular to it. Analytical method characterized by obtaining the main dielectric constant, the thickness of the non-aligned portion in the liquid crystal alignment film, and the dielectric constant. 13. A sample stage on which a liquid crystal alignment film sample is placed and which can be rotated in-plane, and a monochromatic light beam having a predetermined polarization state is incident on the sample surface on the sample stage at a predetermined angle. And means for measuring the phase difference and the intensity ratio of the S-polarized component and the P-polarized component of the reflected light from the sample,
From the anisotropy of the phase difference and the intensity ratio of the S-polarized component and the P-polarized component of the reflected light, the thickness of the molecular alignment layer in the liquid crystal alignment film of the sample,
An apparatus for inspecting a liquid crystal alignment film, which determines an anisotropic main dielectric constant and a thickness and a dielectric constant of a non-aligned portion in the liquid crystal alignment film. 14. A sample table on which a liquid crystal alignment film sample is placed,
A plurality of means for injecting a monochromatic light beam having a predetermined polarization state at a predetermined angle is provided at one point on the surface of the sample on the sample table, and the phase difference and the intensity ratio of the S-polarized component and the P-polarized component of the reflected light from the sample are determined. The thickness of the molecular alignment layer in the liquid crystal alignment film of the sample and the anisotropic main dielectric constant are determined from the anisotropy of the phase difference and the intensity ratio of the S-polarized component and the P-polarized component of the reflected light. An apparatus for inspecting a liquid crystal alignment film, characterized by determining the thickness and the dielectric constant of a non-aligned portion in the liquid crystal alignment film.