Polarization mask for photo-alignment exposure, photo-alignment exposure method, and method for manufacturing optical component or the like

Abstract

[Problem] To provide a manufacturing technology that: is capable of aligning liquid crystal molecules in different directions at different locations; has a high spatial degree of freedom for alignment; and has a high mass productivity. [Solution] Using a waveplate 1407 that contains photonic crystals, a metasurface, or a sub-wavelength structure, the polarization mask for photo-alignment exposure controls the polarization direction of transmitted light at each location in the x-y plane in three-dimensional x,y,z-coordinate space. Through the incidence of polarized light traveling in the z-direction on the polarization mask, a region is present that realizes a different polarization direction at each location in the x-y plane and the molecular orientation of a photoreactive polymer, which is coated on a substrate disposed in parallel to the polarization mask, is collectively controlled at each location.

Description

Polarizing mask for photo-alignment exposure, photo-alignment exposure method, and method for manufacturing optical components, etc.

[0001] The present invention relates to a polarizing mask for photo-alignment exposure, a photo-alignment exposure method using a polarizing mask, and a method for manufacturing an optical component or an optical device using the photo-alignment exposure method.

[0002] In recent years, optical devices that utilize polarized light have been actively developed. One example where this is often used is LCD televisions, which create images by controlling the polarization state of each pixel. More recently, new devices that control light in a more advanced way, such as VR (Virtual Reality) goggles and AR (Augmented Reality) goggles, have been proposed and sold. Miniaturization and weight reduction are major issues for these devices, and the key to their development is how to minimize the number of parts and achieve a thin optical system.

[0003] One solution to this problem is a technology that uses a phenomenon known as the geometric phase (Pancharatnam-Bery phase) to control the wavefront of light (see Non-Patent Document 1). Details are explained below. Assume a situation in which half-wave plates, which use birefringence to create a phase difference of half the wavelength between vertically polarized light and horizontally polarized light, are arranged side by side, with their orientations differing by an angle θ. When circularly polarized light is incident on these plates and passes through each half-wave plate, the phase of the transmitted light differs by 2θ. This phenomenon means that by controlling the orientation of the half-wave plates at different locations, it is possible to control the phase of light at different locations.

[0004] For example, as shown in Figure 1, prepare a wave plate with retardation of π+2(n-1)π (n = 1, 2, 3, ...) where the orientation changes by 180 degrees with a period P (101 in the figure). Circularly polarized light passing through this wave plate will have a phase change of 360 degrees at each location with a period P based on the principle of geometric phase. Therefore, light with wavelength λ is diffracted at an angle Φ shown in Equation 1.

[0005] In FIG. 1, the 180-degree azimuth change of the wave plate is represented by four divisions. In this case, the wave plate orientation 102 changes in increments of π / 4. As a result, the phase of the output light changes in increments of π / 2. Because phases of 0 and 2π are equivalent, the phase of each region in the lower part of FIG. 1 is always π / 2 ahead of the adjacent region to the left. By increasing the number of divisions, the phase change becomes smoother, ultimately resulting in a straight line with a slope of 2π over a distance P. As a result, as shown in FIG. 2, when left-handed circularly polarized light traveling in the +z direction is incident on the wave plate 201 patterned as shown in FIG. 1, each region is a half-wave plate, so the polarization state of the output light becomes right-handed circularly polarized and is diffracted at an angle 202 in the +x direction. Similarly, when right-handed circularly polarized light traveling in the +z direction is incident on the wave plate 201, the polarization state of the output light becomes left-handed circularly polarized and is diffracted at an angle 203 in the -x direction. In this way, it functions as a diffraction grating with polarization dependency.

[0006] When the number of divisions is very large and the orientation of the wave plate changes continuously, the diffraction efficiency reaches 100%. When the number of divisions is small, the efficiency does not reach 100%, and the remaining energy is not diffracted but rather exits in a straight line in the direction of the incident light. In diffraction gratings using conventionally shaped concave and convex structures, the steps of the concave and convex structures provide a phase change at each location. However, when the phase change is large, it is difficult to achieve a continuous change in the steps, and the steps are usually reset every 2π. However, the large steps result in unnecessary diffraction of the light, reducing the diffraction efficiency in the intended direction. However, with the type of diffraction grating shown in Figures 1 and 2, the phase change at each location is achieved by changing the orientation of the wave plate from 0 to π, making it possible to provide a continuous phase change from 0 to 2π, eliminating the need for resetting every 2π and achieving high diffraction efficiency.

[0007] For example, AR glasses require the eye to receive an image signal while viewing the surroundings with the naked eye. Proposed methods include using a half mirror to project the image onto the eye from a display, or scanning the retina with a laser. One method uses a light guide plate and a diffraction grating to send an image signal into the glass of the glasses, and then the light is extracted by a diffraction grating located just in front of the eyeball and transmitted to the eye. Here, the diffraction efficiency is a parameter that greatly influences the device's performance. In other words, higher diffraction efficiency means that the image that reaches the eye is brighter, the viewing angle is wider, and the brightness of the display, or in other words, there is no need to increase the power, resulting in significant benefits such as longer operating time when powered by a rechargeable battery.

[0008] Because the light that comes out of a display that uses liquid crystal is already polarized, it is easy to circularly polarize the light that enters diffraction grating 302 from display 301 with a configuration such as that shown in Figure 3. The light that enters diffraction grating 302 is bent, propagates while undergoing total reflection inside light guide plate 303, is bent in the -z direction by diffraction grating 304, and enters the naked eye 305. An image is formed on the retina by the lens of the naked eye. As mentioned above, the diffraction gratings described in Figures 1 and 2 can easily achieve high diffraction efficiency, and are therefore suitable optical elements for AR glasses.

[0009] In this way, it is possible to bend light using a plate-shaped element, and by appropriately selecting the change in orientation, it is possible to realize a curved wavefront instead of the linear wavefront shown in Figure 1, and it is easy to infer that a lens can also be realized. In other words, a thin plate-shaped lens can be realized. In this case, the axial orientation of the waveplate must be changed within a very small range of several hundred nanometers to several microns for visible light, as can be seen from Equation 1.

[0010] For example, consider a pattern in which regions are divided into annular shapes as shown in Figure 4, and the axis orientation 401 of birefringence changes between 0 degrees and 180 degrees as one moves from the center to the outside. The phase change when moving from the center to the outside is shown in the figure. As shown in Figure 4, the width of each region is not equal and changes gradually, so the envelope of the phase distribution does not fall immediately above it, but forms a curve. For example, with respect to the radius r from the center, ２(a is a constant) is a lens. In this case, as can be inferred from Figure 1, it functions as a lens that converges light for one circularly polarized light, and as a lens that diverges light for the other orthogonal circularly polarized light.

[0011] Until now, there have been plate-shaped lenses called Fresnel lenses, which have a sawtooth uneven surface. However, there is an upper limit to the amount of continuous phase change that can be achieved with the uneven surface, and a portion where the shape changes suddenly is required to shift the phase by 2π. As a result, the disturbance of the light wavefront there affects the image, and high-performance lenses have not been realized. However, with the above-mentioned geometric phase, a continuous phase can be achieved indefinitely by repeating the orientation of the wave plate from 0 degrees to 180 degrees, solving this problem and enabling the realization of high-performance flat lenses, which are expected to be used in the above-mentioned AR / VR devices.

[0012] To realize such an optical element, a wave plate, or optical component with birefringence, is required, and as mentioned above, it is necessary to be able to control its axial orientation at each location. One possible way to create such an optical component is to use liquid crystal materials. Liquid crystal materials are characterized by their birefringent molecules and their ability to exhibit liquid-like fluidity under certain conditions, and are already used in many display devices such as monitors. Necessary conditions for consumer products such as AR / VR devices include low cost, guaranteed stable supply, and ease of mass production, and liquid crystal materials meet these conditions.

[0013] To realize such optical elements using liquid crystals, it is necessary to align the liquid crystal molecules in the intended direction. Generally, to control the molecular orientation of liquid crystals, a film called an alignment film is applied to the substrate, and this is given directionality through rubbing or other processes. When liquid crystals are applied on top of this, the liquid crystal molecules line up either parallel or perpendicular to the direction, thereby aligning the liquid crystal molecules. However, this technology does not allow for the orientation of such small areas to be changed.

[0014] To address this issue, a technology called photo-alignment has been proposed. This is because when polarized light (mainly ultraviolet light) is irradiated onto an alignment film, the electric field component parallel to the molecular bonds is absorbed. In other words, polarized components parallel to the molecular bonds are more absorbed, promoting a reaction. By irradiating polarized light, it is possible to selectively react molecular bonds in that direction. As a result, this technology generates anisotropy in the alignment film. By using this method and controlling the polarization direction at each location, it is possible to realize optical elements that utilize the above-mentioned geometric phase.

[0015] One way to achieve this is to control the polarization direction of a narrowly converged light beam while scanning it, which clearly allows for controlling the polarization at each location and achieving any desired optical alignment pattern. However, although this method offers a high degree of freedom in the light polarization pattern, because it relies on beam manipulation, there are significant issues with throughput, even when multiple beams are used.

[0016] Another method is to generate an orientation pattern using two-beam interference (Non-Patent Document 3). For example, as shown in Figure 5, incident light is split by a polarizing beam splitter 501 into polarized light 502 parallel to the plane of the paper and polarized light 503 perpendicular to the plane of the paper, and each light passes through quarter-wave plates 504 and 505 with their axes oriented at 45 degrees to the polarization direction, and is reflected by mirrors 506 and 507, respectively, and irradiated onto a substrate 508.

[0017] When right-handed and left-handed circularly polarized light beams are incident on the substrate from two symmetrical oblique directions relative to the direction perpendicular to the substrate, the right-handed and left-handed circularly polarized light beams overlap to form linearly polarized light. The orientation of this linearly polarized light is determined by the phase difference between the right-handed and left-handed circularly polarized light beams. Circularly polarized light beams incident on the substrate at an angle have different phase differences at different locations. Because two circularly polarized beams with different phases overlap at different locations, their interference results in a state on the substrate where the polarization orientation changes periodically, as shown in the lower part of Figure 5. (The lower part of Figure 5 shows the amplitude of the electric field. In actual exposure, energy (proportional to the square of the electric field) is important, and the direction of the arrow becomes meaningless.) This period can be controlled by the wavelength and angle of incidence of the light. By increasing the beam diameter, a large-area photoalignment pattern can be achieved at once, making this a highly productive method. However, because it uses interference, while it is effective at achieving a pattern in which the orientation rotates 180 degrees with a period P (reference number 101) as shown in Figure 1, it is not effective at changing only part of the pattern. Therefore, its applicability to anything other than uniform periodic patterns is low, and its uses are limited. In addition, it is necessary to use a laser as the light source, and when attempting to irradiate a large area at once, there are many elements that make the equipment large, such as a beam expander and a reflective optical system, and it tends to be expensive for mass production manufacturing equipment.

[0018] In this way, liquid crystal molecules can be oriented in different directions at different locations, and a manufacturing technology that provides a high degree of spatial freedom in orientation and is suitable for mass production has not yet been proposed.

[0019] S. Pancharatnam, “Generalized theory of interference and its applications,” Proc. Indian Acad. Sci. A 44, 398 (1956).Nanfang Yu and Federico Capasso. 2014. “Flat optics with designer metasurfaces.” NATURE MATERIALS, 13, 2, Pp. 139-150.Fabrication of Polarization Grating on N-Benzylideneaniline Polymer Liquid Crystal and Control of Diffraction Beam, M. Kondo, K. Fujita, T. Sasaki, M. Sakamoto, H. Ono and N. Kawatuki, Crystals 12(2022)273.M. Hasegawa and Y. Taira, J. Photopolym. Sci. Technol. 8, 241(1995).

[0020] Japanese Patent No. 3288976 Japanese Patent Laid-Open No. 2001-51122 Japanese Patent No. 3325825 Japanese Patent No. 4975162 Japanese Patent No. 3766844

[0021] Therefore, the main objective of the present invention is to propose a manufacturing technology that can align liquid crystal molecules in different directions at different locations, has a high degree of spatial freedom in the alignment, and is highly suitable for mass production. Specifically, in order to realize a manufacturing method that allows for free optical alignment patterns and is highly suitable for mass production, the present invention aims to provide a technology and exposure apparatus for optical alignment that uses a polarization control mask, such as a photonic crystal, to realize a polarization state in which the polarization orientation is controlled at different locations.

[0022] Although the above description has focused on applications to optical elements using geometric phases in AR / VR devices, it is clear that the photoalignment technology of this patent is also effective as a technology for aligning liquid crystal molecules in a variety of applications.

[0023] A first aspect of the present invention relates to a polarizing mask for photo-alignment exposure. The polarizing mask according to the first embodiment of the present invention is configured with a polarizer including a photonic crystal or a wire grid. In such a polarizer-type polarizing mask, the polarization direction of light transmitted by the polarizer is controlled for each location within the x-y plane in a space of three-dimensional coordinates x, y, and z. The polarizing mask according to this embodiment has an area where non-polarized light traveling in the z direction is incident on the polarizing mask, thereby realizing different polarization directions for each location within the x-y plane. This allows for simultaneous control of the molecular orientation of a photoreactive polymer coated on a substrate placed parallel to the polarizing mask for each location.

[0024] A polarizing mask according to a second embodiment of the present invention is configured with a wave plate including a photonic crystal, a metasurface, or a subwavelength structure. In such a wave plate-type polarizing mask, the polarization orientation of light transmitted through the wave plate is controlled at each location within the x-y plane in the three-dimensional coordinate space of x, y, and z. The polarizing mask according to this embodiment has an area where light polarized in the z direction is incident on the polarizing mask, resulting in different polarization orientations at each location within the x-y plane. This allows for simultaneous control of the molecular orientation of a photoreactive polymer coated on a substrate placed parallel to the polarizing mask.

[0025] The polarizer type polarizing mask and the waveplate type polarizing mask described above preferably have a plurality of the above-mentioned regions, each of which has an independent polarization distribution.

[0026] In the waveplate-type polarizing mask according to the present invention, the waveplate may be a half-waveplate, in which case it is preferable that the polarized light incident on the polarizing mask and traveling in the z-direction is linearly polarized light, and that the light exiting the polarizing mask is linearly polarized light.

[0027] In the waveplate-type polarizing mask according to the present invention, the waveplate may be a quarter-waveplate, in which case it is preferable that the polarized light incident on the polarizing mask and traveling in the z-direction is circularly polarized light, and the light emerging from the polarizing mask is linearly polarized light.

[0028] A second aspect of the present invention relates to a photo-alignment exposure method. In this photo-alignment exposure method, the aforementioned waveplate-type polarizing mask is used. First, the waveplate is assumed to be a half-waveplate. Furthermore, in the region of the polarizing mask (the region that realizes different polarization orientations for incident light at different locations within the x-y plane), the axial orientation of the waveplate is rotated approximately periodically in one direction within the x-y plane. Furthermore, a portion of the light transmitted through the region is separated into right-handed circularly polarized light and left-handed circularly polarized light. As a result, each of the right-handed circularly polarized light and the left-handed circularly polarized light has a component traveling in the z direction and a component traveling in opposite directions parallel to the direction in which the axial orientation of the waveplate periodically changes within the x-y plane, and these components interfere with each other to become linearly polarized light. Furthermore, the polarization orientation of each of the right-handed circularly polarized light and the left-handed circularly polarized light rotates every half period of the period of the axial orientation of the waveplate, and the polarization distribution is formed across the z direction. This allows the molecular orientation of a photoreactive polymer to be controlled collectively, at different locations, and across the light propagation direction.

[0029] The photo-alignment exposure method according to the present invention preferably uses a laser as a light source, thereby forming a desired electric field distribution even at a position distant from the polarizing mask, and achieving photo-alignment of photoreactive polymers even without contact with the polarizing mask.

[0030] The photo-alignment exposure method according to the present invention preferably uses an LED as a light source, polarizes the light through a linear polarizer, and brings the polarizing mask and the polymer into close contact or close enough to each other that the coherence of the LED is maintained, thereby forming a desired electric field distribution and achieving photo-alignment of the photoreactive polymer.

[0031] The photo-alignment exposure method according to the present invention may use a mercury lamp as a light source, limit the wavelength using a bandpass filter, and polarize the light through a linear polarizer. In this case, too, by bringing the polarizing mask and the polymer into close contact or close enough to each other to maintain the coherence of the mercury lamp, a desired electric field distribution can be formed, thereby realizing the photo-alignment of the photoreactive polymer.

[0032] A third aspect of the present invention relates to a method for manufacturing an optical component or an optical device. The manufacturing method according to the present invention includes a step of carrying out the photo-alignment exposure method according to the second aspect. Examples of optical devices obtained by the present invention include AR devices or VR devices, and examples of optical components obtained by the present invention include polarization diffraction gratings and other optical components for AR / VR devices.

[0033] By using the present invention, a photo-alignment pattern that could previously be achieved by manipulating a light beam can be achieved with a single exposure, thereby realizing high mass productivity.

[0034] Furthermore, the pattern freedom that was previously limited by interference exposure is no longer limited, allowing for free light orientation patterns to be realized. As a result, compact, lightweight, and highly functional optical elements can be realized for AR / VR devices, which will greatly contribute to the widespread use of AR / VR devices and realize new lifestyles.

[0035] [Correction pursuant to Rule 91 31.01.2025] Figure 1 is an explanatory diagram of a polarization diffraction grating using the concept of geometric phase. Figure 2 is an explanatory diagram of a polarization diffraction grating using the concept of geometric phase. Figure 3 is a conceptual diagram of an optical component for an AR device using a polarization diffraction grating and a light guide plate. Figure 4 is an explanatory diagram of a lens using the concept of geometric phase. Figure 5 is an explanatory diagram of an interference exposure method for fabricating a polarization diffraction grating. Figure 6 is an explanatory diagram of a photonic crystal polarizer fabricated by the autocloning method. Figure 7 is a simulation result of a photonic crystal polarizer fabricated by the autocloning method. Figure 8 is an example of a pattern of a photonic crystal polarizer fabricated by the autocloning method. Figure 9 is another example of a patterned polarizer. Figure 10 is an explanatory diagram of a photonic crystal wave plate fabricated by the autocloning method. Figure 11 is an explanatory diagram of a wave plate with a subwavelength structure. Figure 12 is a simulation result of a photonic crystal wave plate fabricated by the autocloning method. Figure 13 is an example of a pattern of a photonic crystal wave plate fabricated by the autocloning method. Fig. 14 is an example of a diagram showing the state of light transmitted through a photonic crystal wave plate fabricated by the autocloning method with a periodic pattern. Fig. 15 is an example of an image display system using a polarization diffraction grating and a light guide plate by photo-alignment. Fig. 16 is a diagram explaining a lens pattern created using a photonic crystal wave plate fabricated by the autocloning method. Fig. 17 is another example of a patterned wave plate. Fig. 18 is an example of a photo-alignment exposure apparatus using a polarizing mask.

[0036] Hereinafter, embodiments of the present invention will be described with reference to the drawings. The present invention is not limited to the embodiments described below, and includes appropriate modifications of the embodiments described below within the scope obvious to those skilled in the art.

[0037] [1. Principles of Polarizing Masks for Photo-Alignment Exposure] Recently, many technologies have been proposed for controlling light using artificial microstructures, such as photonic crystals and metasurfaces. For example, photonic crystals, with their periodic structures of different refractive indices, can control the effective refractive index perceived by light, or phenomena such as transmission and reflection, by controlling the parameters of the periodic structure. As shown in the lower part of Figure 6, materials 602 and 603 with different refractive indices in the z direction are alternately stacked on a substrate 601. In a structure such as area 604, which has periodicity only in the x direction, the structural anisotropy can result in different behaviors for light polarized in the x direction and light polarized in the y direction. For example, when light travels in the z direction, a polarizer can be realized that transmits light polarized in the x direction and reflects light polarized in the y direction (Patent Document 1). Furthermore, while both polarized light beams can be transmitted, differences in the effective refractive index for each polarized light beam can be achieved. By transmitting light through an appropriate thickness, a phase difference of 1 / 2 wavelength or 1 / 4 wavelength can be achieved (Patent Document 2), allowing the polarizer to function as a wave plate.

[0038] Such structures can be fabricated by forming a pattern on a substrate such as quartz using lithography, as shown in the upper part of Figure 6, and then using the autocloning method (Patent Document 3) to create a photonic crystal on top of it. Because the pattern created on the substrate can be freely designed, it is easy to create adjacent structures with periodicity in the x direction, such as area 604, and with periodicity in the y direction, such as area 605. This allows for flexible control of the polarizer or waveplate pattern, as shown in more complex Figure 9. Furthermore, since structures with high transmittance can be realized even in the ultraviolet region (Patent Document 4), this is advantageous for realizing photoalignment.

[0039] Although photonic crystals are a well-known technology, their fabrication method will be explained below for completeness. A photonic crystal is a structure in which the refractive index changes periodically at a period shorter than the operating wavelength of propagating light. A specific method for fabricating a photonic crystal is to sequentially and periodically stack two or more materials (transparent bodies) 602 and 603 with different refractive indices on a substrate 601 having periodic one- or two-dimensional irregularities, and then fabricate an optical element (wave plate) by using sputter etching alone or simultaneously with film formation on at least a portion of the stack. This method is called the autocloning method. A photonic crystal formed by this autocloning method is called an autocloning photonic crystal. Note that a technique for constructing a wave plate using an autocloning photonic crystal is well-known. When a periodic groove pattern is prepared on a substrate and a periodic structure is formed thereon using the autocloning method, the structure exhibits structural anisotropy in the groove direction and in a direction perpendicular to the groove direction. As a result, the effective refractive index experienced by the polarized light component parallel to the groove and the component perpendicular to the groove differs. In other words, birefringence is realized. The birefringence retardation can be controlled by the number of layers. Furthermore, the axis orientation follows the pattern of the substrate, so if grooves with different directions are formed on the substrate at different locations, a wave plate with different axis orientations can be realized at different locations. This manufacturing method is highly practical for realizing patterned wave plates with axis orientations that vary depending on the location.

[0040] Other methods for forming patterned wave plates include irradiating glass with a femtosecond laser to create periodic voids, but these methods are similar in the sense that they artificially create a birefringence distribution.

[0041] The multiple transparent materials forming the autocloned photonic crystal are preferably silicon, germanium, oxides such as niobium pentoxide, tantalum pentoxide, titanium oxide, hafnium oxide, silicon dioxide, and aluminum oxide, or fluorides such as magnesium fluoride and calcium fluoride. Two or more of these materials with different refractive indices can be selected and used for the photonic crystal. For example, combinations of amorphous silicon and silicon dioxide, niobium pentoxide and silicon dioxide, and tantalum pentoxide and silicon dioxide are desirable, but other combinations are also possible. Specifically, the autocloned photonic crystal has a structure in which high-refractive-index materials and low-refractive-index materials are alternately stacked in the z-direction. The high-refractive-index materials are preferably tantalum pentoxide, niobium pentoxide, amorphous silicon, titanium oxide, hafnium oxide, or a combination of two or more of these materials. The low-refractive-index materials are preferably silicon dioxide, aluminum oxide, or a fluoride such as magnesium fluoride, or a combination of two or more of these materials.

[0042] The wavelength used for the optical alignment of liquid crystals in the present invention is mainly ultraviolet light. Therefore, tantalum pentoxide and hafnium oxide are preferred as high refractive index materials, and silicon dioxide, aluminum oxide, or fluorides such as magnesium fluoride are preferred as low refractive index materials.

[0043] The unit period between grooves of the in-plane periodic structure forming each wave plate and the unit period in the thickness direction of the wave plate are both equal to or less than half the wavelength of light incident on the optical element. It is assumed that the wavelength of light incident on the optical element is usually selected from the range of 200 nm to 400 nm. Therefore, the period of the substrate is selected appropriately within the range of 200 nm or less depending on the operating wavelength.

[0044] Reflection occurs between the substrate and the multilayer film, or between the multilayer film and the upper material due to mismatch in refractive index, but by appropriately selecting the thickness of the top few layers and the bottom few layers of the multilayer film, it is possible to achieve the same effect as a commonly used anti-reflective coating.

[0045] Another technology that has recently attracted attention is metasurfaces. This concept also involves controlling the behavior of light by controlling structures smaller than the wavelength created on the surface of a substrate. The difference between this concept and photonic crystals is that periodicity is not necessarily required. However, incorporating structures that are too random can result in unwanted scattering and diffraction, so a periodic structure is considered desirable. For example, if a structure like the one shown in Figure 11 is realized, the effective refractive index of light traveling in the z direction differs between x- and y-polarized light, resulting in structural birefringence, making it possible to create a wave plate. It is clear that by changing the orientation of this structure at different locations, the polarization direction can be controlled at different locations, as in the case of photonic crystals. Note that the basic concept of structures previously referred to as subwavelength structures is quite similar, and it is difficult to clearly draw a boundary between them, including photonic crystals.

[0046] Commonly used metasurfaces exhibit their functions by utilizing the difference in refractive index between the unevenness of a high-refractive-index material created on the surface and the surrounding air. On the other hand, the anisotropy of photonic crystals is manifested within the multilayer film, as shown in Figure 6. Therefore, the properties of a metasurface change significantly when it comes into contact with something with a refractive index other than air, but in the case of photonic crystals, the properties do not change significantly even if the surface comes into contact with something, because the function is manifested within the multilayer film. This is thought to be a major advantage in, for example, contact exposure, where a mask comes into contact with the exposure target.

[0047] By using this fine structure and optical elements that can control polarization at each location, a polarizing mask can be provided to achieve a polarization distribution for patterned photo-alignment. Furthermore, by combining this polarizing mask with a light source and a mechanism for positioning the substrate, a patterned photo-alignment exposure device can be realized. This photo-alignment can also be used to pattern alignment films for liquid crystals.

[0048] When a wave plate is formed using a photonic crystal or metasurface as described above, the phenomenon that occurs at the output portion will be explained using Figure 14. The upper part of Figure 14 shows a wave plate in which one period 1401 is divided into eight regions, and the axial orientation rotates by π in one period 1401. Note that in the multiple regions included in this one period, different polarization orientations are realized for the incident light at different locations within the x-y plane.

[0049] The light incident in the z direction is assumed to be linearly polarized in the y direction. A plane wave with a phase that is aligned in the xy plane is assumed to be incident. As shown in the upper part of Figure 14, we consider the case where light enters a half-wave plate that is patterned so that the orientation 1402 of the wave plate changes periodically in the x direction.

[0050] When linearly polarized light is incident on such a half-wave plate, if the direction of the linearly polarized light is shifted by θ from the axial orientation of the wave plate, the polarization direction is rotated by 2θ. Considering the above, the electric field vector at a certain moment immediately after exiting the patterned wave plate can be represented by the arrow 1403 in the middle of Figure 14. When the polarization direction is parallel to the fast axis orientation and when it is perpendicular to it, the phase is delayed by π, as shown by the arrows 1404 and 1405. Note that here, arrows are used to indicate the direction of the electric field vector to represent a phase shift of π. If the directions of the electric field vectors differ by π, the phases differ by π. In other words, the electric field vectors in the region indicated by the reference numeral 1404 and the region indicated by the reference numeral 1405 have opposite directions, which means that the phases differ by π (180 degrees). By periodically arranging regions with such a phase difference, the light exiting the wave plate is separated into right-handed and left-handed circularly polarized light.

[0051] What is important here is that, as shown in the middle of Figure 14, the direction of the electric field vector of the emitted light rotates by π with a period of 1406. When molecular bonds react to an electric field, the electric field vector oscillates in the direction indicated by the arrow 1403 while remaining parallel (for example, an arrow parallel to the y direction oscillates repeatedly between the +y and -y directions). Therefore, the instantaneous direction of the electric field vector (arrow) is irrelevant; only the direction of the electric field oscillation has an effect. Therefore, in the middle of Figure 14, the electric field vector indicated by the upward arrow and the electric field vector indicated by the downward arrow have the same meaning from the perspective of the molecules being oriented. Therefore, by using a patterned half-wave plate (top of Figure 14) whose axial orientation changes with a period of 1401, an electric field distribution (middle of Figure 14) with a pattern of period 1406 can be formed, and molecular orientation can be changed with half the period of the axial orientation of the wave plate.

[0052] Furthermore, this structure is a so-called polarization grating, and when linearly polarized light is incident, the outgoing light becomes circularly polarized light with opposite rotations traveling in the ±x directions. The wavefronts of the two circularly polarized light with opposite rotations are shown by dotted and dashed lines in the bottom of Figure 14, respectively.

[0053] As shown in the lower part of Figure 14, in the region close to the patterned wave plate 1407, two counterclockwise circularly polarized light beams overlap each other and travel independently. Therefore, their overlapping generates an electric field distribution below the wave plate 1407, which is the same as the electric field distribution shown by 1403 in the middle part of Figure 14. In other words, this distribution appears not only on the surface of the wave plate 1407, but also in the z direction as long as two counterclockwise circularly polarized light beams generated by the wave plate 1407 and traveling in opposite directions overlap. Therefore, even if the material to be oriented is not in close contact with the surface of the wave plate 1407, an electric field distribution such as that shown in the lower part of Figure 14 can be formed in the alignment film, thereby achieving optical alignment. For example, the polarization distribution on the plane 1407 at a coordinate z, located away from the wave plate 1407, is a distribution in which the orientation of linearly polarized light periodically changes from 0 degrees to 180 degrees, as indicated by the arrow 1403 in the middle part of Figure 14.

[0054] As mentioned above, for this phenomenon to occur, it is important that the incident light is in phase, i.e., has high coherence. This phenomenon is unlikely to occur with incoherent light, such as fluorescent light or sunlight. On the other hand, this situation can be easily created by using a light source with high temporal and spatial coherence, such as a laser. Even when using a light source that is said to have low coherence, such as an LED, it is possible to increase spatial coherence by, for example, suppressing the beam divergence angle using a beam expander or the like, thereby limiting the angular components of the beam. Therefore, even LEDs are not completely incoherent, and so they can be used for interference exposure as long as the conditions of use are considered.

[0055] If there is a space between the wave plate and the alignment film and this space is filled with air (with a refractive index of 1), there is a lower limit to the period 1401 of the patterned wave plate. In principle, diffraction does not occur for periods smaller than λ. However, if the space between the wave plate and the alignment film can be filled with a material with a refractive index n, the period can be reduced to λ / n. For example, the refractive index of water is about 1.3, so it can be reduced to about 1 / 1.3 compared to air. This is the same principle as the use of immersion exposure using pure water when exposing a pattern onto a resist in the semiconductor manufacturing process. It should be noted that the average refractive index n of the alignment film ａｖ If is smaller than n, then λ / n ａｖ is the lower bound.

[0056] The light source used is a laser, mercury lamp, or LED. The wavelength is related to the absorption spectrum of the photoreactive polymer and is expected to be in the ultraviolet region. Because the light emitted from a mercury lamp or LED is usually unpolarized, it must be passed through a polarizer to align the polarization direction. When using a mercury lamp as the light source, it is recommended to first limit the wavelength using a bandpass filter and then polarize the light through a linear polarizer. Polarizers that can be used include photonic crystal polarizers, wire grids, polarizers using oblique incidence of multilayer films, and polarizers combining birefringent optical crystals. However, because light emitted from a light source generally has a wide angular component, photonic crystals or wire grids, which have little dependence on the angle of incidence, are preferred. Furthermore, photonic crystals can also be used to create patterned polarizers. In this case, even if the light source is unpolarized, a patterned polarization distribution can be achieved by passing the light through a patterned polarizer.

[0057] [2. Photonic Crystal Polarizer] A photonic crystal polarizer using the autocloning method is described using Figure 6. This polarizer can be used as a polarizing mask for the photo-alignment exposure described above. First, periodic grooves are formed on a substrate transparent to the wavelength of the incident light using electron beam lithography or nanoimprinting. If the incident light is ultraviolet, a quartz substrate is used as the substrate material, for example. The groove period is equal to or less than the wavelength of the light used, preferably equal to or less than half. The groove depth is approximately half the period. Materials with different refractive indices are alternately layered on top of this using the autocloning method. As a result, as shown in Figure 6, it is possible to stack a multilayer film while maintaining the angle of the slope. By layering this substrate period and the thickness of each layer as described in Patent Document 2 (JP 2001-51122 A), a reflective polarizer can be realized that transmits polarized light parallel to the grooves and reflects polarized light perpendicular to the grooves. Therefore, even when unpolarized light from a high-pressure mercury lamp, LED, or the like is introduced, 50% of the light is reflected, while the remaining 50% of the light is polarized by the polarizer and emitted.

[0058] The transmission axis of this polarizer is determined by the direction of the grooves in the substrate. Therefore, if a substrate with grooves oriented differently at different locations is prepared, as shown in Figure 6, a polarizer with different transmission polarization orientations at different locations can be realized, making it possible to freely control the polarization direction within the plane.

[0059] The operating wavelength of this polarizer is determined by the groove pitch of the substrate, the thickness of each layer, and the refractive index of each material. Since photoalignment typically uses wavelengths in the ultraviolet region of 400 nm or less, the substrate pitch is 200 nm or less. Silicon dioxide (SiO ２ , and tantalum pentoxide (Ta) as a high refractive index material. ２ O ５ ), or hafnium oxide (HfO ２ ), aluminum oxide (Al ２ O ３ ) are possible options.

[0060] Since the function of the polarizer is realized by internal reflection in the multilayer film, the function as a polarizer does not change even if another material is attached to the surface. Normally, reflection occurs because there is a difference in the effective refractive index between the multilayer film and the outer material. To prevent this reflection, it is possible to perform a so-called anti-reflection treatment, which cancels out the reflection by optimizing the thickness of one or two layers. For example, the function as a polarizer does not change even if another material is attached to a surface designed for air.

[0061] The light source can be an unpolarized, incoherent light source such as a high-pressure mercury lamp or an LED. Even an unpolarized light source can be polarized in the position direction by passing it through a polarizer. Methods using optical interference require the use of light that can be interfered with, i.e., a coherent laser light source. As a result, the cost of the light source increases, and there are significant benefits to using an inexpensive unpolarized light source, especially in applications where a large area needs to be irradiated at once.

[0062] [3. Polarization Diffraction Grating Mask Used in Photonic Crystal Polarizer] A design example of a polarization mask using the aforementioned photonic crystal polarizer will be described. The central wavelength is assumed to be 355 nm. The pattern period is 145 nm, and the thickness of each layer is tantalum pentoxide (Ta ２ O５ ) was changed to 79 nm, and silicon dioxide (SiO ２ ) is set to 109 nm, and a calculation example for 40 periods (80 layers) is shown in Figure 7. In the structure of region 604 in Figure 6, polarized light parallel to the grooves is proposed as TE polarized light, and polarized light perpendicular to the grooves is proposed as TM polarized light. It can be seen that it acts as a polarizer around a wavelength of 355 nm.

[0063] The operating wavelength can be other than 355 nm, and the substrate pattern and the film thickness of each layer can be designed accordingly. Also, by gradually shifting the thickness of each layer and the operating band, the overall operating band can be widened (Patent Document 5).

[0064] Then, as shown in Figure 8, a pattern was prepared in which the groove orientation changes by 180 degrees at a period of 5 μm, and tantalum pentoxide (Ta ２ O ５ ) and silicon dioxide (SiO ２ ) materials are alternately stacked to create a photonic crystal polarizer. Here, it is assumed that the colored parts will become recesses in the substrate. The pattern can be formed by EB exposure or nanoimprinting.

[0065] While Fig. 6 shows a polarizer in which vertical and horizontal patterns are adjacent to each other, it is also possible to realize polarizers with various patterns as shown in Fig. 9. Fig. 9 is a schematic diagram showing the structure of a polarizer made by the autocloning method.

[0066] [4. Photonic Crystal Waveplate] A photonic crystal waveplate fabricated by the autocloning method will be described using Figure 10. This photonic crystal waveplate can be used as a polarizing mask for the photo-alignment exposure described above. Similar to the polarizer described above, a multilayer film is laminated by the autocloning method on a substrate with periodic grooves. As a result, as shown in Figure 10, it is possible to laminate the multilayer film while maintaining the angle of the slope. The difference between a waveplate and the polarizer described above is the thickness of each layer. In the case of a waveplate, it is important that the thickness of each layer is sufficiently thin compared to the wavelength of the incident light. For example, a thickness of less than 1 / 10 of the wavelength of the incident light is required. With this thickness, multiple reflections occurring at the interface between each layer in the multilayer film do not occur, and both polarized light beams are transmitted. However, looking at the structure 1005 in the left half of the lower diagram of Figure 10, the structure is periodic in the x direction but not in the y direction. This structure exhibits anisotropy in the optical properties, resulting in a difference in refractive index between light polarized parallel to the grooves and light polarized perpendicular to them (birefringence). Therefore, by stacking an appropriate number of layers, a phase difference of 1 / 2 wavelength can be generated between orthogonal polarized light beams. In a half-wave plate formed in this manner, if there is a difference of θ between the direction of the incident polarized light and the axial direction of the wave plate, the polarization direction rotates by 2θ, and therefore the polarization direction can be changed using this element.

[0067] Alternatively, a quarter-wave plate can be realized by adjusting the number of layers. When circularly polarized light is input to a quarter-wave plate, it becomes linearly polarized light at a 45-degree angle to the axial direction of the input light. Therefore, even in this case, the polarization direction of the output light can be controlled by the axial orientation of the wave plate.

[0068] In the autocloning method, the axial orientation of this wave plate can be controlled by the substrate pattern, so that the direction of uniform linearly polarized light can be freely changed within the plane.

[0069] The function of the wave plate is realized by the internal structural anisotropy of the multilayer film, so even if another material is attached to the surface, the function as a wave plate will not change. Normally, there is a difference in the effective refractive index between the multilayer film and the outer material, which causes reflection. To prevent this reflection, it is possible to perform a so-called anti-reflection treatment that cancels out the reflection by optimizing the thickness of one or two layers. For example, even if another material is attached to a surface designed for air, the function as a wave plate will not change.

[0070] Note that a half-wave plate is merely an element that changes the polarization direction, so when unpolarized light enters, only unpolarized light comes out. Therefore, when used for photo-alignment, the light from the light source must first be transmitted through a linear polarizer, and then the light must be incident on the element as light with only components of the same polarization direction. Examples of linear polarizers that can be used in this case include polarizers using photonic crystals, wire grids, and Brewster angles, prism-type polarizers using anisotropic optical crystals such as Glan-Thompson prisms, and polarized beam splitters using oblique incidence of multilayer films.

[0071] Furthermore, when unpolarized light enters a quarter-wave plate, only unpolarized light emerges. Therefore, when used for photo-alignment, light from a light source must first be transmitted through a linear polarizer, then passed through another quarter-wave plate with a uniform orientation to become circularly polarized light before entering this element (quarter-wave plate). Examples of linear polarizers used in this case include polarizers using photonic crystals, wire grids, and Brewster angles, prism-type polarizers using anisotropic optical crystals such as Glan-Thompson prisms, and polarized beam splitters using oblique incidence of multilayer films. Other quarter-wave plates include those using crystals with inconspicuous birefringence such as quartz and calcite, those using the birefringence of polymers, and those that exhibit phase difference due to reflection, such as Fresnel rhombic plates.

[0072] [5. Metasurface Waveplate] The metasurface waveplate will be explained using FIG. 11. This metasurface waveplate can be used as a polarization mask for the optical alignment exposure described above. For example, in a structure in which grooves are arranged at a period sufficiently smaller than the wavelength of the incident light as shown in FIG. 11, structural anisotropy occurs as with the photonic crystal described above, and if the depth of the grooves is optimized, it will function as a waveplate. For example, if the refractive index of the material is n, the refractive index of the air between them is 1, the width of the grooves is w, and the period of the grooves is p, then the effective refractive index n sensed by polarized light in a direction parallel to the grooves is ｓ is expressed by the following equation (2). In addition, the effective refractive index n ｆ is expressed by the following equation (3). If n=1.5, then n ｓ = 1.275, n ｆ = 1.177, so to realize a half-wave plate at a wavelength of 350 nm, a groove with a depth of 1.78 μm is required. To realize a quarter-wave plate, a groove with a depth of 0.89 μm is required. In order to realize structural birefringence with such a structure, the period p of the structure needs to be a fraction of the wavelength. In other words, it is necessary to create a structure with a fairly high aspect ratio. Therefore, it is conceivable that the surface would be easily destroyed if force is applied to it.

[0073] This type of structure is called a subwavelength structure or metasurface. Metasurface is a concept that improves functionality by adding a fine structure to a surface, and it is also called this when referring to a structure consisting of round pillars, rather than a structure consisting of an array of plates as shown in Figure 11. Anisotropy can also be expressed by making the cross section of these pillars elliptical. Although anisotropy is necessary to control polarization, this is sufficient as a principle of operation. In the structure shown in Figure 11, by changing the direction of the grooves depending on the location, this is equivalent to changing the axial orientation of the wave plate depending on the location, and the polarization orientation can be freely controlled within the plane.

[0074] Since the function of the wave plate is realized in the space with the grooves, the function does not change even if a material with a different refractive index is placed on top of it. However, in reality, a structure like the one in Figure 11 is fragile and will break immediately if force is applied, so it is unrealistic to use it in contact with something.

[0075] However, since this element merely changes the polarization direction, if unpolarized light enters, only unpolarized light will come out. Therefore, when used for photo-alignment, the light from the light source must first pass through a linear polarizer, and then enter the element as light with only the components of the same polarization direction. Examples of linear polarizers that can be used in this case include polarizers using photonic crystals, wire grids, and Brewster angles, prism-type polarizers using anisotropic optical crystals such as Glan-Thompson prisms, and polarized beam splitters using oblique incidence of multilayer films.

[0076] Furthermore, when unpolarized light enters a quarter-wave plate, only unpolarized light emerges. Therefore, when used for photo-alignment, light from a light source must first be transmitted through a linear polarizer, then passed through another quarter-wave plate with a uniform orientation to become circularly polarized light before entering this element (quarter-wave plate). Examples of linear polarizers used in this case include polarizers using photonic crystals, wire grids, and Brewster angles, prism-type polarizers using anisotropic optical crystals such as Glan-Thompson prisms, and polarized beam splitters using oblique incidence of multilayer films. Other quarter-wave plates include those using crystals with inconspicuous birefringence such as quartz and calcite, those using the birefringence of polymers, and those that exhibit phase difference due to reflection, such as Fresnel rhombic plates.

[0077] [6. Polarization Diffraction Grating Mask Using Photonic Crystal Wave Plate] The design of a polarization mask using the above-mentioned photonic crystal wave plate will be described. For example, the pattern period is 100 nm, and the thickness of each layer is tantalum pentoxide (Ta ２ O ５ ) is 20 nm, silicon dioxide (SiO ２ The thickness is set to 20 nm. This thickness should be sufficiently thin compared to the wavelength. For example, it is desirable that the thickness is 1 / 10 or less of the wavelength.

[0078] Figure 12 shows the wavelength dependence of retardation calculated by the FDTD method for a 98-layer multilayer structure. It can be seen that the retardation is π, that is, it becomes a half-wave plate, at a wavelength of around 355 nm. For example, as shown in Figure 13, a pattern is prepared in which the groove orientation changes at a period of 1 μm, and tantalum pentoxide (Ta) is applied on top of it. ２ O ５ ) and silicon dioxide (SiO ２ The photonic crystal wave plate is realized by alternately stacking the above materials. The pattern can be formed by EB exposure or nanoimprinting.

[0079] Next, we will discuss the polarization distribution of light transmitted through a photonic crystal wave plate. Consider the case where the period is sufficiently large. Consider vertically linearly polarized light incident on each of the rectangular regions shown in the upper panel of Figure 14. Because each region functions as a half-wave plate, if the difference between the polarization direction and the axial direction is θ, the polarization direction of the emitted light will rotate by 2θ. This results in the polarization distribution shown in the lower panel of Figure 14. This distribution is half the period of the pattern of the photonic crystal wave plate. Therefore, for example, to create a periodic polarization distribution with a period of 300 nm, the structure of the photonic crystal wave plate can be 600 nm. When realizing such a fine structure, the smaller the structure, the more susceptible it is to various disturbances during the process. For example, photoresist patterns have inherent fluctuations, and smaller patterns are more susceptible to these fluctuations. Therefore, the characteristic of requiring a large period works in favor of uniformity and yield during manufacturing. This is an effect unique to polarizing masks using wave plates, an effect that cannot be obtained with polarizer-based polarizing masks.

[0080] Furthermore, assume that the incident linearly polarized light maintains coherence over a spatial area wider than the width of the strip in the upper part of Figure 14. In this case, the polarized light at the exit point has a phase shift of π between the region 1404 and the region 1405. This distribution can be considered as a superposition of left-handed circularly polarized light and right-handed circularly polarized light, each with a positive and negative wave vector in the x direction. As a result, the light from the exit point becomes left-handed circularly polarized light and right-handed circularly polarized light, traveling obliquely, as shown in the lower part of Figure 14. In the region where these polarized light overlap, the polarization distribution on the plane 1408 at a certain coordinate z below the wave plate 1407 (polarizing mask) is also a distribution in which the direction of linear polarization periodically changes from 0 degrees to 180 degrees, as indicated by the arrow in the middle part of Figure 14.

[0081] In other words, when a photonic crystal waveplate is used as a polarizing mask, it is not necessary to directly touch the material to be oriented, and the polarization orientation pattern can be projected onto the alignment film even from a distance. This means that high-precision exposure is possible without directly touching the polarizing mask, and contamination of the polarizing mask surface can be prevented, which is a major advantage for mass production.

[0082] The above effect is not limited to photonic crystal wave plates (polarization masks), but is also applicable to wave plates (polarization masks) that use microstructures such as metasurfaces.

[0083] Furthermore, the possible distance between the alignment film and the mask varies depending on the coherence of the light source. When a highly coherent laser is used, the interference pattern will not disappear even if the laser is placed several hundred microns away. On the other hand, when an LED is used, the coherence length can be several microns. In such cases, the interference pattern can be utilized by closely adhering the mask and alignment film. The ability to use an inexpensive LED light source in this way is an industrial advantage.

[0084] [7. Design Example of Polarization Diffraction Grating by Optical Alignment] Consider an image display system using a light guide plate used in AR glasses, for example. As shown in Figure 15, this image display system includes a microprojector 1501 that outputs an image, a lens 1502 that collimates the light emitted from the microprojector so that it can enter the light guide plate, a first diffraction grating 1503 that introduces the light into the light guide plate, a light guide plate 1504, and a second diffraction grating 1505 that extracts the light from the light guide plate and directs it into the eyeball 1506. A design example will now be described, assuming that the first diffraction grating 1503 and the second diffraction grating 1505 are fabricated using the optical alignment technique of the present invention.

[0085] The light source of the display constituting microprojector 1501 is assumed to be an LED, with RGB wavelengths of λR = 620 nm, λG = 530 nm, and λB = 460 nm. For the configuration in Figure 15 to function, two conditions must be met: 1) diffraction occurs at first diffraction grating 1503 on the entrance side and second diffraction grating 1505 on the exit side, and 2) light is totally reflected within light guide plate 1504.

[0086] From the above condition 1), the refractive index of the light guide plate 1504 is set to n ｇ , the incident angle to the first diffraction grating 1503 is θ ０ , the output angle from the second diffraction grating 1505 is θ １ If the period of the strip-shaped regions of the first diffraction grating 1503 and the second diffraction grating 1505 is P, then the following holds from the law of conservation of wave numbers in the x direction. For diffraction to occur, θ １ <π / 4, that is, sin θ １ <1, the following equation (4) is obtained. In addition, considering the above condition 2), n ｇ sinθ １ >1, the following equation (5) is obtained. From equations (4) and (5), the condition of the following equation (6) must be satisfied. In addition, this is the case for the three wavelengths of RGB, and θ ０ is the range of the viewing angle (±θ ０For example, at the three wavelengths mentioned above, the refractive index n of the light guide plate 1504 must be ｇ When the period P of the strip-shaped regions of the first diffraction grating 1503 and the second diffraction grating 1505 is 1.9, the period P of the strip-shaped regions of the first diffraction grating 1503 and the second diffraction grating 1505 can be satisfied if it is 380 nm. Furthermore, if the concept explained in [6. Polarization diffraction grating mask using photonic crystal wave plate] is used, a pattern having a period P of 760 nm as shown in FIG. 1 is formed.

[0087] [8. Polarizing Mask for Lens Pattern] In Figure 14, a structure in which strip-shaped regions are periodically arranged in one direction is assumed. However, the present invention is not limited to this structure and can also orient two-dimensional patterns. For example, the pattern shown in Figure 16 is divided into circular rings from the center, each with the same width. The pattern is divided into radii 1601, and the orientation θ changes along the direction. To obtain a lens with a focal length f, the phase φ must be given according to the radius r as shown in the following equation (7). Therefore, the orientation θ of the ring at radius r is expressed as follows: When the region is divided into finite widths as in Figure 16, appropriate discretization is required, such as using the average value of the radius of each annular region. However, it is clear that the finer the division width, the more ideal the phase distribution can be obtained.

[0088] The pattern shown in Figure 16 can also be oriented, and as a result, it is possible to realize a plate-shaped lens using birefringence. In this case, too, if we consider a line segment at a certain radius from the center, circularly polarized light and left-handed circularly polarized light travel in opposite directions after the light leaves the polarizing mask, as explained in Figure 14, and a similar interference phenomenon occurs, making it possible to achieve a polarization distribution similar to that on the surface of the polarizing mask even in a space away from the polarizing mask. In the case of a lens with a pattern such as that shown in Figure 16, the period in which the orientation changes is not constant, as in Figure 14, but the magnitude of the change in the period is sufficiently small compared to the period, so it does not have a significant effect.

[0089] Furthermore, it is clear that the phase distribution to be applied is not limited to Equation (7), and other patterns can be realized. For example, various patterns such as those shown in Figure 17 are also possible. If each region is sufficiently larger than the wavelength, the above-mentioned interference effect does not need to be taken into consideration. Figure 17 is a schematic diagram showing the structure of a wave plate made by the autocloning method.

[0090] [8. Photo-Alignment of Polyimide Alignment Films] Photo-decomposition occurs when polyimide is irradiated with ultraviolet light, but it is known that the photo-decomposition becomes anisotropic when polarized light is irradiated (Non-Patent Document 4). The use of polyimide to align liquid crystal molecules is a technique widely used in LCD televisions and other applications. For example, by irradiating a polyimide film with polarized light with different orientations in different locations and then coating it with nematic liquid crystal, it is possible to create regions where the alignment direction of the liquid crystal molecules varies from location to location. This means that the axis orientation of birefringence can be controlled at each location.

[0091] The polarizing mask of the present invention can be used in a method for forming a film for aligning liquid crystal molecules, including polyimide. The polarizing mask of the present invention can achieve a polarization distribution with different orientations at different locations. When the film is irradiated with this light, molecular bonds with a component parallel to the polarization direction of the light absorb and decompose the light in proportion to the proportion of the parallel component, resulting in molecular bonds with a direction dependent on the polarization direction remaining. When liquid crystal molecules are applied to a film in this state, they sense the anisotropy and align in one direction, allowing them to be oriented in a pattern induced by the polarizing mask.

[0092] In this case, since only the molecules on the surface of the alignment film contribute to the alignment of the liquid crystal molecules, the polarization distribution required for alignment only needs to be realized on the surface of the alignment film. For example, if a pattern close to the wavelength is created using a polarizing mask, the pattern will quickly become blurred as the distance from the polarizing mask increases due to the effect of light diffraction. However, if only the outermost surface of the alignment film is oriented, and the liquid crystal molecules sense this and align themselves in that orientation for the required thickness, this problem is solved. Therefore, it is clear that this feature is advantageous when attempting to create alignment patterns with higher resolution.

[0093] Furthermore, when light is incident on a polarizing mask at an oblique angle, the light that passes through the polarizing mask also retains the oblique component, and as a result, the direction of the oriented molecules also has an oblique component, which clearly means that so-called tilted orientation and patterned orientation can be achieved.

[0094] [9. Photoisomerization] Azobenzene has long been known to undergo photoisomerization, and its structure changes when irradiated with ultraviolet light. When irradiated with polarized light, only molecules aligned in a specific direction undergo a structural change due to the anisotropy of the reaction. Therefore, when azobenzene dispersed in a liquid crystal is irradiated with modified light, anisotropy occurs in the structure of some of the azobenzene. For example, by heating the liquid crystal molecules to make them more mobile, they will align in a direction that follows the structural anisotropy of the azobenzene.

[0095] Using the polarizing mask of the present invention, azobenzene is irradiated with patterned polarized light to trigger alignment, ultimately resulting in a liquid crystal film oriented in that pattern. This method is industrially advantageous because it allows the alignment film coating and liquid crystal molecule coating processes to be performed simultaneously compared to conventional liquid crystal film manufacturing methods. However, because azobenzene must be isomerized throughout the entire liquid crystal layer to be oriented, achieving a half-wave plate, for example, with a refractive index difference Δn of the final liquid crystal requires a minimum film thickness T that satisfies λ / 2 = TΔn. For example, for a wavelength of 500 nm and a refractive index difference Δn of 0.2, a film thickness of T = 1250 nm is required.

[0096] The difference between a liquid crystal film oriented in a pattern using the polarizing mask of the present invention and a liquid crystal film oriented with an alignment film is that the present invention can realize a three-dimensional structure in the liquid crystal. By applying light obliquely to the normal of the substrate coated with the alignment film, structural changes in the alignment film can be realized in the thickness direction as well. As a result, obliquely aligned liquid crystal molecules can be realized.

[0097] [10. Regarding the Light Source] When performing optical alignment using the interference of light emitted from a patterned wave plate formed by a half-wave plate (incident light is linearly polarized), the coherence of the light source is particularly important. Light sources with high temporal and spatial coherence, such as lasers, can be used without any particular problems. Care must be taken when using LEDs. The light emitted from LEDs is not as coherent as lasers, and it is not easy to obtain a stable interference pattern over a distance of, for example, centimeters. However, in the case of the present invention, the required thickness of the interference fringes is several microns, so the temporal coherence of a typical LED spectrum (e.g., a wavelength width of ±5% of the center wavelength) is sufficient. The coherence length L of a light source with a wavelength λ and a wavelength spread Δλ is given by the following equation (9): For example, if the wavelength is 355 nm and the wavelength width is 30 nm, then L = 4200 nm. In terms of spatial coherence, it is possible to obtain interference fringes in the required area by collimating the light emitted from the LED using a lens.

[0098] [11. Substrate] The base material of the substrate to which the polymer is applied may be glass or quartz, or may be a film such as PET or PC.

[0099] [12. Photo-Alignment Exposure Apparatus] As shown in FIG. 18 , a photo-alignment exposure apparatus using a polarizing mask basically comprises a light source 1801, an optical system including a lens 1802 that guides the light emitted from the light source 1801 to a sample 1804, a polarizing mask 1803, and a sample stage 1805.

[0100] The light source 1801 is assumed to be an ultraviolet light source that emits light with a wavelength of 400 nm or less, and a high-pressure mercury lamp, LED, or laser can be used. When light coherence is important, a laser light source is most desirable. Although high-pressure mercury lamps and LEDs are basically light sources with low coherence, for example, spatial coherence can be increased by extracting only the components of the light emitted from the light source within a limited solid angle.

[0101] The optical system that guides the light emitted from the light source 1801 is mainly composed of a lens 1802 that converts diffused light into collimated light, and plays a role in making the irradiation intensity on the sample surface uniform.

[0102] Polarizing mask 1803 is as described above and may be either a polarizer type or a waveplate type.

[0103] The sample stage 1805 is a platform on which a sample 1804 (substrate) coated with a material to be photo-aligned, such as liquid crystal, is placed. The sample stage 1805 has a mechanism that can move in three axes: the x-axis, y-axis, and z-axis. The sample stage 1805 is preferably movable in a plane parallel to the polarizing mask 1803 as needed, and has a mechanism that can fine-tune the distance from the polarizing mask 1803. When photo-alignment is performed using the interference of light emitted from the polarizing mask 1803, the coherence of the light source is particularly important, as described above, so it is desirable that the distance between the sample stage 1805 and the polarizing mask be as close as possible. The distance between the sample stage 1805 and the polarizing mask 1803 can be appropriately set taking into account the coherence length of the light source. Note that, although light is irradiated perpendicularly to the sample in FIG. 18 , if light is to be irradiated obliquely onto the sample 1804, the light source 1801 and lens 1802 can be tilted.

[0104] In the exposure device, the exposure conditions are appropriately set depending on the type of light source and the target material. For example, when a high-pressure mercury lamp is used as the ultraviolet light source, the illuminance of the i-line (365 nm) is set to 50 to 200 mW / cm. ２The exposure time can be adjusted within a range of 30 to 180 seconds. When using an LED light source (365 nm), the illuminance can be set to 20 to 100 mW / cm. ２ The exposure time can be adjusted within a range of 60 to 300 seconds. The sample stage 1805 has a temperature control function, which can keep the substrate temperature constant within a range of 20 to 30°C during exposure. It is also desirable to control the temperature of the exposure environment within a range of 23±2°C and humidity within a range of 45±5% RH.

[0105] 101...Period at which the orientation of the wave plate changes 102...Arrow indicating the axial orientation of the birefringence of the wave plate 201...Patterned wave plate shown in Figure 1 202...Diffraction angle 203...Diffraction angle 301...Microdisplay 302...Diffraction grating 303...Light guide plate 304...Diffraction grating 305...Naked eye 401...Arrow indicating the axial orientation of the wave plate 402...Width of the wave plate region having one axial orientation 501...Polarizing beam splitter 502...Symbol indicating polarized light parallel to the paper surface 503...Symbol indicating polarized light perpendicular to the paper surface 504...1 / 4 wave plate 505...1 / 4 wave plate 506...Mirror 507...Mirror 508...Substrate 601...Substrate of photonic crystal 602...Material of multilayer film 603...Material of multilayer film 604...Period of the substrate of the photonic crystal 605...Photonic crystal with periodicity in the x direction 606...Photonic crystal with periodicity in the y direction 801...Concave portion of substrate 1001...Photonic crystal substrate 1002...Multilayer film material 1003...Multilayer film material 1004...Period of photonic crystal substrate 1005...Photonic crystal with periodicity in the x direction 1401...Period at which the orientation of the wave plate changes 1402...Arrow indicating the fast axis direction of the wave plate 1403...Arrow indicating the electric field direction 1404...Arrow indicating the electric field vector 1405...Arrow indicating the electric field vector 1406...Period at which the electric field vector changes 1407...Patterned wave plate with the period of 1401 in the axial direction 1408...Line indicating the surface of a certain z coordinate 1501...Microdisplay 1502...Lens 1503...Diffraction grating 1504...Light guide plate 1505...Diffraction grating 1506...Naked eye 1601... Difference in radius between inner and outer circles of an annular region having the same wave plate orientation 1801... Light source 1802... Lens 1803... Polarizing mask 1804... Sample 1805... Sample stage

Claims

1. A polarizing mask for photo-alignment exposure in which the polarization direction of transmitted light is controlled at each location within the x-y plane by a polarizer including a photonic crystal or a wire grid in the three-dimensional coordinate space of x, y, and z. When unpolarized light traveling in the z direction is incident on the polarizing mask, it has an area where different polarization directions are realized at each location within the x-y plane, and the molecular orientation of a photoreactive polymer coated on a substrate placed parallel to the polarizing mask can be collectively controlled at each location.

2. A polarizing mask for photo-alignment exposure in which the polarization orientation of transmitted light is controlled at each location within the x-y plane by a wave plate including a photonic crystal, metasurface, or subwavelength structure in the three-dimensional coordinate space of x, y, and z, and which has an area where different polarization orientations are realized at each location within the x-y plane when polarized light traveling in the z direction is incident on the polarizing mask, and which can collectively control the molecular orientation of a photoreactive polymer coated on a substrate placed parallel to the polarizing mask at each location.

3. The polarizing mask according to claim 1 or 2, wherein the polarizing mask has a plurality of the regions, each of which has an independent polarization distribution.

4. The polarizing mask of claim 2, wherein the wave plate is a half-wave plate, the polarized light traveling in the z direction that enters the polarizing mask is linearly polarized, and the light that exits the polarizing mask is linearly polarized.

5. The polarizing mask of claim 2, wherein the wave plate is a quarter wave plate, the polarized light traveling in the z direction incident on the polarizing mask is circularly polarized, and the light exiting the polarizing mask is linearly polarized.

6. A photo-alignment exposure method using the polarizing mask of claim 2, wherein in the region of the polarizing mask, the axial orientation of the wave plate is rotated almost periodically in one direction in the x-y plane, the wave plate is a half-wave plate, and a portion of the light transmitted through the region is separated into right-handed circularly polarized light and left-handed circularly polarized light, each of the right-handed circularly polarized light and the left-handed circularly polarized light having a component traveling in the z direction and a component traveling in opposite directions parallel to the direction in which the axial orientation of the wave plate periodically changes in the x-y plane, which interfere with each other to become linearly polarized light, and the polarization orientation rotates every half period of the period of the axial orientation of the wave plate, and the polarization distribution is formed across the z direction, thereby making it possible to control the molecular orientation of a photoreactive polymer collectively, at each location, and across the propagation direction of light.

7. A photo-alignment exposure method using the polarizing mask described in claim 6, wherein by using a laser as a light source, a desired electric field distribution can be formed even at a position distant from the polarizing mask, and photo-alignment of photoreactive polymers can be achieved even without contact with the polarizing mask.

8. A photo-alignment exposure method using the polarizing mask described in claim 6, wherein an LED is used as a light source, the light is polarized through a linear polarizer, and the polarizing mask and polymer are brought into close contact with each other or close enough to maintain the coherence of the LED, thereby forming a desired electric field distribution and achieving photo-alignment of photoreactive polymers.

9. A photo-alignment exposure method using the polarizing mask described in claim 6, comprising: a mercury lamp as a light source; limiting the wavelength using a bandpass filter; polarizing the light through a linear polarizer; and bringing the polarizing mask and polymer into close contact or close enough to each other to maintain the coherence of the mercury lamp, thereby forming a desired electric field distribution and achieving photo-alignment of photoreactive polymers.

10. A method for manufacturing an optical component or an optical device, comprising a step of carrying out the photo-alignment exposure method according to claim 6.

Images