JP2019215435A - Vacuum ultraviolet polarizing element - Google Patents

Abstract

To present a further appropriate configuration of a vacuum ultraviolet polarizing element usable for processing such as optical alignment, etc.SOLUTION: A grid 2 provided on a substrate 1 which is transparent to vacuum ultraviolet light comprises a large number of linear parts 3 extending in parallel, an interval between the linear parts 3 being a space with no filler provided therein. The material of each linear part 3 is an oxide of a group 3 or group 4 element, with which PE obtained by a formula PE=T×log(ER) (where, T represents a transmission factor by grid, ER represents an extinction ratio by grid) is 0.2 or greater in a combination in which the PE is highest in a vacuum ultraviolet region.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a vacuum ultraviolet light polarizing element for polarizing vacuum ultraviolet light having a wavelength of 200 nm or less.

  Among various polarizing elements, a grid polarizing element having a structure in which a fine striped grid is provided on a transparent substrate is widely used because a relatively large irradiation area can be irradiated with polarized light. Among these, in the field of alignment treatment that gives a certain direction to the molecular structure in the member, this is practically performed by irradiation with polarized light, and is generally called photo-alignment.

In photo-alignment, the wavelength of polarized light is shorter in order to irradiate a wavelength with higher energy to improve the processing efficiency. That is, although it was initially a visible short wavelength region, recently, ultraviolet light has been frequently used, and near ultraviolet light such as 365 nm is also being used.
In order to achieve such a short wavelength, the grid polarizing element used to be a reflective type (wire grid polarizing element) that uses a metal such as aluminum as a grid material. Absorption-type grid polarizing elements utilizing absorption have been developed and used.

  In the grid polarizing element, the grid has a stripe shape including a large number of linear portions extending in parallel with each other. When the interval (gap width) between the linear portions is appropriately shortened with respect to the wavelength of the light, linearly polarized light having an electric field component exclusively in the direction perpendicular to the length direction of the linear portions is emitted from the grid. . For this reason, the axis of the polarized light (direction of the electric field component) is directed in the desired direction by controlling the posture of the grid polarization element so that the length direction of each linear portion of the grid is directed in the desired direction. Polarized light is obtained.

  Hereinafter, for convenience of explanation, linearly polarized light whose electric field is directed in the length direction of each linear portion of the grid is called s-polarized light, and linearly polarized light whose electric field is directed in a direction perpendicular to the length direction is p. Called polarized light. Usually, the surface perpendicular to the reflecting surface (the surface perpendicular to the reflecting surface and including the incident light and the reflected light) is called s wave, and the parallel one is called p wave, but the length direction of each linear part is incident. The distinction is made on the assumption that the surface is perpendicular to the surface.

  Basic indices indicating the performance of such a polarizing element are the extinction ratio ER and the transmittance T. The extinction ratio ER is the ratio (Ip / Is) of the intensity (Ip) of p-polarized light to the intensity (Is) of s-polarized light among the intensity of polarized light transmitted through the polarizing element. The transmittance T is the ratio of the energy of outgoing p-polarized light to the total energy Iin of incident s-polarized light and p-polarized light (T = Ip / Iin). An ideal polarizing element has an extinction ratio ER = ∞ and a transmittance T = 50%.

JP 2015-125280 A Japanese Patent No. 4778958

  Grid polarizers are often used for optical processing such as optical alignment, and as described above, the wavelength has been shortened for more efficient processing. Therefore, it is also conceivable to be able to polarize vacuum ultraviolet light (wavelength of 200 nm or less) shorter than the near ultraviolet region. However, when the wavelength region is 200 nm or less, the energy becomes too high, and there is a possibility of causing problems before the desired processing, such as destroying the molecular structure of the object. Vacuum ultraviolet light is a wavelength region often used in the field of light cleaning that decomposes and removes harmful organic substances by light irradiation. From this, vacuum ultraviolet light is also used for optical processing such as photo-alignment. Cannot be used.

For this reason, a grid polarizer that polarizes vacuum ultraviolet light has not been intended so far and has not been studied. For this reason, there is no document that specifically teaches a grid polarizing element that polarizes vacuum ultraviolet light, including the appropriate grid material and characteristics.
In this situation, the inventor can use vacuum ultraviolet light for processing such as photo-alignment, etc., by setting appropriate irradiation conditions. I thought it might be. Based on such a concept, the inventors of the present invention have intensively studied an appropriate configuration of the vacuum ultraviolet light polarizing element, and have come to the invention of this application. Therefore, an object of the present invention is to provide a more appropriate configuration of a vacuum ultraviolet light polarizing element that can be used for processing such as optical alignment.

In order to solve the above problems, the invention according to claim 1 of this application is a vacuum ultraviolet light polarizing element that polarizes vacuum ultraviolet light having a wavelength of 200 nm or less,
A substrate transparent to vacuum ultraviolet light, and a grid provided on the substrate,
The grid is made up of a number of parallel linear parts,
It is a structure where no filler is provided between each linear part,
The material of each linear portion is an oxide of an element belonging to Group 3 or Group 4 and has a formula of PE = T ² × log ₁₀ (ER) (where T is transmittance through a grid, and ER is transmittance through a grid). The extinction ratio is such that PE is 0.2 or more in the combination in which the highest PE in the vacuum ultraviolet region is obtained.
Further, in order to solve the above problem, the invention according to claim 2 has the configuration according to claim 1, wherein the vacuum ultraviolet region includes a wavelength of 172 nm.
In order to solve the above-mentioned problem, the invention according to claim 3 is the invention according to claim 1 or 2, wherein the material forming each linear portion is a part of the element belonging to the third or fourth group. Is substituted by another element, and the substitution ratio is not more than the ratio at which the PE becomes 0.2 in the combination in which the PE is highest in the vacuum ultraviolet region.

As described below, according to the invention described in claim 1 of this application, since each linear portion is formed of an oxide of an element belonging to Group 3 or Group 4, the linear portion has a high level such as ozone due to vacuum ultraviolet rays. Even in an environment where a species having an oxidizing action is present, a change in polarization characteristics can be suppressed to a small value. Further, the material of the grid is selected so that the PE represented by T ² × log ₁₀ (ER) is 0.2 or more, and the dimensions of each part of the grid are selected. It can be suitably used.
According to the second aspect of the present invention, in addition to the above-described effects, a light source that emits vacuum ultraviolet light having a wavelength of 172 nm can be used. It can be practical.
According to the third aspect of the present invention, in addition to the above-described effects, high polarization performance can be obtained in a vacuum ultraviolet region even when the element is substituted with another element in order to improve workability or adjust a refractive index. Can be.

1 is a schematic perspective view of a vacuum ultraviolet light polarizing element according to an embodiment. It is an Ellingham diagram of the oxide of the main elements of Group 3 and Group 4. It is a figure of the result of having investigated the polarization performance of the grid polarizing element for near ultraviolet light sold from each company. It is a figure of the result of having investigated the polarization performance of the grid polarizing element for near ultraviolet light sold from each company. It is a figure which shows the result of the simulation experiment which examined what kind of refractive index n and the light absorption coefficient a materialize PE> = 0.2. FIG. 4 is a graph of the values of n and a in the vacuum ultraviolet region for oxides of Group 3 and Group 4 elements. FIG. In hafnium oxide, changes in n and k are shown when part of the hafnium is replaced by silicon. (1) is a graph of wavelength vs. n, and (2) is a graph of wavelength vs. k. In hafnium oxide, changes in n and k are shown when a part of hafnium is replaced with aluminum. Similarly, (1) is a graph of wavelength vs. n and (2) is a graph of wavelength vs. k. It is the schematic which showed about the manufacturing method of the vacuum ultraviolet light polarization element of embodiment. It is a front schematic diagram of a photo-alignment device carrying a vacuum ultraviolet light polarizing element of an embodiment.

Next, modes (embodiments) for carrying out the invention of this application will be described.
FIG. 1 is a schematic perspective view of a vacuum ultraviolet light polarizing element according to the embodiment. The vacuum ultraviolet light polarizing element shown in FIG. 1 includes a transparent substrate 1 and a grid 2 provided on the transparent substrate 1.
The transparent substrate 1 is “transparent” in the sense that it has sufficient transparency with respect to the target wavelength (the wavelength of light polarized using a polarizing element). In this embodiment, since the wavelength of the vacuum ultraviolet region of 200 nm or less is assumed as the target wavelength, quartz glass (for example, synthetic quartz) is adopted as the material of the transparent substrate 1. The transparent substrate 1 has an appropriate thickness in consideration of mechanical strength for stably holding the grid 2 and ease of handling as an optical element. The thickness is, for example, about 0.5 to 10 mm.

  As shown in FIG. 1, the grid 2 is a stripe having a large number of linear portions 3 extending in parallel. The grid polarizing element performs a polarizing action by alternately and parallelly arranging regions having different optical constants. The space 4 between each linear part 3 is called a gap, and a polarization action is obtained by each linear part 3 and each gap 4. The width w of each linear portion 3 and the width of the gap 4 are appropriately determined so that a polarization action can be obtained with respect to light of the target wavelength. Specifically, the width of the gap 4 is generally equal to or smaller than the target wavelength. In this embodiment, no special filler is provided in the gap 4. Therefore, the refractive index of the gap 4 is the refractive index of the atmosphere in which the polarizing element is placed. Usually, it is air (refractive index 1).

  The vacuum ultraviolet light polarizing element of the embodiment operates in an absorption type model. That is, the electric field of the s-polarized light is divided by the dielectric constant of each linear portion 3 forming the grid 2 and is localized in each linear portion 3 and propagates while being attenuated by absorption. , Does not substantially occur, and propagates without significant attenuation. For this reason, p-polarized light is emitted exclusively from the transparent substrate 1, and a polarizing action is obtained. The operation model of the absorption-type grid polarizer is described in detail in Japanese Patent Application Laid-Open No. H10-157, and thus is omitted.

In the vacuum ultraviolet polarizing element of such an embodiment, a material particularly optimized for the polarization of vacuum ultraviolet light is selected as the material of each linear portion 3. Hereinafter, this point will be described.
It is oxidation resistance that needs to be examined first for the material of each linear portion 3 of the vacuum ultraviolet polarizing element. As is well known, vacuum ultraviolet light is largely absorbed by oxygen molecules in the air and creates abundant species having high oxidizing effects such as oxygen radicals, ozone, and hydroxyl radicals. For this reason, if the oxidation resistance of the material of each linear part 3 is low, when it uses for polarization | polarized-light of vacuum ultraviolet light, each linear part 3 will oxidize in a short period, and a characteristic will change. The change in the characteristics cannot be obtained as expected in the polarization characteristics such as transmittance and extinction ratio, that is, appears as deterioration.

  In consideration of this point, the vacuum ultraviolet light polarizing element of the embodiment first selects a material having high oxidation resistance as a grid material (material of each linear portion 3). At this time, in this embodiment, the oxidation resistance is reconsidered in consideration of the absorption type grid polarizing element. That is, in the absorption type grid polarizing element, a material that appropriately absorbs light of a target wavelength is used as the grid material, and in the ultraviolet region, a metal oxide such as titanium oxide is often used. Considering this point, the oxidation resistance is reconsidered not as “not easily oxidized” but as “not further oxidized”. That is, the stability of oxidation state (oxidation stability) is regarded as oxidation resistance.

  According to the study of the inventor, in general, the transition metal of Group 3 or Group 4 which easily has +2 to +4 valence easily forms a stable oxide, and is an element which forms an oxide for a grid material. Suitable as. However, in practice, it is necessary to consider the relationship with the transparent substrate. Oxide crystals such as quartz, zirconia crystals, and magnesium oxide crystals also have optical transparency and can be used as a material for a transparent substrate of a grid polarizer. In this case, if the oxidation stability is lower than the oxide forming the transparent substrate, oxygen is easily taken and reduced on the transparent substrate side, and then the oxidized species in the atmosphere (oxygen, oxygen radical, ozone, etc.) It is likely to be reoxidized by. As a result of such unstable reduction by the material of the transparent substrate and oxidation by the oxidizing species in the air, the optical characteristics are likely to change. For this reason, it is not preferable to use such a material as a grid material.

The oxidation stability of metal oxides is known as the so-called Ellingham diagram. FIG. 2 is an Ellingham diagram of oxides of the main elements of Group 3 and Group 4. In this embodiment, since the transparent substrate 1 is made of quartz, the standard chemical potential of silicon oxide is also added for comparison. The horizontal axis in FIG. 2 is the absolute temperature, and the vertical axis is the standard cast energy.
As shown in FIG. 2, it can be seen that titanium oxide, zirconium oxide, hafnium oxide, and yttrium oxide have lower standard Gibbs energy and higher oxidation stability than silicon oxide. Therefore, these materials can be candidates for the grid material of the vacuum ultraviolet light polarizing element.

On the other hand, as a grid material of a vacuum ultraviolet light polarizing element, it is not enough to simply have high oxidation stability, and it is necessary to sufficiently exhibit basic performance (transmittance and extinction ratio) as a polarizing element. As an index for examining this point, the inventor has come to conceive a quantity PE to be combined by the equation PE = T ² × log ₁₀ (ER). Hereinafter, this point will be described in detail.

  In a grid polarizer, the transmittance and the extinction ratio generally have a trade-off relationship. When trying to increase the transmittance, the extinction ratio decreases, and conversely, increasing the extinction ratio decreases the transmittance. This point is the same in the absorption type grid polarizer as in the embodiment. If a material having a large absorption with respect to light of the target wavelength is used, the extinction ratio is increased, but the overall transmittance is lowered. When a material with low absorption is used, the transmittance increases, but the extinction ratio decreases.

Therefore, the overall performance of the grid polarizing element (hereinafter referred to as PE) should be expressed by the product of the transmittance T and the extinction ratio ER. In this case, the extinction ratio ER has a large change due to parameters such as line width, gap width, and aspect ratio (hereinafter referred to as grid dimensions), so a common logarithm should be taken. PE has transmittance T and extinction ratio ER. Should be expressed as T × log ₁₀ (ER).

  As a polarizing element for ultraviolet light, grid polarizing elements for near ultraviolet light of 200 nm to 400 nm are sold by several companies. The inventor obtained the products of several companies that have been evaluated as almost the same performance as a grid polarizing element for near ultraviolet light, and measured the transmittance and the extinction ratio. The results are shown in FIGS. 3 and 4. FIGS. 3 and 4 are diagrams showing the results of examining the polarization performance of grid polarizers for near ultraviolet light sold by various companies.

3 and 4, the abscissa represents the extinction ratio represented by a common logarithm, and the ordinate represents the transmittance. As shown in FIGS. 3 and 4, the performances of the grid polarizers of each company are similar, although there is some variation in the combination of the extinction ratio and the transmittance. What is interesting here is that although the extinction ratio and the transmittance have a trade-off relationship, the performance of the grid polarizers of each company is plotted as shown in FIG. 3 by T × log ₁₀ (ER). For some reason I don't get on the line. Then, the inventor tried to write a line of T ² × log ₁₀ (ER), and it was found that the line almost entered as shown in FIG.

Where this result implies, in assessing the overall polarization performance of the grid polarizer, rather than evaluating the value of _{T × log 10 (ER),} T 2 × log value of 10 (ER) It is desirable to evaluate with. Based on this finding, the inventor further continued the research and found that the value of T ² × log ₁₀ (ER) (hereinafter, expressed as PE as the overall polarization performance) was practically 0.2 or more. It has turned out to be preferable.

PE ≧ 0.2 means that if the transmittance is 0.2 (20%), for example, the extinction ratio is 10 ⁵ or more. Conversely, if the extinction ratio is 10, for example, the transmittance is A transmittance of about √ (0.2) ≈0.45 (45%) is required.
The inventor has further researched on titanium oxide, zirconium oxide, hafnium oxide and yttrium oxide materials selected as grid material candidates for the vacuum ultraviolet light polarizing element, assuming such basic performance as a polarizing element. . Hereinafter, this point will be described.

  The extinction ratio and the transmittance can be obtained by simulation if the target wavelength, the optical constants (n, k) of the material, and the grid dimensions are known. In other words, by virtually determining the optical constant and the grid size, the extinction ratio and transmittance at each wavelength can be obtained, and what optical constant can be used to satisfy PE ≧ 0.2. FIG. 5 shows the result of conducting this study as a simulation experiment.

In this study, a grid size that is considered typical for a vacuum ultraviolet polarizing element was assumed. Specifically, the line width w = 20 nm, the grid height h = 100 nm, and the pitch p = 100 nm. Therefore, the aspect ratio (h / w) is 5, and the gap width is 80 nm.
In the simulation experiment whose results are shown in FIG. 5, the transmittance T and the extinction ratio ER were calculated on the premise that the grids having the above dimensions were used and various combinations were adopted while n and k were changed one after another. The calculation is based on the FDTD (Finite-Difference Time-Domain) method, and the software used is MATLAB (registered trademark) of Mathworks (Massachusetts, USA).

In various n and k combinations, n and k at which PE = T ² × log ₁₀ (ER) was 0.2 or more were examined. The result is shown in FIG. In FIG. 5A, the vertical axis represents the refractive index (real part) n, and the horizontal axis represents the wavelength. In FIG. 5B, the vertical axis represents the absorption coefficient a obtained from the extinction coefficient k, and the horizontal axis represents the wavelength. The extinction coefficient a is obtained by a = 4πk / λ (λ is a wavelength). In FIG. 5 (1), the line where PE = 0.2 is indicated by a broken line, and the line where PE is the maximum value is indicated by a solid line. Also in FIG. 5 (2), a line where PE = 0.2 is indicated by a broken line, and a line where PE is the maximum value is indicated by a solid line.

  The results shown in FIGS. 5A and 5B show that n can be used as a grid material of a vacuum ultraviolet light polarizing element if n is higher than a certain level at a wavelength of 200 nm or less and a is within a certain range. I have. Note that the value of a has an upper limit and a lower limit, but the polarization performance is not exhibited unless there is a certain amount of absorption, but if k is too large, the absorption increases and the transmittance becomes too small. Guessed.

  The inventors further studied based on the results shown in FIG. 5 and examined materials satisfying PE ≧ 0.2. FIG. 6 shows the result. FIG. 6 is a graph showing the values of n and a in the vacuum ultraviolet region for the oxides of the above-described Group 3 and Group 4 elements. Similarly, FIG. 6A shows wavelength vs. n (real part of refractive index), and FIG. 6B shows wavelength vs. a (extinction coefficient).

  As shown in FIG. 6, it can be seen that in the vacuum ultraviolet region, hafnium oxide and yttrium oxide have n and a satisfying PE ≧ 0.2. In the case of hafnium oxide, PE is less than 0.2 for a at about 180 nm or more. However, as will be described later, at 172 nm, which is important as the spectrum of vacuum ultraviolet light, exceeds PE = 0.2, it can be regarded as an effective grid material. With respect to titanium oxide, zirconium oxide, and lanthanum oxide, either n or a is below the line of PE = 0.2, which indicates that it is unsuitable as a vacuum ultraviolet grid material.

Therefore, from the results of the above experiments and investigations, it can be concluded that hafnium oxide and yttrium oxide are promising as grid materials in the vacuum ultraviolet region.
Such a grid material may be partially substituted with other elements for the purpose of improving workability and adjusting the refractive index. Also in this case, it is desirable not to fall below PE = 0.2. Hereinafter, this point will be described by taking hafnium oxide as an example.

FIG. 7 shows changes in n and k when a part of hafnium is replaced with silicon in hafnium oxide. (1) is a graph of photon energy vs. n, and (2) is a graph of photon energy vs. k. It is. FIG. 8 shows a change in n and k when a part of hafnium is replaced with aluminum in hafnium oxide. Similarly, (1) is a graph of photon energy vs. n, and (2) is a photon energy. 6 is a graph of energy versus k.
As shown in FIG. 7, as the replacement amount of silicon increases, both n and k decrease in the vacuum ultraviolet region. The substitution amount in this case is that the composition ratio _is the value of x in the _{Hf 1-x Si x O 2} .

In FIG. 7, a line of PE = 0.2 is added for evaluation. In FIG. 7A, when x = 0.6, the refractive index n falls below the line of PE = 0.2 when the photon energy is about 7 eV. Since there is a relationship of λ = 1240 / E between the photon energy E and the wavelength λ, this is a case where the wavelength is about 180 nm. Regarding the extinction ratio k, when x = 0.6, PE is less than 0.2 at about 7.8 eV. This corresponds to about 160 nm. Therefore, when x is 0.6 or more, PE becomes 0.2 or more on the shorter wavelength side than about 160 to 180 nm, and when hafnium is substituted in hafnium oxide, the composition ratio of silicon is 0.6 or less. It is preferable to do so. Further, when x = 0.4, the wavelength region where PE is 0.2 or more is more preferable because it extends to the longer wavelength side. As for hafnium silicate, there are cases where the oxidation number is 4 (HfSiO ₄ ) and 1 (HfSiO).

Also, in FIG. 8 showing the case of substitution with aluminum (the case of hafnium aluminate), a line of PE = 0.2 is added for evaluation. As shown in FIG. 8A, when the composition ratio x of aluminum is ３, PE is 0.2 when the photon energy is in the range of about 6.8 eV (≒ 182 nm) to 7.4 eV (≒ 168 nm). Surpass. Regarding the extinction coefficient k, when x is 1/3, when the photon energy is larger than about 7.2 eV (when the ≒ wavelength is shorter than about 172 nm), PE exceeds 0.2. Therefore, if x is set to 0.3 or less, it is estimated that PE becomes 0.2 or more in the range of about 180 to 150 nm for both n and k. That is, when substituting with aluminum, the composition is preferably set to 0.3 or less (Hf _1-x Al _x O ₂ , 0 ≦ x ≦ 0.3).
It should be noted that yttrium may be replaced with other elements by silicate or aluminate. However, it is preferable to make the addition ratio to achieve PE ≧ 0.2 in the vacuum ultraviolet region. From the viewpoint of obtaining

Further, in comparison between hafnium oxide and yttrium oxide, hafnium oxide has an advantage that workability is better. Transition metal oxides such as hafnium oxide and yttrium oxide are generally known as difficult-to-work materials because of their low volatility when they are converted to metal / halogen compounds and strong bonds between metal and oxygen. Nevertheless, hafnium oxide has been studied as a material for a gate insulating film in a semiconductor device, and can be etched by BCl ₃ plasma. In the future, if a hafnium oxide etching apparatus is developed as an apparatus for manufacturing semiconductor devices, it will be possible to divert it. On the other hand, it has been reported that yttrium oxide exhibits high resistance to fluorocarbon plasma, and its use as a protective film for a portion exposed to plasma in a plasma etching apparatus is also being studied. For this reason, it is assumed that the situation in which the workability is inferior to that of hafnium oxide will continue in the future.

Next, a method for manufacturing such a vacuum ultraviolet light polarizing element will be described.
FIG. 9 is a schematic view illustrating a method for manufacturing the vacuum ultraviolet light polarizing element of the embodiment. When manufacturing the vacuum ultraviolet light polarizing element of the embodiment, a process of forming a sacrificial layer as an intermediate structure is suitably adopted. FIG. 9 shows an example of this process.
When manufacturing the vacuum ultraviolet light polarizing element of the embodiment, first, a film 51 for a sacrificial layer is formed on the transparent substrate 1 (FIG. 9A). As the material of the sacrifice layer, a material having a high etching selectivity with respect to the grid material is suitably used, and for example, silicon is used as the material of the sacrifice layer. Various methods can be employed for forming the film 51 for the sacrificial layer. For example, plasma CVD is employed.

Next, a resist is applied on the sacrificial layer film 51 and patterned by photolithography to form a resist pattern 52. Since the resist pattern 52 is a manufacture of a grid polarizing element, it is striped (line and space). However, the pitch of the resist pattern 52 (indicated by p ′ in FIG. 9A) is twice the pitch of the final grid.
Next, the film 51 is etched using the resist pattern 52 as a mask, and then the resist pattern 52 is removed by ashing. Thus, a sacrificial layer 53 is formed as shown in FIG. The etching is anisotropic etching in a direction perpendicular to the transparent substrate 1. The sacrificial layer 53 is also striped and formed of a large number of linear portions extending in parallel.

  Next, a step of forming a film 54 for the grid is performed. The grid film 54 is formed on each side surface and each upper surface of each linear portion of the sacrificial layer 53. The film 54 is preferably formed by ALD (Atomic Layer Deposition). For example, when a hafnium oxide film is formed as the film 54, TEMAH (tetrakisethylmethylaminohafnium) is used as the precursor gas, and water (water vapor) is used as the oxidizing agent. The temperature of the susceptor on which the transparent substrate 1 is placed is set to about 200 to 400 ° C. (for example, 250 ° C.), and water vapor and a precursor that has been heated to about 75 to 95 ° C. in the chamber at pulse intervals of 200 to 500 milliseconds. To form a hafnium oxide film. The pressure in the chamber is about 100 mTorr to 500 mTorr. Ozone may be introduced as an oxidant. Nitrogen or argon is used as the carrier gas or purge gas. As shown in FIG. 9C, the grid film 54 is formed on each side surface and each upper surface of each linear portion of the sacrificial layer 53.

After forming the film 54 in this way, as shown in FIG. 9D, the film 54 is partially etched. “Partial” refers to etching that removes only the portion of each sacrificial layer 53 that is placed on the upper surface and the portion that is directly deposited on the transparent substrate 1 (the bottom of the gap). As described above, this etching is performed by BCl ₃ plasma etching in the case of hafnium oxide. For example, partial etching of the film 54 is performed by ECR plasma or IC (capacitive coupling) plasma of BCl ₃ using argon as a buffer gas. At this time, a substrate bias is applied to set an electric field perpendicular to the transparent substrate 1 and anisotropic etching is performed. This is in order not to etch the portion deposited on each side surface of the sacrificial layer 53. In some cases, plasma etching is performed by adding oxygen gas or chlorine gas to BCl ₃ gas. Thus, each linear portion 3 constituting the grid 2 is formed.

Thereafter, etching for removing the sacrificial layer 53 is performed. At this time, only the material of the sacrificial layer 53 is selectively etched. For example, when the sacrificial layer 53 is silicon, only the sacrificial layer 53 can be selectively removed by plasma etching using a gas such as CF ₄ . By removing the sacrificial layer 53, the vacuum ultraviolet light polarizing element of the embodiment is completed as shown in FIG. The pitch p of each linear portion 3 in the completed polarizing element is half the pitch p ′ of the resist pattern.

  In the above-mentioned manufacturing method, the height of the sacrificial layer 53 formed in the middle determines the final height of the grid, and therefore requires a particularly high precision. Further, the aspect ratio of the sacrifice layer 53 is a factor that determines the aspect ratio of the grid. To increase the aspect ratio, the sacrifice layer 53 also needs to have a high aspect ratio. For this reason, a film of carbon or the like is formed as a mask layer on the sacrificial layer film 51 and patterned by photolithography, and the sacrificial layer film 51 is etched using the mask layer as a mask. is there. Since the mask itself has a high aspect ratio, it can withstand long-time anisotropic etching, and the sacrificial layer 53 having a uniform height can be formed.

Such a vacuum ultraviolet light polarizing element is suitably used for applications of optical alignment. Hereinafter, this point will be described.
FIG. 10 is a schematic front view of an optical alignment device equipped with the vacuum ultraviolet light polarizing element of the embodiment. The photo-alignment device shown in FIG. 10 is a device for obtaining a photo-alignment layer for a liquid crystal display. This is a device for forming a layer.

  This apparatus includes a lamp house 6 including a light source 61 that emits vacuum ultraviolet light, and a work transfer system 7 that transfers the work 10 to a vacuum ultraviolet light irradiation region R. As the light source 61, an excimer lamp, a low-pressure mercury lamp, or the like can be used. In particular, an excimer lamp is a lamp that emits light that can be regarded as having a single wavelength, and is preferable because it does not unnecessarily heat the work 10 or cause a reaction. For example, an excimer lamp having a wavelength of 172 nm in which xenon is sealed as a discharge gas is used.

Note that a mirror 62 is arranged behind the light source 61. Since the light source 61 has a rod shape elongated in the direction perpendicular to the paper surface, the mirror 62 has a substantially gutter shape. A mechanism for forcibly cooling the light source 61 and the mirror 62 may be provided.
The vacuum ultraviolet light polarizing element 8 is mounted on the light emitting side of the lamp house 6. For example, the vacuum ultraviolet light polarization element 8 is mounted on the frame 81 by being held as a unit and fitted into the light exit of the lamp house 6.

  The work 10 has a transparent plate shape in this example, and is placed on the stage 71 and transported. Therefore, the work transfer system 7 includes a mechanism for transferring the stage 71 through the irradiation area R. As the work 10, a work whose surface is coated with a film material to be a photo-alignment layer may be used. The work 10 is conveyed so as to pass through the irradiation region R, and at the time of conveyance, the work 10 is irradiated with polarized light of vacuum ultraviolet light to be subjected to optical alignment processing. The work transfer system 7 includes a linear guide 72 for guiding the linear movement of the stage 71, a linear drive source (not shown), and the like.

Note that the inside of the lamp house 6 may be purged with nitrogen gas in order to suppress absorption of vacuum ultraviolet light. Nitrogen gas may be flowed for the purpose of cooling the vacuum ultraviolet light polarizing element 8 or preventing adhesion of foreign substances such as siloxane to the vacuum ultraviolet light polarizing element 8.
The irradiation distance (indicated by L in FIG. 10) from the vacuum ultraviolet light polarizing element 8 to the work 10 is preferably about 1 to 40 mm. If it is longer than 40 mm, the illuminance may be reduced to a limit or less due to absorption of vacuum ultraviolet light by the atmosphere (air). If it is shorter than 1 mm, there arises a problem that extremely high accuracy is required for the transfer position by the work transfer system 7.

The irradiation region R longer than the width of the work 10 (the length in the direction perpendicular to the paper surface of FIG. 10) is irradiated with the polarized light of the vacuum ultraviolet light. It is determined by the length, the speed of passing through the irradiation area R, and the illuminance. It is preferable that the irradiation amount is about 40 mJ / mm ^{2 to} 4000 mJ / mm ² . If it is less than 40 mJ / mm ² , the amount of irradiation may be insufficient and the photo-alignment may be insufficient. When it is more than 4000 mJ / mm ² , the workpiece 10 may be deteriorated by high energy of vacuum ultraviolet light.

In the above embodiment, as the structure of the vacuum ultraviolet light polarizing element, a structure in which an antireflection layer or a protective layer is formed on the incident side of the grid 2 may be used. For example, a silicon oxide layer may be formed as a protective layer so as to cover the grid 2. The protective layer may be provided in consideration of adhesion of foreign substances such as siloxane, and the protective layer is provided so that the foreign substances can be removed by a method such as wiping.
Moreover, about a photo-alignment apparatus, a sheet-like film | membrane material may become a workpiece | work. In this case, a mechanism for conveying a workpiece by a roll-to-roll conveyance method may be employed as the workpiece conveyance system.

DESCRIPTION OF SYMBOLS 1 Transparent substrate 2 Grid 3 Linear part 4 Gap 53 Sacrificial layer 6 Lamp house 61 Light source 7 Work transfer system 8 Vacuum ultraviolet light polarizing element

Claims (3)

A vacuum ultraviolet light polarizing element that polarizes vacuum ultraviolet light having a wavelength of 200 nm or less,
A substrate transparent to vacuum ultraviolet light, and a grid provided on the substrate,
The grid is made up of a number of parallel linear parts,
It is a structure where no filler is provided between each linear part,
The material of each linear portion is an oxide of an element belonging to Group 3 or Group 4 and has a formula of PE = T ² × log ₁₀ (ER) (where T is transmittance through a grid, and ER is transmittance through a grid). A vacuum ultraviolet light polarizing element comprising a material having a PE of 0.2 or more in a combination in which PE obtained by extinction ratio is highest in a vacuum ultraviolet region.   The vacuum ultraviolet light polarization element according to claim 1, wherein the vacuum ultraviolet region includes a wavelength of 172 nm.   In the material forming each of the linear portions, a part of the element of the third or fourth group is replaced by another element, and the substitution ratio is determined in a combination in which the PE is highest in a vacuum ultraviolet region. 3. The vacuum ultraviolet light polarizing element according to claim 1, wherein the PE is 0.2 or less.

Images