TW201939741A - Composite structure, semiconductor manufacturing apparatus and display manufacturing apparatus provided with composite structure - Google Patents

Abstract

Disclosed is provision of a ceramic coat having an excellent low-particle generation as well as a method for assessing the low-particle generation of the ceramic coat. A composite structure including a substrate and a structure which is formed on the substrate and has a surface, wherein the structure includes a polycrystalline ceramic and the composite structure has luminance Sa satisfying a specific value calculated from a TEM image analysis thereof, can be suitably used as an inner member of a semiconductor manufacturing apparatus required to have a low-particle generation.

Description

Composite structure, evaluation method, semiconductor manufacturing device and display manufacturing device including composite structure

The present invention relates to a composite structure in which polycrystalline ceramics are coated on the surface of a substrate to impart a function to the substrate. The present invention also relates to a semiconductor manufacturing apparatus and a display manufacturing apparatus including the composite structure. In particular, the present invention relates to a composite structure having excellent particle resistance used in a semiconductor manufacturing device member or the like exposed to a corrosive plasma environment, an evaluation method, and a semiconductor manufacturing device provided with the composite structure. And display manufacturing equipment.

A technique is known in which a ceramic is coated on the surface of a substrate to impart a function to the substrate. Examples of such ceramic coatings include ultra-smooth coatings such as plasma-resistant coatings for structural components of reaction chambers in semiconductor manufacturing equipment, insulating coatings in heat-dissipating substrates, and optical mirrors. Scratch resistance, wear resistance, etc. in coatings, sliding members, and the like. Along with the high level of functionality of such components, the level of requirements for such components is high. In this type of ceramic coating, not only the material composition but the physical structure, especially the microstructure, often dominates its performance.

As a method for obtaining such a ceramic coating, an aerosol deposition method (AD method), and a PVD (Physical Vapor) thickened by plasma or ion-assisted have been developed. Deposition: physical vapor deposition (PEPVD) method (Plasma-Enhanced Physical Vapor Deposition) method, IAD (Ion Assisted Deposition) method, suspension using fine raw materials (suspension) Various ceramic coating technologies such as suspension spraying method.

Ceramic coatings that are carefully manufactured by these methods are also controlled to some degree with microstructure. In the reports so far, the porosity confirmed by an image analysis method such as SEM (Scanning Electron Microscope) is 0.01 to 0.1%.

For example, Japanese Patent Application Laid-Open No. 2005-217351 (Patent Document 1) discloses that a layered structure made of yttria polycrystalline having a hole occupancy ratio of 0.05 area% or less is used as a member for a semiconductor manufacturing device having plasma resistance. Structure. This layered structure has appropriate plasma resistance.

Furthermore, Korean Patent 20170077830A Publication (Patent Document 2) discloses a YF ₃ transparent fluorine-based film having high resistance to plasma and corrosive gases. Since this YF ₃ film has a dense porosity of 0.01-0.1%, it has high resistance to plasma and the like. Moreover, the withstand voltage is 50-150V / μm.

Japanese Patent Publication No. 2016-511796 (Patent Document 3) discloses a ceramic coating film of Y ₂ O ₃ and the like including constituent particles having a particle diameter in a range of _{200 to} 900 nm and constituent particles having a particle diameter in a range of 900 nm to 10 μm. The coating film has a dense porosity of 0.01-0.1%, and has high resistance to plasma and the like. Moreover, the withstand voltage is 80-120V / μm.

In the Journal of the Chemical Society of Japan, 1979, (8), p. 1106 to 1108 (Non-Patent Document 1), the optical characteristics of yttrium oxide sintered bodies as transparent plate-shaped samples reveal the refractive index and reflectance (see FIG. 10). .

In the field of semiconductor manufacturing equipment, the miniaturization of semiconductor devices has been promoted year by year. If EUV (Extreme ultraviolet lithography) is put into practical use, it is estimated that it can reach several nm. According to the IEEE (The Institute of Electrical and Electronics Engineers, Inc: Institute of Electrical and Electronics Engineers) IRDS (International Roadmap for Devices and Systems: International Device and Systems Development Roadmap), the post-Moore White Paper 2016 version (MORE MOORE WHITE PAPER 2016EDITION) It is predicted that although the half-pitch (half pitch) between devices in 2017 is 18.0nm, it will be reduced to 12.0nm in 2019 and to less than 10.0nm after 2021.

Furthermore, the circuit line width and the circuit pitch have been reduced for the purpose of increasing the integration of such semiconductors. Further, for example, a corrosive plasma such as a fluorine-based plasma such as CF ₄ or NF ₃ or a chlorine-based plasma is used in the etching process. In addition, in the future, treatments using high-density plasmas will be carried out, and various components in semiconductor manufacturing devices will be required to have a higher level of particle resistance (low-particle generation).

In the past, it has been thought that the plasma resistance of ceramic coatings is related to their porosity. As long as the consumption of ceramic coatings caused by plasma erosion is suppressed, the generation of fine particles can be suppressed. According to this idea, the porosity of the structure can be made as small as 0.01, for example. ~ 0.1% solves the problem caused by particles. However, according to the knowledge obtained by the present inventors, when the miniaturization is further advanced, even if the structure has a very small porosity, the problem of suppressing the generation of fine particles cannot be solved. That is to say, it is understood that not only the consumption of the ceramic coating layer based on the porosity index, but also the generation of fine particles must be controlled with higher precision from another viewpoint.

In other words, in recent years, in the miniaturization of components and the increase in the density of plasma, even for ceramic structures with porosity of 0.01 to 0.1%, which contains almost no pores, the problem of fine particles still exists. The structure is required. In addition, in the future, semiconductor devices having a line width of several nanometers and fine elements are required to have a ceramic structure capable of solving the problem of fine particles.

[Patent Document 1] Japanese Patent Publication No. 2005-217351 [Patent Document 2] Korean Patent Publication 20170077830A [Patent Document 3] Japanese Patent Publication No. 2016-511796

Non-Patent Document 1: Journal of the Chemical Society of Japan 1979, (8), p. 1106 to 1108 [Refractive index and reflectance of yttrium oxide sintered body]

The present inventors have succeeded in obtaining a composite structure this time. For example, in a ceramic coating used in a semiconductor manufacturing device or the like exposed to a corrosive plasma environment, the influence of particles can be extremely reduced. In addition, several indexes were found to have a high correlation with the particle resistance performance at a very high level. In addition, it was successfully made: with these indexes, specifically, in accordance with the first to fifth aspects described later Structure with excellent particle resistance specified by the index.

Therefore, an object of the present invention is to provide a composite structure including a polycrystalline ceramic structure with a controlled microstructure, and more particularly, to provide a substrate with a substrate capable of solving the problem of high-density and high-density plasma. A composite structure of a polycrystalline ceramic structure that can also suppress the generation of particles.

It is another object of the present invention to provide a semiconductor manufacturing apparatus and a display manufacturing apparatus including the composite structure .

Composite Structure The basic structure of a composite structure according to the present invention will be described using FIG. FIG. 1 is a schematic cross-sectional view of a composite structure 100 according to the present invention. The structure 10 is disposed on the surface 70 a of the substrate 70. The structure 10 includes a surface 10a. The surface 10 a is a surface exposed to the environment in an environment that requires physical properties and characteristics imparted by the structure 10 to the composite structure. Therefore, for example, when the composite structure according to the present invention is a composite structure having physical properties and characteristics imparted to the particles by the structure 10, the surface 10 of the composite structure is a surface exposed to a corrosive gas such as plasma. In the present invention, the structure 10 includes a polycrystalline ceramic. In addition, the structure 10 provided in the composite structure according to the present invention displays a predetermined value under an index in the first to fifth aspects described later.

The structure 10 provided in the composite structure according to the present invention is a so-called ceramic coating. By applying a ceramic coating, 70 kinds of physical properties and characteristics can be imparted to a substrate. In addition, in this specification, a structure (ceramic structure) and a ceramic coating are used synonymously unless otherwise specified.

According to an aspect of the present invention, the structure 10 includes a polycrystalline ceramic as a main component. The content of the polycrystalline ceramic is 70% or more, preferably 90% or more, and more preferably 95% or more. Most preferably, the structure 10 is made of 100% polycrystalline ceramic.

Furthermore, according to one aspect of the present invention, although the structure 10 may include a polycrystalline region and an amorphous region, it is more preferable that the structure 10 is composed of only a polycrystal.

Method for measuring crystallite size In the present invention, the size of the polycrystalline ceramic constituting the structure 10 is an average crystallite size obtained under the following measurement conditions of 3 nm to 50 nm. The upper limit is more preferably 30 nm, even more preferably 20 nm, and still more preferably 15 nm. Moreover, a preferable lower limit thereof is 5 nm. In the present invention, the [average crystallite size] is a transmission electron microscope (TEM) image taken at a magnification of 400,000 or more. In this image, 15 crystallites are formed in an approximately circular shape. The average value of the diameter is calculated. At this time, the thickness of the sample during the processing of a focused ion beam (FIB: Focused Ion Beam) was sufficiently thin to about 30 nm. This allows the crystallites to be more clearly identified. The photography magnification can be appropriately selected in the range of more than 400,000 times. FIG. 11 is an example of a TEM image for measuring crystallite size. Specifically, FIG. 11 is a TEM image of the structure 10 at a magnification of 2 million times. The area indicated by 10c in the figure is a microcrystal.

The ceramics constituting the structure 10 may be appropriately determined according to the physical properties and characteristics desired to be imparted to the substrate 70 as described above, and may be a metal oxide, a metal fluoride, a metal nitride, a metal carbide, or a mixture thereof. According to one aspect of the present invention, as the compound having excellent resistance to fine particles, oxides, fluorides, acid fluorides (LnOF) of rare earth elements, or mixtures thereof can be cited as materials. More specifically, examples of the rare earth element Ln include Y, Sc, Yb, Ce, Pr, Eu, La, Nd, Pm, Sm, Gd, Tb, Dy, Ho, Er, Tm, and Lu.

In the case of the insulating structure 10, Al ₂ O ₃ , ZrO ₂ , AlN, SiC, Si ₃ N ₄ , cordierite, forsterite, and mullite ( mullite), silica, and other materials.

In the present invention, the film thickness of the structure 10 may be appropriately determined in consideration of required applications, characteristics, film strength, and the like. Generally, it is in the range of 0.1 to 50 μm, and the upper limit may be, for example, 20 μm, 10 μm, or 5 μm, and may be 1 μm or less. Here, the film thickness of the structure 10 can be confirmed, for example, by cutting the structure 10 and performing SEM observation of the fracture surface.

In the present invention, the substrate 70 is an object to which a function is provided by the structure 10, and it may be appropriately determined. Examples of the material include ceramics, metals, and resins, and may also be composites thereof. Examples of the composite include a composite substrate of resin and ceramic, a composite substrate of fiber reinforced plastics and ceramic, and the like. The shape is not particularly limited, and it may be a flat plate, a concave surface, a convex surface, or the like.

According to an aspect of the present invention, it is preferable that the surface 70 a of the substrate 70 bonded to the structure 10 is smooth to form a good structure 10. According to one aspect of the present invention, the surface 70a of the substrate 70 is subjected to at least any one of, for example, blast, physical polishing, chemical mechanical polishing, lapping, and chemical polishing, and the surface 70a is removed. Of bumps. Such unevenness is removed so that, for example, the subsequent surface 70a has a two-dimensional arithmetic mean roughness (Ra) of 0.2 μm or less, more preferably 0.1 μm or less, or a two-dimensional arithmetic average height (two-dimensional Arithmetic mean height) Rｚ is preferably performed so that it is 3 μm or less.

The composite structure according to the present invention can be used as various components in a semiconductor manufacturing apparatus, and is particularly suitable as a component used in an environment exposed to a corrosive plasma environment. As described above, the components inside the semiconductor manufacturing apparatus are required to be resistant to particles. This is because the structure including the polycrystalline ceramic included in the composite structure according to the present invention has high particle resistance.

When the composite structure according to the present invention is used as a member used in an environment exposed to a corrosive plasma environment, the components of the ceramic constituting the structure 10 include Y ₂ O ₃ and yttrium oxyfluoride ( yttrium oxyfluoride) (YOF, Y ₅ O ₄ F ₇ , Y ₆ O ₅ F ₈ , Y ₇ O ₆ F ₉ and Y ₁₇ O ₁₄ F ₂₃ ), (YO _0.826 F _0.17 ) F _1.174 , YF ₃ , Er ₂ O ₃ , Gd ₂ O ₃ , Nd ₂ O ₃ , Y ₃ Al ₅ O ₁₂ , Y ₄ Al ₂ O ₉ , Er ₃ Al ₅ O ₁₂ , Gd ₃ Al ₅ O ₁₂ , Er ₄ Al ₂ O ₉ , ErAlO ₃ , Gd ₄ Al ₂ O ₉ , GdAlO ₃ , Nd ₃ Al ₅ O ₁₂ , Nd ₄ Al ₂ O ₉ , NdAlO ₃ and the like.

Plasma resistance The composite structure according to the present invention is provided with the plasma resistance by the below-mentioned indicators defined in accordance with the first to fifth aspects. This plasma resistance performance was evaluated using the plasma resistance test described below as a reference method. This plasma resistance test is hereinafter referred to as a "reference plasma resistance test" in this specification.

As an aspect of the present invention, a composite structure having an arithmetic average height Sa of the surface 10a of the structure 10 after the [reference plasma resistance test] is 0.060 or less, and more preferably a composite structure of 0.030 or less. The arithmetic mean height Sa is described later.

As the plasma etching apparatus used for the [reference plasma resistance test], inductively coupled plasma reactive ion etching (Muc-21 Rv-Aps-Se / Sumitomo Precision Industry Co., Ltd.) was used. The conditions for plasma etching are power output. The output of ICP (Inductively Coupled Plasma) is 1500W, the bias output is 750W, and the process gas is CHF ₃ gas 100ccm and O ₂ gas 10 ccm mixed gas, the pressure is 0.5 Pa, and the plasma etching time is 1 hour.

The state of the surface 10a of the structure 10 after the plasma irradiation is photographed with a laser microscope (for example, OLS4500 / Olympus). The details of the observation conditions and the like will be described later.

The arithmetic mean height Sa of the surface after plasma irradiation was calculated from the obtained SEM image. Here, the arithmetic mean height Sa refers to a roughness that expands the two-dimensional arithmetic mean roughness Ra three-dimensionally, and is a three-dimensional roughness parameter (three-dimensional height direction parameter). Specifically, the arithmetic mean height Sa is the volume of the portion surrounded by the surface shape curved surface and the average surface divided by the measurement area. That is, if the average plane is the xy plane, the longitudinal direction is the ｚ axis, and the measured surface shape curve is ｚ (x, y), the arithmetic mean height Sa is defined by the following formula. Here, [A] in the formula (1) is a measurement area. [Formula 1]

Although the arithmetic mean height Sa is basically a value that does not depend on the measurement method, it is calculated under the following conditions in the "reference plasma resistance test" in this specification. The calculation of the arithmetic mean height Sa is performed using a laser microscope. Specifically, a laser microscope [OLS4500 / Olympus] was used. The objective lens uses MPLAPON100xLEXT (numerical aperture) 0.95, working distance 0.35mm, focal spot diameter 0.52 μm, and measurement area 128 × 128 μm), and the magnification is 100 times. The λc filter from which the moiré component was removed was set to 25 μm. The measurement was performed at three arbitrary locations, and the average value was used as the arithmetic mean height Sa. In addition, it is appropriate to refer to the international standard for three-dimensional surface properties ISO25178.

The first aspect of the present invention serves as the basis of the first aspect of the present invention. The present inventors have succeeded in the following aspects: For example, a semiconductor manufacturing apparatus and the like are used in a state exposed to a corrosive plasma environment. The effect of particles is extremely reduced in the composite structure of the crystalline ceramic structure. Furthermore, it has been found that by using the amount of hydrogen contained in a structure as an index measured by a secondary ion mass spectrometry (Dynamic-Secondary Ion Mass Spectrometry_D-SIMS method), it is possible to evaluate the resistance to particles at an extremely high level.

A composite structure according to a first aspect of the present invention is a composite structure including a substrate and a structure provided on the substrate and having a surface. The structure includes a polycrystalline ceramic. The number of hydrogen atoms per unit volume in a measurement depth of 500 nm or 2 μm measured by mass spectrometry (Dynamic-Secondary Ion Mass Spectrometry_D-SIMS method) is 7 * 10 ²¹ atoms / cm ³ or less.

Fig. 18 is a schematic sectional view for explaining the microstructure of the composite structure. In FIG. 18, (a) is a conventional composite structure 110, and (b) is a composite structure 100 related to the present invention. In the figure, 80 indicates a sparse structure at a nano level, and 90 indicates a water molecule (OH group). In FIG. 18, for the sake of easy understanding, although the size of the nano-sized sparse structure 80 is increased, in practice, any of the composite structures 100 and 110 is a porosity by a conventional evaluation method such as SEM. It is 0.01 to 0.1%.

The inventors of the present invention focused on the fact that even a structure having a porosity as low as 0.01 to 0.1% still cannot solve the problem of fine particles, and successfully obtained a novel structure that can solve the problem of fine particles at a higher level. As a reason why the problem of fine particles cannot be solved by the conventional structure, it is considered that the microstructure in the structure has nano-scale unevenness, and the sparse structure 80 has a lower plasma resistance than the dense structure. Furthermore, it is considered that, for example, water molecules (OH groups) included in the atmosphere are present in the slightly sparse structure 80 at the nanometer level, and the correlation with the particle resistance is obtained by performing the quantification. In other words, it has been found that by quantifying the amount of hydrogen in water molecules (OH groups), it is possible to specify a structure of a novel structure that can solve the problem of fine particles at a higher level.

Specifically, in FIG. 18, it is considered that the composite structure 100 of the present invention has fewer sparse structures 80 and fewer water molecules (OH groups) existing in the sparse structure 80 than the conventional composite structure 110. The amount of hydrogen (the number of hydrogen atoms per unit volume) of the structure 10 can be quantified by the secondary ion mass spectrometry (D-SIMS method) described later, and it can be correlated with the particle resistance.

The sample for measuring the amount of hydrogen in the first aspect of the invention In the first aspect of the present invention, the sample for measuring the amount of hydrogen can be prepared by the following method, for example.

First, a composite structure including the structure 10 is cut out by a cutting machine or the like in advance. At this time, a portion corresponding to the sample obtaining portion 40 of FIGS. 8 and 9 is cut out. Although the size is arbitrary, it is, for example, 3 mm × 3 mm to 7 mm × 7 mm and a thickness of about 3 mm. In addition, the thickness of the sample may be appropriately determined according to a measurement device or the like to be used, and is adjusted by cutting off a surface or the like on the side where the structure 10 is not formed in the base material 70. The surface 10a of the structure 10 is ground so that the arithmetic mean roughness Ra of the two-dimensional surface roughness parameter becomes 0.1 μm or less, and more preferably 0.01 μm. The thickness of the structure 10 is at least 500 nm, preferably 1 μm or more, and more preferably 3 μm or more.

In the D-SIMS method of the hydrogen content measurement method in the present invention, measurement is generally performed using a standard sample. As the standard sample, a sample having the same composition and the same structure as the sample to be measured is preferably used. For example, at least two samples are prepared, and one of them can be used as a standard sample. The details of the standard sample will be described later.

The state of the sample before the hydrogen content measurement is described.

As described above, in the present invention, the specific method for the nanometer-level sparse structure of the structure 100 uses the amount of hydrogen. Therefore, it is important to manage the sample before measuring the hydrogen amount for a predetermined time in a constant temperature and humidity chamber. Specifically, in the present invention, the amount of hydrogen is measured after the sample is left at room temperature for 20-25 ° C., humidity of 60% ± 10%, and atmospheric pressure for more than 24 hours.

Measurement of the amount of hydrogen in the first aspect of the present invention Next, a method for measuring the amount of hydrogen in the first aspect of the present invention will be described.

In the present invention, the amount of hydrogen is measured by secondary ion mass spectrometry: Dynamic-Secondary Ion Mass Spectrometry (D-SIMS method). As the device, for example, IMF-7f manufactured by CAMECA is used.

Next, the measurement conditions are described.

First, conductive platinum (Pt) is deposited on the surface of the structure. As a measurement condition, a cesium (Cs) ion was used as a primary ion. The primary acceleration voltage was 15.0 kV, and the detection area was 8 μmφ. The measurement depth was 500 nm and 2 μm.

The resistance to particles depends greatly on the properties of the surface of the structure directly exposed to the plasma environment. Therefore, by quantifying the amount of hydrogen in a region having a depth of at least about 500 nm from the surface in the structure, the amount of hydrogen and the resistance to particulates can be effectively connected. On the other hand, when the thickness of the structure is very large and the microstructure from the surface to the depth of the measurement object is slightly homogeneous, the reliability of the quantitative result can be improved by using the area up to about 2 μm as the measurement object. ). In addition, the structure 10 has a lower region 10b on the substrate 70 side and an upper region 10u on the surface 10a side (see FIG. 1). For example, in the upper region 10u, it is considered to be a laminated structure having higher particle resistance than the lower region 10b In the case where the measurement depth is set so that the amount of hydrogen in only the upper region 10u can be quantified. From this viewpoint, the amount of hydrogen specified in the present invention may be any one of the measurement depths of 500 nm and 2 μm. The amount of hydrogen that satisfies the present invention is preferably at any one of the measurement depths of 500 nm and 2 μm.

The measurement of the amount of hydrogen is to prepare a measurement sample and a standard sample.

The standard sample is designed to cancel the factors related to the measurement conditions. In order to normalize the signal intensity of the ion species to be analyzed based on the signal intensity of the ion species of the matrix element of the sample, the standard The SIMS method is used. More specifically, using: a standard sample of evaluation sample Evaluation sample, equivalent to the matrix elements having an evaluation sample (matrix component) with a sample, and the Si single crystal, Si single crystal and the criteria used Sample. The standard sample for the evaluation sample refers to a sample in which deuterium is injected into a sample having a matrix component equivalent to the evaluation sample. At this time, heavy hydrogen is also injected into the Si single crystal, and it is assumed that the equivalent heavy hydrogen is injected into the standard sample and the Si single crystal for the evaluation sample. Then, the amount of deuterium injected into the Si single crystal was identified using a standard sample for the Si single crystal. For the standard sample used for the evaluation sample, the secondary ion mass intensity of the heavy hydrogen and the constitutive element was calculated using the secondary ion mass spectrometry (D-SIMS method), and the relative sensitivity coefficient was calculated. ). The hydrogen content of the evaluation sample was calculated using the relative sensitivity coefficient calculated from the standard sample for the evaluation sample. For other, please refer to ISO 18114_ "Determining relative sensitivity factors from ion-implanted reference materials" (International Organization for Standardization, Geneva) , 2003).

In the present invention, each of the unit volumes included in the structure constituting the composite structure has a measurement depth of 500 nm or 2 μm or less as measured by a Dynamic-Secondary Ion Mass Spectrometry_D-SIMS method The number of hydrogen atoms is 7 * 10 ²¹ atoms / cm ³ or less.

In the structure of the present invention, the surface is directly exposed to a plasma environment, for example. Therefore, in particular, the properties of the structure surface become important. The present inventors have considered the results of reviewing the results of the inability to solve the problem of particles due to the influence of nanometer-level microstructures even on structures with almost no pores having a porosity of 0.01 to 0.1%, and successfully controlled the nanometers. Advanced microstructure, a novel structure that can solve the problem of particles at a higher level is obtained. Furthermore, it was newly discovered that the structure of the novel structure can be specified using the amount of hydrogen on the surface (the number of hydrogen atoms per unit volume) as an index. Furthermore, the present invention was conceived by finding the correlation between the amount of hydrogen on the structure surface and the resistance to fine particles.

Specifically, for example, in the atmosphere, it is considered that a hydrogen atom exists in a state such as a hydroxyl group (-OH). The size of the molecule is about 3 Å for water molecules and about 1 Å for hydroxyl groups, and it is considered to be the aforementioned sparse structure 80 slightly existing in the structure. By using this amount of hydrogen as an index, a nano-scale microstructure can be displayed.

In the present invention, as the amount of hydrogen of the structure, the number of hydrogen atoms per unit volume in a measurement depth of 500 nm or 2 μm measured by D-SIMS is 7 * 10 ²¹ atoms / cm ³ or less. The number of hydrogen atoms per unit volume is preferably 5 * 10 ²¹ atoms / cm ³ or less.

In addition, although it is considered that the smaller the amount of hydrogen in the structure of the composite structure according to the present invention is, the better and better, it is obvious to those skilled in the art that there are also practical measurement limits. Therefore, the lower limit of the amount of hydrogen in this aspect is taken as the measurement limit. This point is the same in the second aspect below.

Although the second aspect of the present invention uses the amount of hydrogen as an indicator in the second aspect of the present invention as the first aspect of the present invention, the second aspect of the present invention uses a forward scattering spectrum by hydrogen. Method (HFS: Hydrogen Forward-Scattering Spectroscopy) -Raseford Backscattering Spectrometry (RBS-HFS method) and proton-hydrogen forward scattering spectroscopy (p-RBS method) Volume as an indicator. That is, the present inventors have succeeded in a composite structure provided with a structure containing Y (yttrium element) and O (oxygen element) used in a semiconductor manufacturing device or the like exposed to a corrosive plasma environment, for example. Minimize the effects of particles. It was also found that by using the forward hydrogen scattering method (HFS) -Rasefort backscattering method (RBS) (RBS-HFS method) and the proton-hydrogen forward scattering method (p-RBS method) The amount of hydrogen contained in the structure measured as an index can be used to evaluate the resistance to particulates at an extremely high level.

A composite structure according to a second aspect of the present invention is a composite structure including a substrate and a structure disposed on the substrate and having a surface. The structure includes a polycrystalline ceramic, and is forwarded by hydrogen. The hydrogen atom concentration measured by scattering spectrometry (HFS) -Rasefort backscattering spectroscopy (RBS) (RBS-HFS method) and proton-hydrogen forward scattering spectroscopy (p-RBS method) is 7 atomic% or less.

The second aspect of the present invention is different from the first aspect of the present invention in the method for measuring the amount of hydrogen. Except for cases where changes are caused due to different changes, the description of the first aspect in this specification becomes the first aspect. Description of the second invention.

Sample for measuring hydrogen amount in the second aspect of the present invention In the second aspect of the present invention, the sample for measuring the amount of hydrogen can be prepared, for example, by the following method.

First, a composite structure including the structure 10 is cut out by a cutting machine or the like in advance. At this time, a portion corresponding to the sample obtaining portion 40 of FIGS. 8 and 9 is cut out. Although the size is arbitrary, it is, for example, 20 mm × 20 mm and a thickness of about 5 mm. In addition, the thickness of the sample may be appropriately determined according to a measurement device or the like to be used, and is adjusted by cutting off a surface or the like on the side where the structure 10 is not formed in the base material 70. The surface 10a of the structure 10 is ground so that the arithmetic mean roughness Ra of the two-dimensional surface roughness parameter is 0.1 μm or less, and more preferably 0.01 μm. The thickness of the structure 10 is at least 500 nm, preferably 1 μm or more, and more preferably 3 μm or more.

The state of the sample before the hydrogen content measurement is described.

As described above, in the present invention, the amount of hydrogen is used in a specific method of nanoscale sparse structure of the structure 100. Therefore, it is important to manage the sample before measuring the hydrogen amount for a predetermined time in a constant temperature and humidity chamber. Specifically, in the present invention, the amount of hydrogen is measured after the sample is left at room temperature for 20-25 ° C., humidity of 60% ± 10%, and atmospheric pressure for more than 24 hours.

Measurement of the amount of hydrogen in the second aspect of the present invention Next, a method for measuring the amount of hydrogen will be described.

In the present invention, the measurement system for the amount of hydrogen is a combination of: Hydrogen Forward scattering Spectrometry (HFS) method / Rutherford Backscattering Spectorometry (RBS) method (hereinafter referred to as RBS-HFS method) and the RBS method (hereinafter referred to as p-RBS) using a proton. As the device, for example, Pelletron 3SDH manufactured by National Electrostatics Corporation can be used.

The method for quantifying the amount of hydrogen will be further described.

The RBS-HFS method using a helium (He) element was implemented. The structure is irradiated with helium (He atom), and He atoms that are backward scattered and H atoms that are forward scattered are detected. Regarding the energy spectrum of the back-scattered He atom, fitting is performed sequentially from the element with the largest energy spectrum, and the scattering intensity is calculated. Furthermore, the The energy spectrum is also fitted to calculate the scattering intensity. From the calculated respective scattering intensities, the ratio of the average number of atoms of the elements in the structure can be calculated. For example, when the structure is yttrium oxide, fitting of the Y element with the largest energy spectrum of the detected He atom is performed to calculate the scattering intensity, and then fitting of the O element is performed to calculate the scattering intensity. In addition, for the identification of the element with the largest energy spectrum, other methods such as a combined energy dispersive X-ray analysis (EDX) method are preferred.

Although the ratio of the average atomic number of the elements in the structure is measured by the RBS-HFS method described above, in order to improve the measurement accuracy, in the present invention, the structure is measured again by the p-RBS method using a proton (H ⁺ ). The ratio of the average number of atoms other than the aforementioned hydrogen in. When performing the calculation, as in the RBS-HFS method, fitting is performed sequentially from the element with the largest measured energy spectrum to calculate the scattering intensity. Based on the calculated respective scattering intensities, the ratio of the average number of atoms of the elements in the structure is calculated. In addition, by combining the ratio of the average number of atoms measured by the p-RBS method with the element with the largest energy spectrum of the detected He atom measured by the RBS-HFS method (for example, when the structure is yttrium oxide, The element with the largest energy spectrum of the detected He atom is Y) and the ratio of the average number of hydrogen atoms, and the hydrogen amount is calculated as the hydrogen atom concentration (atomic%).

In the present invention, the order of performing the p-RBS method and the RBS-HFS method is not particularly asked.

Next, the measurement conditions are described.

The RBS-HFS incident ion uses 4He ⁺ . Incident energy was 2300KeV, incident angle was 75 °, scattering angle was 160 °, and recoil angle was 30 °. The sample current was 2 nA, the beam diameter was 1.5 mmφ, and the irradiation dose was 8 μC. No in-plane rotation.

The p-RBS method uses hydrogen ions (H ⁺ ) as incident ions. The incident energy was 1740KeV, the incident angle was 0 °, the scattering angle was 160 °, and there was no recoil angle. The sample current was 1 nA, the beam diameter was 3 mmφ, and the irradiation amount was 19 μC. No in-plane rotation.

In addition to the RBS-HFS method, the combination of the p-RBS method can further improve the measurement accuracy of the hydrogen atom concentration, making it possible to quantify the correlation between the amount of hydrogen (hydrogen atom concentration) and fine particles.

The hydrogen atom concentration contained in the structure constituting the composite structure of the present invention is 7 atomic% or less. In the structure of the present invention, the surface is directly exposed to a plasma environment, for example. Therefore, in particular, the properties of the structure surface become important. The present inventors have considered the results of reviewing the results of the inability to solve the problem of particles due to the influence of nanometer-level microstructures even on structures with almost no pores having a porosity of 0.01 to 0.1%, and successfully controlled the nanometers. Advanced microstructure, a novel structure that can solve the problem of particles at a higher level is obtained. Furthermore, it has been newly discovered that the structure of the novel structure can be specified using the amount of hydrogen on the surface (hydrogen atom concentration) as an index. Furthermore, the present invention was conceived by finding the correlation between the amount of hydrogen on the structure surface and the resistance to fine particles.

Specifically, for example, in the atmosphere, it is considered that a hydrogen atom exists in a state such as a hydroxyl group (-OH). The size of the molecule is about 3 Å for water molecules and about 1 Å for hydroxyl groups, and it is considered to be the aforementioned sparse structure 80 slightly existing in the structure. By using this amount of hydrogen (hydrogen atom concentration) as an index, a nano-scale microstructure can be displayed.

In the present invention, a method of combining the p-RBS method and the RBS-HFS method is used as a specific method of the amount of hydrogen. In these methods, helium ions or hydrogen ions are made incident on the sample surface, and hydrogen is scattered forward by elastic scattering, and helium is scattered backward, and the amount of hydrogen is quantified by detecting the hydrogen. At this time, the measurement depth of the amount of hydrogen is 400 to 500 nm from the surface 10a. Therefore, in the structure 10, it is possible to appropriately quantify the microstructure of the surface 10a that most affects the fine particles.

The third aspect of the present invention As the basis of the third aspect of the present invention, it has been discovered that the new index of brightness Sa has a high correlation with particle resistance at an extremely high level. In addition, a structure having excellent particle resistance with a brightness Sa of a predetermined value or less was successfully produced. That is, a structure having extremely high particle resistance was obtained, and it was further found that the particle resistance can be quantified by the brightness Sa. Furthermore, an evaluation method for obtaining the brightness Sa was further established.

Furthermore, the present inventors have found that the brightness Sa related to the fine particle resistance is an index capable of evaluating a finer structure (microstructure) in a ceramic structure, particularly a structure in which the porosity is evaluated to be 0.01 to 0.1%.

A composite structure according to a third aspect of the present invention is a composite structure including a base material and a structure provided on the base material and having a surface, characterized in that the structure is made of polycrystalline ceramics A method for obtaining the brightness Sa by a brightness Sa value of 19 or less calculated by the following method includes: (i) a procedure for preparing a transmission electron microscope (TEM) observation sample of the structure; (ii) preparing the aforementioned A procedure for observing a digital black and white image of a bright field image of a sample by a TEM; (iii) obtaining the brightness of color data of each pixel in the aforementioned digital black and white image with a tone value (Iv) a procedure for correcting the brightness value; (v) a procedure for calculating the brightness Sa using the corrected brightness value; in the procedure (i), the TEM observation sample is prepared from the structure at least 3, each of the aforementioned at least 3 TEM observation samples was made using a focused ion beam method (FIB method: Focused Ion Beam method) to suppress processing damage, and during the FIB processing, charging was provided on the surface of the structure For the carbon layer and tungsten layer for sample protection, when the FIB processing direction is the longitudinal direction, the thickness of the upper part of the sample is 100 ± 30nm for the length in the minor axis direction of the structure surface on the plane perpendicular to the longitudinal direction. In the aforementioned procedure (ii), the digital black-and-white images are obtained for each of the at least three TEM observation samples, and each of the digital black-and-white images is accelerated using a transmission electron microscope (TEM) at a magnification of 100,000 times. The voltage is 200 kV and includes the structure, the carbon layer, and the tungsten layer. In each of the digital black and white images, a brightness obtaining area with a longitudinal length of 0.5 μm in the longitudinal direction from the surface of the structure is set. In each of the at least three TEM observation samples, a plurality of the aforementioned digital black and white images are obtained, so that the total area of the luminance acquisition area is 6.9 μm ² or more. In the procedure (iv), regarding the luminance value, the carbon The brightness value of the layer is 255, and the brightness value of the tungsten layer is relatively corrected to obtain a corrected brightness value. In the program (v), the brightness obtaining area is obtained. Each, using a least squares method (least squares method) calculates the average of absolute difference in luminance value after the correction of each of the pixel to luminance such as the average Sa.

Moreover, the evaluation method according to the third aspect of the present invention is an evaluation method including a polycrystalline ceramic and a microstructure of a structure having a surface, which is characterized by: (i) preparing a transmission electron microscope of the structure ( (TEM) procedures for observing samples; (ii) procedures for preparing digital black and white images of bright field images of the aforementioned TEM observation samples; (iii) obtaining color data representing the color data of each pixel in the aforementioned digital black and white images with numerical values of hue. Procedure for brightness value; (iv) procedure for correcting the brightness value; (v) procedure for calculating brightness Sa using the brightness value after correction; in the procedure (i), the TEM observation sample is prepared by the structure At least three, and each of the at least three TEM observation samples is made by using a focused ion beam method (FIB method) to suppress processing damage. During the FIB processing, a structure for preventing electrification and protecting the sample is provided on the surface of the structure. For the carbon layer and tungsten layer, when the FIB processing direction is the longitudinal direction, the thickness of the upper part of the sample on the length of the minor axis direction of the structure surface on the plane perpendicular to the longitudinal direction is 100 ± 30 nm. In (ii), the digital black and white images are obtained for each of the at least three TEM observation samples, and each of the digital black and white images uses a transmission electron microscope (TEM) at a magnification of 100,000 times and an acceleration voltage of 200 kV. The structure includes the structure, the carbon layer, and the tungsten layer, and in each of the digital black and white images, a brightness acquisition area is set with a longitudinal length of 0.5 μm as an area from the surface of the structure in the longitudinal direction. For each of the TEM observation samples, a plurality of the aforementioned digital black-and-white images were obtained so that the total area of the luminance acquisition area was 6.9 μm ² or more. In the aforementioned procedure (iv), regarding the luminance value, the luminance of the carbon layer was used. The value is 255, and the brightness value of the tungsten layer is relatively corrected to obtain the corrected brightness value. In the program (v), each of the pixels is calculated using the least square method for each of the brightness acquisition areas. The average of the absolute values of the differences in the brightness values after the aforementioned correction is used as the brightness Sa.

Brightness Sa in the third aspect of the present invention In the third aspect of the present invention, the microstructure of the structure is represented by an index called [luminance Sa]. This [brightness Sa] is an index obtained by quantifying pixel information of a digital black and white image of a bright field image of the structure obtained by a transmission electron microscope (TEM) as described in detail below. The structure of the composite structure according to the present invention is characterized in that the brightness Sa is 19 or less, and preferably 13 or less.

As described above, the present inventors have found that the brightness Sa related to the fine particle resistance is an index for evaluating a finer structure in a ceramic structure, particularly a structure in which the porosity is evaluated to be 0.01 to 0.1%. Therefore, in the present invention, "microstructure" means fine particles in a region where the porosity is evaluated to be 0.01 to 0.1%, and has a higher level of resistance to fine particles, and a difference in brightness Sa occurs. structure.

[Brightness value] in this specification refers to a value that represents the color data of each pixel in a digital black and white image with a value of hue (0 to 255). Here, [hue] is the level of brightness. Specifically, the value representing the color data of each pixel in a black-and-white image with a numerical value corresponding to 256 different brightness levels is referred to as a brightness value (refer to [Image Processing Device and Use Method], Nikkan Kogyo Shimbun, 1989 , First edition, p. 227). The contrast of black and white in a TEM black-and-white image is represented by a brightness value.

In this specification, [brightness Sa] refers to an index for applying the concept of Sa (Arithmetical Mean Height of the Surface) defined by the international standard ISO25178 on three-dimensional surface properties to image processing of digital TEM images. The brightness value of the digital TEM image will be specifically described using FIG. 5 showing a three-dimensional display. Figure 5 (a) is a digital black and white image of a structure of a TEM bright-field image. Regarding the digital black-and-white image, the color data of each pixel is represented by the value of the hue (0 to 255), and the value represented in the Z-axis direction is shown in FIG. 5 (b). That is, in FIG. 5 (b), the Z axis is the brightness value, and the brightness value of each pixel on the X-Y plane is represented in three dimensions. The three-dimensional image of the brightness value is compared to a three-dimensional image of a surface property defined by ISO25178 (for example, refer to the following URL. Https://www.keyence.co.jp/ss/3dprofiler/arasa/surface/). For the evaluation area, The least square method is used to calculate the average of the absolute values of the differences in the luminance values of each pixel, and regard it as [luminance Sa].

Next, [luminance Sa] in this specification is roughly calculated as follows.

In the calculation of the brightness Sa in the present invention, a TEM observation sample for obtaining a digital black-and-white image was prepared using a focused ion beam method (FIB method) to suppress processing damage. During FIB processing, a carbon layer and a tungsten layer for preventing the charging and protecting the sample are provided on the surface of the structure. When the FIB processing direction is the longitudinal direction, the thickness of the upper part of the sample on the length of the minor axis direction of the structure surface on the plane perpendicular to the longitudinal direction is 100 ± 30 nm. Prepare at least 3 TEM observation samples from one structure. Digital black and white images were obtained for each line of at least 3 TEM observation samples. Digital black and white images were acquired using a transmission electron microscope (TEM) at a magnification of 100,000 times and an acceleration voltage of 200 kV. The digital black and white image includes a structure, a carbon layer, and a tungsten layer. In the digital black-and-white image, a brightness acquisition area is set from the surface of the structure in the longitudinal direction with a longitudinal length of 0.5 μm as the area. A plurality of the aforementioned digital black and white images were acquired from each of at least three TEM observation samples so that the total area of the luminance acquisition area was 6.9 μm ² or more, and each of the acquired digital black and white images was represented by a numerical value of hue. The brightness value of the color data of the pixel is corrected relative to the brightness value of the carbon layer and the brightness value of the tungsten layer to 0. Using the corrected brightness value, the brightness Sa is calculated as shown below. In other words, for each of the brightness acquisition regions, an average of the absolute values of the differences in the brightness values after correction for each pixel is calculated using the least square method, and the average of these is used as the brightness Sa.

Hereinafter, the calculation method of the brightness Sa will be described in more detail with reference to FIGS. 2 to 9.

FIG. 2 is a flowchart of a calculation method of the display brightness Sa. The following is a description along this flowchart. (i): Preparation of TEM observation sample This procedure is a procedure for preparing a sample for TEM observation. This procedure will be described with reference to FIG. 3. The TEM observation sample was prepared by a focused ion beam method (FIB method, Focused Ion Beam method). In the processing by the FIB method, the film can be thinned to the place for observation purposes ([Surface Analysis Technology Series Penetrating Electron Microscope], edited by Japan Surface Chemical Society, Maruzen Co., Ltd., issued March 30, 1999) .

First, the structure 10 is cut out by a cutting machine or the like in advance. At this time, a portion corresponding to the sample obtaining portion 40 of FIGS. 8 and 9 described later is cut out first. Then, the structure surface 10a is subjected to FIB processing to a shape shown in FIG. 3. At this time, the arrow L in FIG. 3 is taken as the vertical direction. The longitudinal direction L is slightly parallel to the thickness direction of the structure 10. The longitudinal direction L is slightly the same as the longitudinal direction defined in the aforementioned FIB processing direction.

The FIB processing will be described in more detail. A carbon layer 50 for suppressing charge-up and protecting the surface 10a of the structure 10 is deposited on the surface 10a of the structure 10 after the cutting process. The thickness of the carbon layer 50 is about 300 nm. Here, it is preferable to smooth the structure surface 10a by grinding or the like before the carbon layer 50 is formed.

Next, the structure 10 having the carbon layer 50 deposited thereon is thinned using a focused ion beam (FIB, Focused Ion Beam) apparatus. Specifically, first, the carbon layer 50 is directed upward to irradiate the periphery of the thinned portion with a Ga ion beam, and a part of the structure 10 is cut out together with the carbon layer 50. The structure 10 cut out is fixed to the TEM sample stage for FIB by a FIB pick-up method using a tungsten deposition function. Next, in order to obtain a TEM observation sample 90, the cut structure 10 is thinned. In this thinning process, a tungsten layer 60 is first formed by a tungsten deposition process on the carbon layer 50 of the structure 10 and at the thinned part for TEM observation. By disposing the tungsten layer 60, it is possible to suppress damage to the surface of the TEM observation sample due to Ga ion beam during processing. The thickness of the vapor-deposited tungsten layer 60 is 500 to 600 nm. Then, the structure was cut off from both sides by Ga ions in the thinned portion, and a TEM observation sample 90 having a predetermined thickness (length along the arrow T in the figure) was prepared.

In the present invention, when the TEM observation sample 90 is produced, it is prepared so as to suppress processing damage such as unevenness on the processed surface. Specifically, the acceleration voltage during FIB processing starts from 40kV at the maximum voltage. Finally, in order to avoid damage to the processing surface of the structure or the formation of an amorphous layer as much as possible, finishing processing is performed at the lowest voltage of 5kV. Or finally, the damaged layer is removed by Ar ions. Observation can also be performed after the surface is cleaned by ion milling. For details of these FIB processes, please refer to [Fine Processing Using FIB Equipment] (Muro Hiroyuki, Technical Report of University of Tsukuba 24: 69-72, 2004), [FIB and Ion Grinding Techniques Q & A] (by Hiraka Masao, Asakura Kentaro , Agne Chengfeng Club).

3 (a) and 3 (b) are schematic diagrams of the TEM observation sample 90 obtained as described above. As shown in FIGS. 3 (a) and 3 (b), the TEM observation sample 90 has a thin rectangular parallelepiped shape. In FIG. 3, regarding the two directions perpendicular to the longitudinal direction L, the long axis direction is taken as the lateral direction W (arrow W in the figure), and the short axis direction is taken as the thickness direction T (arrow T in the figure). As shown in FIG. 3 (a), the electron beam penetrates the thickness direction T in the TEM observation.

As shown in FIG. 3 (a), since the thinning by Ga ions is performed from the upper side of the drawing, the sample 90 has a tendency that the lower thickness 90 b is larger than the upper thickness 90 u. Here, the upper thickness 90u refers to the length in the thickness direction T of the sample 90 on the surface 10a side. TEM observation of the thickness of the sample 90 affects the penetrability of the electron beam. Specifically, when the thickness of the sample is too large, the sensitivity (sensitivity) of the brightness Sa is slow, and there is a possibility that the correlation with the particle resistance performance cannot be obtained. When the thickness of the sample is too small, it is difficult to control the thickness during processing, and thickness unevenness may occur in the TEM observation sample 90, so that the correlation with the resistance to fine particles may not be obtained. In the present invention, the upper thickness 90u is 100 nm ± 30 nm, and more preferably 100 nm ± 20 nm.

Furthermore, in the present invention, since the brightness Sa is calculated from the image analysis using a TEM digital black and white image, the difference in thickness of the sample 90 along the longitudinal direction L (the difference between the upper thickness 90u and the lower thickness 90b) is observed by TEM as much as possible Make it smaller. Generally, the sample is in the form shown in FIG. 3 (a). The sample height 90h (length in the longitudinal direction L) is about 10 μm, and the sample width 90w (length in the transverse direction W) is about 10 μm to several tens μm.

The method for confirming the upper thickness 90u of the TEM observation sample 90 is shown below. Regarding the TEM observation sample 90, a secondary electron image was obtained using a scanning electron microscope (SEM) (see FIG. 3 (c)), and an upper thickness of 90 u was obtained from the secondary electron image. For the SEM, for example, S-5500 manufactured by Hitachi is used. SEM observation conditions were 200,000 times magnification, 2kV acceleration voltage, 40 seconds scanning time, and 2560 * 1920 pixels. At this time, the SEM image is made a plane perpendicular to the longitudinal direction L. Using a scale bar of the SEM image, an upper thickness of 90 u was obtained. At this time, the upper thickness 90u is an average of 5 times.

If the confirmation method of the upper thickness 90u is further displayed, it will be as follows. That is, regarding the digital image of the secondary electron image, the luminance value is measured along the thickness direction T to obtain a luminance line profile. At this time, the line width is 11 pixels, and the average value of the luminance values of 11 pixels in the line width direction is used. An example of the luminance spectrum profile thus obtained is shown in Fig. 3 (d). Next, the luminance spectrum contour is differentiated once, and the maximum and minimum values of the first-order differentiation are used as the ends of the structure 10 to obtain an upper thickness 90u of the structure 10 (see FIG. 3 (e)). At this time, the upper thickness 90u is an average of five luminance spectral contours.

In the calculation of the brightness Sa in the present invention, at least three TEM observation samples 90 described above are prepared from one composite structure. The preparation of at least three TEM observation samples will be further described with reference to FIGS. 8 and 9.

8 and 9 are schematic diagrams showing examples of a case where the composite structure 100 is used as a semiconductor manufacturing device member 301. In FIG. 8, in the semiconductor manufacturing device member 301, a structure 10 is disposed on a surface 70 a of a cylindrical substrate 70. In FIG. 9, in the semiconductor manufacturing device member 302, a structure 10 is disposed on a surface 70 a of a cylindrical base material 70 having a hole 31 in the center.

In the semiconductor manufacturing device members 301 and 302, the surface 10a of the structure 10 is exposed to a corrosive plasma. The semiconductor manufacturing apparatus members 301 and 302 are members constituting the inner wall of an etching reaction chamber such as a shower plate, a focus ring, a window, and a sight glass. When the structure 10 has a plasma-irradiated area 30 a and a plasma non-irradiated area 30 b that is not exposed to the plasma, the sample location 40 for measuring the brightness Sa is set to correspond to the plasma-irradiated area 30 a. At this time, if there are particularly many areas irradiated with plasma, the correlation between the particle resistance and the brightness Sa can be improved by setting the structure surface 10a corresponding to the area as the sample obtaining portion 40.

In the present invention, at least three sample acquisition sites 40 for brightness Sa measurement for preparing the TEM observation sample 90 are prepared. At this time, as shown in FIG. 8 and FIG. 9, the plurality of sample acquisition sites 40 are evenly arranged in the plasma irradiation area 30 a. Accordingly, a high correlation between the brightness Sa of the composite structure 100 and the fine particle resistance can be guaranteed.

(ii): Acquisition of TEM image G (bright field image) In this procedure, each of the at least three TEM observation samples 90 obtained by (i) is used to observe each of the samples 90 at a photographic magnification of 100,000 times and an acceleration voltage. The section was observed at 200 kV, and a TEM image G including the structure 10, the carbon layer 50, and the tungsten layer 60 was obtained (see FIG. 4). The cross section of the TEM observation sample 90 refers to the cross section of the composite structure 100 shown in FIG. 1, and more specifically, includes the cross section near the surface 10 a of the structure 10. At this time, a bright field image is obtained. Bright field image refers to an image that only transmits a transmitted wave through the objective aperture ([Surface Analysis Technology Series Penetrating Electron Microscope] Japan Surface Chemical Society, Maruzen Co., Ltd., 1999 (Publication pages 43-44 on March 30).

The TEM image G is captured using, for example, a transmission electron microscope (H-9500 / manufactured by Hitachi High-Technologies Corporation). The accelerating voltage is 200kV, with a digital camera (made by One View Camera Model 1095 / Gatan), the shooting pixels are 4096 × 4096 pixels, the capture speed is 6fps, the exposure time is 2sec, and the image capture mode (image The setting of the capture mode is performed under the same conditions as those for shooting with an exposure time and a camera position bottom mount. As shown in FIG. 4, when the image G is acquired, the structure 10, the structure surface 10a, the carbon layer 50, and the tungsten layer 60 are brought into the same field of view.

In the present invention, the brightness Sa is calculated by image analysis of the brightness information of the digital image. Therefore, focus accuracy in photography becomes extremely important. Therefore, for example, when a 100,000-times TEM image is acquired, a 100,000-times TEM image is acquired after focus adjustment is performed at a high magnification of 300,000-times or more.

The TEM image G of the digital black-and-white image is further described using FIG. 4. FIG. 4 is a schematic diagram of a TEM image G (bright-field image). In the figure, a quadrangular portion having a length G1 and a width Gw is a TEM image G. In FIG. 4, an image corresponding to the carbon layer 50 and the tungsten layer 60 deposited in the process (i) is provided on the surface 10 a of the structure 10. The portion lower than the surface 10 a corresponds to the structure 10. Further, in the TEM image G, the carbon layer 50 is white to light gray, and the tungsten layer 60 is black.

In addition, in this specification, the direction in which the structure 10, the carbon layer 50, and the tungsten layer 60 are arranged in the TEM image G, that is, the direction indicated by the arrow L in FIG. 4 is referred to as [longitudinal]. ] The direction of the arrow W in the vertical figure is called [transverse]. This longitudinal L corresponds to the longitudinal L of FIG. 3.

The physical properties and characteristics of the composite structure according to the present invention, such as particle resistance, are governed by properties near the surface of the structure. The present inventors have found that the brightness Sa in a region having a longitudinal length dL of 0.5 μm in a region along the longitudinal direction L from the structure surface 10 a is most related to physical properties and characteristics such as resistance to particles. Although the vertical image length G1 and horizontal image length Gw of the TEM image obtained at a magnification of 100,000 times also depend on the camera, they are usually about 1.5 μm to 2.0 μm, respectively. Therefore, in the present invention, a 100,000-times TEM image is used, and the luminance acquisition region R is set with the vertical length dL of the region at 0.5 μm, and the luminance Sa is obtained from the luminance value in the luminance acquisition region R.

In the present invention, a brightness acquisition area R is set for each image G. Furthermore, a plurality of digital black and white images were acquired from each of the at least three TEM observation samples so that the total area of the brightness acquisition region R was 6.9 μm ² or more. At this time, the number of images G obtained by observing the samples by each TEM is made the same. The details of the total area of the luminance acquisition region R will be described later.

(iii): Get brightness value

Next, in this program, each [vertical] and [horizontal] coordinate in the TEM image G of the digital black and white image obtained in the aforementioned program (ii) is used to obtain the brightness value of each pixel corresponding thereto. Here, the luminance value of the tungsten layer and the luminance value of the carbon layer in the image G are also included. FIG. 5 (a) is a plan view showing an example of the obtained TEM image G. FIG. An arrow Y in FIG. 5 (a) corresponds to the longitudinal direction L in FIGS. 3 and 4. The arrow X in FIG. 5 (a) corresponds to the horizontal direction W in FIGS. 3 and 4. FIG. 5 (b) is a diagram showing the luminance value of each pixel in the three-dimensional direction of the arrow Z in the image G. FIG.

(iv): Procedure for Correcting the Brightness Value Next, in this procedure, a correcting operation for the luminance value of the TEM image G obtained in the aforementioned procedure (iii) is performed. The specific content will be described using FIG. 6. FIG. 6 (a) is a diagram that three-dimensionally displays the luminance value of the TEM image G obtained in the above-mentioned procedure (iii) corresponding to the luminance value of each pixel of the vertical and horizontal coordinates of the structure 10. Here, as in the procedure (iii), the luminance value of the tungsten layer and the luminance value of the carbon layer in the image G are also included. In this program, the brightness value of each pixel is set to the brightness value of the tungsten layer 60 in the image G and the brightness value of the carbon layer 50 in the image G to 255. Each pixel relatively corrects the corresponding structure. The brightness value of the object 10. As described above, in the TEM image G, the tungsten layer 60 becomes black, and the carbon layer 50 becomes white to light gray. In this program, between the brightness value 0 of the tungsten layer and the brightness value 255 of the carbon layer, the brightness value of each pixel of the structure 10 is used as a relative value for correction and calculation. FIG. 6 (b) is a diagram showing the luminance value of the image G after three-dimensional display correction. Compared with FIG. 5, it can be seen that the tungsten layer 60 is based on black, and the carbon layer 50 is based on white to light gray, and the brightness value in the structure 10 is relatively corrected.

In the TEM image G, the brightness value of each pixel of the carbon layer 50 / tungsten layer 60 is usually slightly uneven. In order to eliminate the influence of the unevenness, the value of the brightness value of the tungsten layer 0 and the brightness value of the carbon layer 255 are determined as follows. Regarding the brightness value of the tungsten layer 60, the average value of the brightness values of 10,000 pixels is continuously used in the order of the smallest value of the brightness value of the tungsten layer 60 in the image G. For the brightness value of the carbon layer 50, the average value of the brightness values of 100,000 pixels is used in the descending order of the maximum value of the brightness value in the carbon layer 50. Each average value obtained here is used as the luminance value of the carbon layer 50 / tungsten layer 60 before correction. The luminance values are similarly corrected for the plurality of TEM images G as well.

(v): Calculation of brightness Sa In this program, the brightness Sa is calculated from the brightness value of the image G obtained by the program (iv). Specifically, for each of the aforementioned luminance acquisition regions R (hereinafter, each luminance acquisition region is represented as R _n ), the average of the absolute values of the differences in the corrected luminance values of each pixel is calculated by the least square method, and such as the average luminance of the region R _n Sa _n. Then, about _{+ ·} total brightness acquisition region R _{1 + 2 +} area becomes an average value of _n is set 6.9μm ² or more made of a plurality of luminance brightness region R _n is calculated as Sa _n structure 10 Brightness Sa.

In the present invention, in order to improve the correlation between the brightness Sa and the physical properties and characteristics of the structure, such as resistance to fine particles, when the brightness Sa is calculated, the area of the brightness acquisition region R is 6.9 μm ² or more. By determining the brightness Sa based on the area exceeding this area, the structure 10 can accurately and appropriately display even if there is a possibility that unevenness in physical properties and characteristics occurs in the limited area. Correlation between brightness Sa and physical properties.

The size of one TEM observation sample 90 is very small relative to the surface area of the structure 10, that is, the plasma irradiation area. On the other hand, the physical properties and characteristics imparted to the structure, such as resistance to particles, are in principle required for the entire structure surface. In this regard, in the present invention, it is required that the total area of the brightness acquisition region R is 6.9 μm ² or more, and that at least three TEM observation samples are prepared from the structure 10. That is, in the present invention, a plurality of TEM observation samples are obtained equally from the surface of the structure, so as to capture as much information as possible on the entire surface of the structure. In addition, when obtaining at least three TEM observation samples, it is necessary to consider information on the entire surface of the structure.

The brightness Sa is related to the physical properties and characteristics possessed by the structure, such as particle resistance. Therefore, according to the present invention, by calculating the brightness Sa of the structure, instead of evaluating the actual fine particle resistance of the structure, it is possible to grasp the properties such as the fine particle resistance of the structure before use.

Focusing on the description of the area of the brightness acquisition area R, the procedure for calculating the brightness Sa will be described in more detail.

In order to improve the correlation between the brightness Sa and the resistance to fine particles, the brightness acquisition region R in the image G is set so that the total area thereof becomes 6.9 μm ² or more. Specifically, a plurality of (n) images G _{n are} obtained from at least three TEM observation samples, and a region R _{n is} set for each image. Then, the average value of the brightness Sa _n of each image Gn is used as the brightness Sa of the structure.

A method for calculating the brightness Sa ₁ from one image G ₁ will be described with reference to FIGS. 4, 6 (a), and 7 (a) to (b).

4 and FIG. 6 (a), the image G ₁ is set to a region R _1. In one region R ₁ , the region longitudinal length dL is 0.5 μm. As described above, it is considered that the properties near the surface of the structure are most related to the physical properties and characteristics of the structure, such as particle resistance. In addition, the region lateral length dW, which is the lateral W in the region R ₁ , is set to be the longest in the video G. As an example, the lateral length dW of the region is about 1.5 μm to 2.0 μm at a magnification of 100,000 times. That is, the area of the region R _{1 that} can be set for one TEM image G ₁ captured at a magnification of 100,000 times is 0.75 to 1.0 μm ² .

Therefore, in this example, the number of images G required for the total area of the brightness acquisition region R _n is 6.9 μm ² to 7 to 9 sheets. On the other hand, as described above, in the present invention, at least three TEM observation samples 90 are prepared, and the same number of TEM images G are obtained from each of the TEM observation samples 90. Therefore, in this example, three images G are obtained for each TEM observation sample 90. When a plurality of TEM images G _n are acquired from one TEM observation sample 90, as shown in FIG. 3 (b), the plurality of images G _n are continuously acquired in the lateral direction W. In addition, the sizes of the plurality of images G _n (the image vertical length G1 and the image horizontal length Gw) are slightly the same. Accordingly, for example, the influence of measurement unevenness among observers can be reduced, and the correlation between the brightness Sa and the resistance to particles can be further improved.

The present inventors have found that the brightness Sa in a region of 0.5 μm from the surface of the structure 10a is most related to the physical properties and characteristics of the structure, such as resistance to particulates, as described above. Therefore, in the region R when the brightness Sa is calculated, the region vertical length dL is 0.5 μm. Furthermore, in the procedure (i), as described with reference to FIG. 3, since the processing is performed from the surface 10a direction of the structure 10, it becomes a sample of the TEM observation sample 90 even if the processing accuracy is improved. The thickness of the lower part 90d is larger than the thickness of the upper part of the sample 90u. Therefore, the greater the depth from the surface 10a of the structure 10, the more difficult it is for the electron beam to penetrate, and the sensitivity of the luminance value becomes dull. That is, in the image G, the image becomes darker as a whole with respect to the surface 10a side as it goes toward the depth direction, that is, a blackish image. Therefore, in order to sufficiently reduce the influence of the sample thickness, the longitudinal length dL of the region needs to be 0.5 μm.

When the brightness acquisition region R is set, the region vertical length dL is set with the vicinity of the surface 10 a of the structure 10 as a base point. In the TEM image G, when a gap is observed between the surface 10 a and the carbon layer 50, it is necessary to set the region R while avoiding the gap. [Near the surface 10a] means a range of about 5 to 50 nm from the surface 10a. The details of the specific setting can be performed with reference to the examples described later.

The processes in the above procedures (iii) to (iv) can be performed continuously and collectively in the image analysis software. As such software, WinROOF2015 (available from Mitani Corporation) can be used.

In addition, although it is considered that the smaller the brightness Sa of the structure of the composite structure according to the present invention is, the better, but the existence of a critical value in manufacturing is also apparent to those skilled in the art. Since such a manufacturing limit becomes the lower limit value of the brightness Sa in the present invention, the lower limit value is not specifically specified and does not make the present invention according to the third aspect unclear. This point is the same in the fourth aspect below.

Fourth aspect of the present invention In the fourth aspect of the present invention, it is characterized in that although the brightness Sa is taken as an index as in the third aspect of the present invention, a method for obtaining the brightness Sa in the first aspect The procedure (iv) in (i.e., the procedure for correcting the luminance value) includes a procedure for removing noise components. Therefore, the description of the third aspect in this specification other than the procedure for removing the noise component becomes the description of the fourth invention.

Moreover, the composite structure according to the fourth aspect of the present invention is a composite structure including a substrate and a structure disposed on the substrate and having a surface, wherein the structure includes a polycrystalline ceramic Thus, a method for obtaining the brightness Sa by a brightness Sa value of 10 or less calculated by the following method includes: (i) a procedure for preparing a transmission electron microscope (TEM) observation sample of the structure; (ii) A procedure for obtaining a digital black and white image of a bright field image of the aforementioned TEM observation sample; (iii) a procedure for obtaining a brightness value of a color data of each pixel in the digital black and white image by a hue value; (iv) correcting the brightness (V) a program for calculating the brightness Sa using the corrected brightness value. In the program (i), the TEM observation sample is prepared from the structure at least three, and the at least three TEM observation tests are prepared. Each of the samples was manufactured using a focused ion beam method (FIB method) to suppress processing damage. During the FIB processing, a carbon layer and a tungsten layer for preventing charge and sample protection were provided on the surface of the structure. When the processing direction is the longitudinal direction, the thickness of the upper part of the sample on the length of the minor axis direction of the surface of the structure on the longitudinal vertical plane is 100 ± 30nm. In the foregoing procedure (ii), the digital black and white image is at least 3 Each of the TEM observation samples was obtained. Each of the aforementioned digital black and white images uses a transmission electron microscope (TEM) at a magnification of 100,000 times and an acceleration voltage of 200 kV, and includes the structure, the carbon layer, and the tungsten layer. In each of the digital black-and-white images, a brightness acquisition area having a longitudinal length of 0.5 μm as an area in the longitudinal direction from the surface of the structure is set, and a plurality of the digital black-and-white images are obtained from each of the at least three TEM observation samples. In order to make the total area of the brightness acquisition area 6.9 μm ² or more, in the above-mentioned procedure (iv), the brightness value of the carbon layer is 255, and the brightness value of the tungsten layer is 0. Correct the brightness value to obtain the corrected brightness value. Regarding the digital black and white image with the brightness value corrected, a low-pass filter is used. The cut-off frequency in the noise removal using the aforementioned low-pass filter is 1 / (10 pixels). In the aforementioned procedure (v), the minimum level is used for each of the brightness acquisition regions. The method calculates an average of the absolute values of the differences in the brightness values after the corrections for each of the pixels, and uses these averages as the brightness Sa.

In addition, the evaluation method according to the present invention is an evaluation method of a microstructure of a structure including a polycrystalline ceramic and having a surface, which comprises: (i) preparing a transmission electron microscope (TEM) observation sample of the structure; (Ii) a procedure for obtaining a digital black-and-white image of the bright field image of the aforementioned TEM observation sample; (iii) a procedure for obtaining a brightness value of a color data of each pixel in the digital black-and-white image by a hue value; (iv) a procedure for correcting the brightness value; (v) a procedure for calculating the brightness Sa using the brightness value after the correction; in the procedure (i), the TEM observation sample is prepared from the structure at least 3; Each of the at least three TEM observation specimens was produced using a focused ion beam method (FIB method) to suppress processing damage. During the aforementioned FIB processing, a carbon layer and a tungsten layer for preventing charging and protecting the sample were provided on the surface of the structure. When the FIB processing direction is the longitudinal direction, the thickness of the upper part of the sample in the minor axis direction of the surface of the structure on the longitudinal vertical plane is 100 ± 30 nm. In the foregoing procedure (ii), the The digital black-and-white images are obtained for each of the at least three TEM observation samples. Each of the digital black-and-white images uses a transmission electron microscope (TEM) at a magnification of 100,000 times and an acceleration voltage of 200 kV. In the carbon layer and the tungsten layer, in each of the digital black-and-white images, a brightness acquisition region is set with a longitudinal length of 0.5 μm as a region from the surface of the structure in the longitudinal direction, and the sample is observed from the at least 3 TEMs. Each of the plurality of digital black-and-white images is obtained so that the total area of the luminance acquisition area becomes 6.9 μm ² or more. In the foregoing procedure (iv), regarding the luminance value, the luminance value of the carbon layer is 255, and The brightness value of the tungsten layer is relatively corrected to obtain a corrected brightness value. The digital black and white image with the brightness value corrected is subjected to noise removal using a low-pass filter. The cut-off frequency in noise removal is 1 / (10 pixels). In the above-mentioned program (v), for each of the brightness acquisition areas, the least square method is used to calculate each An average of the absolute values of the differences in the brightness values of the aforementioned pixels after the correction, and these averages are used as the brightness Sa.

As such, the composite structure according to the fourth aspect is characterized in that the brightness Sa is 10 or less, and preferably 5 or less.

In the fourth aspect of the present invention, as in the third aspect of the present invention, the following procedure is performed in order to obtain the brightness Sa: Procedure (i), that is, preparing a transmission electron microscope (TEM) observation of a structure Sample procedure; procedure (ii), that is, a procedure for preparing a digital black and white image of the bright field image of the sample under TEM observation; and then procedure (iii), that is, obtaining each pixel in the digital black and white image as a hue value Procedure for the brightness value of color data. Furthermore, in the program (iv), a program for removing noise components in an image so as to make the brightness Sa more accurately and appropriately represent the microstructure of the structure is performed as necessary. The TEM image G contains noise having a high frequency component, and the noise is removed by a filter in this additional procedure. In this aspect, a low-pass filter (LPF) is used for image G to perform noise removal. For details on noise removal in image processing, please refer to [Image Processing-From Its Basics to Application 2nd Edition] (by Hiro Ozaki and Keiji Taniguchi, Kyoritsu Publishing Co., Ltd.).

In this aspect, the cutoff frequency in noise removal using a low-pass filter is 1 / (10 pixels). That is, the cut-off period is 10 pixels. For example, when using WinROOF2015 as image processing software, use the noise removal command to set the cutoff frequency.

Figs. 7 (c) and (d) are the noise-corrected image G of Fig. 7 (a) and (b). The cut-off frequency is 1 / (10 pixels) and noise is removed using a low-pass filter. Examples of images. By performing noise removal, the influence of the focusing accuracy of the TEM image G and the like can be eliminated, and the correlation between the physical properties and characteristics possessed by the structure such as the resistance to fine particles and the brightness Sa can be improved.

In the fourth aspect of the present invention, the program (v) for calculating the brightness Sa using the thus obtained corrected brightness value is performed later. This program (v) may be the same as the third aspect of the present invention.

A fifth aspect of the present invention is different from the first to fourth aspects of the present invention in the fifth aspect of the present invention, and its object is limited to a structure including Y (yttrium element) and O (oxygen element) Moreover, the microstructure of the structure is expressed by using the refractive index as an index. That is, the present inventors succeeded in a composite structure provided with a structure containing Y (yttrium element) and O (oxygen element) used in a semiconductor manufacturing device or the like exposed to a corrosive plasma environment, for example. , Extremely reduce the impact of particles. In addition, it was found that by using the refractive index as an index, the particle resistance performance can be evaluated at an extremely high level.

A composite structure according to a fifth aspect of the present invention is a composite structure including: a substrate and a structure provided on the substrate and having a surface, the structure including: Y (yttrium element) and O (Polyoxygen) polycrystalline ceramics have a refractive index greater than 1.92 at a wavelength of 400 nm to 550 nm. The refractive index is calculated by reflection spectroscopy using a microspectral film thickness meter, and is used as a measurement condition to measure the spot size. 10 μm, the average surface roughness Ra of the substrate surface and the surface of the composite structure Ra ≦ 0.1 μm, the thickness of the structure ≦ 1 μm, the measurement wavelength range is 360 to 1100 nm, and the analysis wavelength range is 360 to 1100 nm. Optimization method and least square method.

In addition, a fifth aspect of the composite structure according to the present invention is a composite structure including: a substrate, and a structure disposed on the substrate and having a surface, wherein the structure includes: Y ( Polycrystalline ceramics with yttrium) and O (oxygen) have a refractive index of at least 1.99 at a wavelength of 400 nm, 1.96 or more at a wavelength of 500 nm, 1.94 or more at a wavelength of 600 nm, 1.93 or more at a wavelength of 700 nm, or 1.92 or more at a wavelength of 800 nm In either case, the refractive index is calculated by reflection spectroscopy using a microspectral film thickness meter, and the measurement condition is a measurement spot size of 10 μm, and the average surface roughness Ra of the surface of the substrate and the surface of the composite structure Ra ≦ 0.1 μm. The thickness of the aforementioned structure is ≦ 1 μm, the measurement wavelength range is 360 to 1100 nm, and the analysis wavelength range is 360 to 1100 nm. An optimization method and a least square method are used.

In general, the average refractive index of Y ₂ O ₃ is 1.92 (Edited by the Chemical Society of Japan [Chemical Handbook], Maruzen (1962), p919, [Ceramic Chemistry] p220, Table 8-24, Japan Ceramic Association, September 30, 1994 Revised edition, etc.). Furthermore, in the Journal of the Japanese Chemical Society, 1979, (8), p. 1106 to 1108 (Non-Patent Document 1), the optical characteristics of the yttrium oxide sintered body as a transparent plate-shaped sample reveal the refractive index and reflectance (see Figure 20). In contrast, the composite structure according to the fifth aspect of the present invention has a refractive index greater than 1.92 at a wavelength of 400 to 550 nm, for example. Or at least any one of 1.99 or more at a wavelength of 400 nm, 1.96 or more at a wavelength of 500 nm, 1.94 or more at a wavelength of 600 nm, 1.93 or more at a wavelength of 700 nm, and 1.92 or more at a wavelength of 800 nm or more.

In the fifth aspect, although the basic structure of the composite structure is the same as the composite structures according to the first to fourth aspects, in the basic structure of FIG. 1, the structure 10 includes: containing Y (yttrium element) and O (oxygen) polycrystalline ceramic (hereinafter referred to as [YO compound] as appropriate). The structure 10 included in the composite structure exhibits a predetermined refractive index.

Therefore, the structure 10 including Y (yttrium element) and O (oxygen element) included in the composite structure according to the fifth aspect is a so-called Y-O compound coating. By applying a Y-O compound coating, 70 kinds of physical properties and characteristics can be imparted to a substrate. In addition, in this aspect, ceramic structures and ceramic coatings are also used synonymously unless otherwise specified.

According to a preferred aspect, the structure 10 is mainly composed of polycrystalline ceramics containing YO compounds, preferably containing more than 50% of YO compounds, more preferably 70% or more, still more preferably 90% or more, 95% %the above. Most preferably, the structure 10 is composed of a Y-O compound.

In this aspect, the YO compound means, for example, an oxide of yttrium. Examples include Y ₂ O ₃ and Y _α O _β (nonstoichiometric composition). In addition to the Y element and the O element, other elements may be included. For example, an YO compound containing at least any one of an F element, a Cl element, and a Br element may be mentioned. The structure 10 contains, for example, Y ₂ O ₃ as a main component. The content of Y ₂ O ₃ is 70% or more, preferably 90% or more, and more preferably 95% or more. Most preferably, the structure 10 is composed of 100% of Y ₂ O ₃ .

Refractive Index in the Fifth Aspect The measurement of the refractive index may be performed by calculation using a reflection spectrometer using a microspectral film thickness meter (for example, OPTM-F2, FE-37S manufactured by Otsuka Electronics). The measurement conditions were a measurement spot size of 10 μm and a measurement wavelength range of 360 to 1100 nm. In addition, as an analysis condition, an analysis wavelength range of 360 to 1100 nm was used, and an optimization method and a least square method were adopted.

In the fifth aspect of the composite structure according to the present invention, the refractive index at a wavelength of 400 nm to 550 nm is greater than 1.92, preferably the refractive index at 400 nm to 600 nm is greater than 1.92, and more preferably the refractive index at 400 nm to 800 nm is greater than 1.92. Furthermore, in the composite structure according to the fifth aspect of the present invention, the refractive index satisfies 1.99 or more at a wavelength of 400 nm, 1.96 or more at a wavelength of 500 nm, 1.94 or more at a wavelength of 600 nm, 1.93 or more at a wavelength of 700 nm, or 1.92 or more at a wavelength of 800 nm or more. At least any one. The present inventors have newly discovered a novel structure that increases the refractive index in a composite structure containing a Y-O compound as described above. Furthermore, it was unexpectedly found that the composite structure containing a Y-O compound having a very high refractive index was extremely excellent in fine particle resistance, and the present invention was conceived. In a preferred aspect of the present invention, the upper limit of the refractive index is 2.20.

Method for preparing a composite structure The composite structure according to the present invention can be manufactured by various manufacturing methods as long as it is a composite structure capable of realizing the indicators having the first to fifth aspects described above. According to one aspect of the present invention, the composite structure according to the present invention can be preferably manufactured by forming a structure on a substrate by using an aerosol deposition method (AD method). According to a preferred aspect of the present invention, the structure of the composite structure according to the present invention can be realized by the AD method. The AD method refers to spraying an aerosol of fine particles of a brittle material such as ceramics and a gas on the surface of the substrate, causing the fine particles to collide with the substrate at a high speed, and pulverizing or deforming the fine particles by the collision to the substrate. A method for forming a structure (ceramic coating) on the substrate.

Apparatus for performing the AD method Although the apparatus used for the AD method for manufacturing the first to fifth composite structures according to the present invention is not particularly limited, it has a basic configuration shown in FIG. 10. That is, the device 19 used in the AD method includes a reaction chamber 14, an aerosol supply unit 13, a gas supply unit 11, an exhaust unit 18, and a pipe 12. Inside the reaction chamber 14 are arranged a stage 16 on which the substrate 70 is arranged, a driving unit 17, and a nozzle 15. The position of the substrate 70 and the nozzle 15 disposed on the stage 16 can be relatively changed by the driving unit 17. In this case, the distance between the nozzle 15 and the base material 70 may be constant, or may be made variable. In this example, although the driving unit 17 is shown driving the stage 16, the driving unit 17 may drive the nozzle 15. The driving direction is, for example, the XYZθ direction.

In the apparatus of FIG. 10, the aerosol supply unit 13 is connected to the gas supply unit 11 via a pipe 12. In the aerosol supply unit 13, the aerosol in which the raw material fine particles and the gas are mixed is supplied to the nozzle 15 through a pipe 12. The device 19 further includes a powder supply unit (not shown) that supplies raw material fine particles. The powder supply section may be arranged inside the aerosol supply section 13 or may be arranged separately from the aerosol supply section 13. Further, in addition to the aerosol supply unit 13, an aerosol-forming unit may be provided in which raw material particles and a gas are mixed. By controlling the supply amount from the aerosol supply unit 13 so that the amount of fine particles ejected from the nozzle 15 becomes constant, a homogeneous structure can be obtained.

The gas supply unit 11 supplies nitrogen, helium, argon, air, and the like. When the gas to be supplied is air, for example, compressed air with little impurities such as moisture or oil is used, or an air treatment unit for removing impurities by air is more preferred.

Next, the operation of the device 19 used in the AD method will be described. In a state in which the substrate is arranged on the platform 16 in the reaction chamber 14, the inside of the reaction chamber 14 is decompressed to an atmospheric pressure or lower, specifically to several hundreds Pa by an exhaust portion 18 such as a vacuum pump. about. On the other hand, the internal pressure of the aerosol supply unit 13 is set to be higher than the internal pressure of the reaction chamber 14. The internal pressure of the aerosol supply unit 13 is, for example, several hundreds to tens of thousands of Pa. The powder supply unit may be at atmospheric pressure. The differential pressure in the reaction chamber 14 and the aerosol supply unit 13 accelerates the fine particles in the aerosol, so that the ejection speed of the raw material particles from the nozzle 15 is in a range from subsonic to supersonic (50 to 500 m / s). The injection speed is appropriately controlled by the flow rate of the gas supplied from the gas supply unit 11, gaseous species, the shape of the nozzle 15, the length and inner diameter of the piping 12, and the discharge volume of the exhaust unit 18. For example, a supersonic nozzle such as a Laval nozzle may be used as the nozzle 15. The fine particles in the aerosol sprayed from the nozzle 15 collide with the substrate, pulverize or deform, and deposit a structure on the substrate. By changing the relative position of the substrate and the nozzle 15, a composite structure including a structure having a predetermined area on the substrate is formed. The specific manufacturing conditions such as the pressure in the reaction chamber, the gas flow rate, and the like are changed by the combination of various devices, and these conditions can be appropriately adjusted within the range in which the structure of the present invention can be formed.

In addition, a cracking unit (not shown) that releases agglutination of fine particles before ejection from the nozzle 15 may be provided. Any method can be selected as the method of lysing in the lysing section as long as it meets the collision form of the fine particles on the substrate described later. Examples include mechanical cracking such as vibration and collision; and well-known methods such as static electricity, plasma irradiation, and classification.

In addition to the above, the composite structure according to the present invention is suitably used in various applications as follows. In other words, coatings and slidings are used in aerospace industries such as electric vehicles, touch panels, LEDs, solar cells, dental implantations, and satellite mirrors. Components, corrosion-resistant coatings in chemical plants, all-solid-state batteries, thermal barrier coatings, high refractive index applications such as optical lenses, Optical mirrors (optical mirrors), optical elements, jewelry, and the like are suitably used. [Example]

The following examples further illustrate the present invention, but the present invention is not limited to these examples.

1. Sample preparation 1-1 Raw material particles As raw material particles, yttrium oxide, yttrium oxyfluoride, yttrium fluoride, alumina powder, and zirconia powder were prepared. Table 1 shows the average particle size and particle state of various powders.

1-2 Film Formation of Samples The composite structures of samples a to d and f to n were prepared using the raw material particles and the substrate shown in Table 1. As the sample e, a commercially available sample prepared by an ion plating method was used, and other samples were prepared by an aerosol deposition method.

The basic structure of the device used in the AD method is the same as that shown in FIG. 10. The raw material particles, substrate, gas species, and gas flow rate of each sample are shown in Table 1, and the spray speed from the nozzle was 150 m / s or more. In addition, the formation thickness of the structures is about 5 μm. The formation of the structure is performed at room temperature (about 20 ° C).

2.Characterization of structures (part one) 2-1 average crystallite size

For sample e and sample c, the average crystallite size was calculated from the TEM image taken at a magnification of 400,000. Specifically, the average crystallite size was calculated using an image obtained at a magnification of 400,000 times and an average value consisting of approximately 15 circular crystallites.

Regarding the sample c prepared by the AD method, the average crystallite size calculated from the TEM image was 9 nm.

In the sample e prepared by ion plating, the average crystallite size calculated from the TEM image was 1 nm.

2-2 Porosity measurement About samples a to h, the porosity was calculated from the image analysis using the image analysis software WinRoof2015 using an image obtained by a scanning electron microscope (SEM). The magnification is from 5000 to 20,000 times. This porosity measurement is conventionally used as a method for evaluating the density of a structure.

As a result, in all the samples a to h, the porosity was 0.01% or less. The SEM images of samples a, c, e, f, and g are shown in Fig. 12, respectively. As will be described later, with respect to such samples having different microparticle resistance, in the measurement of porosity that has been conventionally performed, it is impossible to specify the difference in the structure of the samples.

3. Characterization and analysis of structures (part two) 3-1. Hydrogen measurement by D-SIMS method As a sample for hydrogen measurement, the following method was used. First, two samples a to c, f, i to k, m, and n were prepared. The sample size is 3mm × 3mm, and the thickness is 3mm. Regarding each of the samples, one was used as a standard sample, and the other was used as a measurement sample. For each sample, the two-dimensional average surface roughness Ra of the surface 10a of the structure 10 was adjusted to 0.01 μm by grinding or the like. Next, for each sample, the amount of hydrogen was measured by D-SIMS after being left for more than 24 hours at a room temperature of 20-25 ° C, a humidity of 60% ± 10%, and an atmospheric pressure.

The measurement of the amount of hydrogen by a secondary ion mass spectrometry method: Dynamic-Secondary Ion Mass Spectrometry (D-SIMS method) was performed using IMF-7f manufactured by CAMECA as a device.

The production of standard samples is shown below. Prepared are: an evaluation sample, a standard sample for an evaluation sample that is equivalent to a sample having a matrix component equivalent to the evaluation sample, a Si single crystal, and a standard sample for a Si single crystal. For each of the two prepared samples, one was used as an evaluation sample and the other was used as a standard sample for evaluation. The standard sample for the evaluation sample is a sample in which heavy hydrogen is injected into a sample having a matrix component equivalent to the evaluation sample. At this time, heavy hydrogen was also injected into the Si single crystal, and equivalent heavy hydrogen was injected into the standard sample and the Si single crystal for the evaluation sample. Then, the amount of deuterium implanted into the Si single crystal was identified using a standard sample for the Si single crystal. For the standard sample used for the evaluation sample, the secondary ion intensity of the deuterium and the constituent element was calculated using the secondary ion mass spectrometry (D-SIMS method), and the relative sensitivity coefficient was calculated. The hydrogen content of the evaluation sample was calculated using the relative sensitivity coefficient calculated from the standard sample for the evaluation sample. For other departments, refer to ISO 18114_ "Determining relative sensitivity factors from ion-implanted reference materials" (International Organization for Standardization, Geneva). , 2003).

Conductive platinum (Pt) was vapor-deposited on the structure surface of each of the measurement sample and the standard sample. As a measurement condition of D-SIMS, cesium (Cs) ion was used as a primary ion. The primary acceleration voltage was 15.0 kV, the detection area was 8 μmφ, and the measurement depth was 3 levels of 500 nm, 2 μm, and 5 μm.

The obtained number of hydrogen atoms (atoms / cm ³ ) per unit volume is shown in Table 2 described later.

3-2. Measurement of hydrogen amount by RBS-HFS method and p-RBS method First, regarding the samples a, c, f, and j, the surface 10a of the structure 10 was polished so that the two-dimensional average surface roughness Ra became 0.01 μm. . Next, the sample was left at room temperature of 20-25 ° C., humidity of 60% ± 10%, and atmospheric pressure for 24 hours or more, and then the amount of hydrogen (hydrogen atom concentration) was measured.

The hydrogen amount was measured by combining RBS-HFS and p-RBS. As the device, Pelletron 3SDH manufactured by National Electrostatics Corporation was used.

The measurement conditions of the RBS-HFS method are shown below. Incident ions: 4He ⁺ incident energy: 2300KeV, incident angle: 75 °, scattering angle: 160 °, recoil angle: 30 ° Sample current: 2nA, beam diameter: 1.5mmφ, irradiation amount: 8μC In-plane rotation: None

The measurement conditions of the p-RBS method are shown below. Incident ion: Hydrogen ion (H ⁺ ) Incident energy: 1740KeV, incidence angle: 0 °, scattering angle: 160 °, recoil angle: No sample current: 1nA, beam diameter: 3mmφ, irradiation amount: 19μC In-plane rotation: no

The obtained hydrogen atom concentration is shown in Table 2 described later.

3-3-1. Preparation of samples for TEM observation About samples a to n, TEM observation samples were prepared by a focused ion beam method (FIB method, Focused Ion Beam). First, each sample was cut. Then, the surface of the structure of each sample was subjected to FIB processing. First, a carbon layer 50 is deposited on the surface of the structure of each sample. The target thickness of the carbon layer is about 300 nm.

After the carbon layer was evaporated, each sample was thinned using a FIB apparatus. First, a Ga ion beam is irradiated to the periphery of the thinned portion with the carbon layer facing upward, and a part of the structure of each sample is cut out with the carbon layer. The structure cut out was fixed to the TEM sample stage for FIB by the FIB pickup method using a tungsten deposition function. Next, a tungsten layer is formed on the carbon layer 50 and the thinned portion for TEM observation by a tungsten deposition process. The target thickness of the tungsten layer is 500 to 600 nm. Then, the structure of each sample was cut off from both sides by Ga ions in the thinned portion, and a TEM observation sample was produced. The target thickness of the TEM observation sample at this time was 100 nm. The acceleration voltage during FIB machining starts from 40kV at the maximum voltage, and finishes machining at 5kV at the lowest voltage. In this way, three TEM observation samples were obtained.

Next, it was confirmed that the upper thickness of the TEM observation sample 90 was 90u. A secondary electron image was obtained using a TEM observation sample and a scanning electron microscope (SEM), and an upper thickness of 90 u was obtained from the secondary electron image. For the SEM, S-5500 manufactured by Hitachi was used. SEM observation conditions were 200,000 times magnification, 2kV acceleration voltage, 40 seconds scanning time, and 2560 * 1920 pixels. Using the scale of this SEM image, the upper thickness of each TEM observation sample was 90u from an average of 5 times. The upper thickness 90u of each sample is shown in Table 1. There is a tendency that when the thickness of the upper portion of the sample is more than 90 ± 100 ± 30nm, the brightness Sa becomes smaller, and when the thickness of the upper portion of the sample is more than 100 ± 30nm, the brightness Sa becomes larger. Regarding the sample g, although the thickness of the upper portion of the sample is 138 nm, which is larger than the predetermined range, that is, the brightness Sa is smaller than originally, it was also confirmed that this is also outside the scope of the present invention.

3-3-2, Photography of TEM bright-field image About samples a to n obtained by FIB processing, photography of bright-field image using TEM was performed. A transmission electron microscope H-9500 (manufactured by Hitachi High-tech Co., Ltd.) was used with an acceleration voltage of 200 kV and an observation magnification of 100,000 times. A digital camera (made by OneView Camera Model 1095 / Gatan) was used at 4096 × 4096 pixels for photography pixels, The capture speed is 6fps, the exposure time is 2sec, and the image capture mode is set to shoot with the exposure time and camera position base. A TEM digital black-and-white image was obtained in such a manner that the carbon layer and the tungsten layer vapor-deposited when the sample was prepared were included in the same field of view.

FIG. 13 is a TEM image of samples a to g, i, j, and l taken at a magnification of 100,000 times. The TEM images are obtained by continuously acquiring three TEM images for each TEM observation sample, and a total of nine TEM images are obtained for one sample.

3-3-3. Calculation of brightness Sa by brightness value acquisition and image analysis From the obtained TEM bright field image, the image brightness software is used to obtain the brightness value of the image using WinROOF2015. Specifically, as shown in FIG. 13, regarding each image G, the vicinity of the structure surface 10 a is used as a base point, so that the vertical length dL of the region is 0.5 μm, and the horizontal length dw of the region is substantially the same as the horizontal length Gw of the video G The brightness acquisition area R is set. For each brightness obtaining area R, the brightness value of each pixel is obtained, and the brightness value is relatively corrected with the brightness of the tungsten layer in the area R being 0 and the brightness of the carbon layer being 255. Here, for the brightness value of the tungsten layer, the average value of the brightness values of 10,000 pixels is continuously used in the order of the smallest value among the brightness values in the tungsten layer in the TEM image. In addition, as for the brightness value of the carbon layer, the average value of the brightness values of 100,000 pixels is used in the descending order from the maximum value of the brightness value in the carbon layer. The average value obtained here is used as the brightness value of the carbon layer / tungsten layer before correction.

With respect to each of the brightness acquisition regions R, an average of absolute values of differences in brightness values after correction for each pixel is calculated using a least square method. In this way, the values obtained from the nine TEM images were averaged and used as the brightness Sa. The total area of the brightness acquisition region R at this time is 6.9 μm ² or more.

Further, as shown in FIGS. 13 (e), (f), and (g), when the interface between the surface 10a of the structure 10 and the tungsten layer 50 is not straight, the brightness acquisition region R is set away from this region. Further, as shown in FIG. 13 (e), when the unevenness of the surface 10a of the structure 10 is observed, the region vertical length dL is set with the portion closest to the surface 10a as a starting point.

The obtained luminance Sa value is shown in Table 2 described later.

3-4. Calculation of the brightness Sa without noise components In the above 3-3-3, the procedure for obtaining the brightness Sa by the brightness value acquisition and image analysis, the image analysis software WinROOF2015 is used to remove the noise components. Get the brightness value of the image.

The obtained luminance Sa value is shown in Table 2 described later. 3-5. Calculation of refractive index Regarding samples a and c, the surface 10a of the structure 10 was polished so that the two-dimensional average surface roughness Ra was 0.1 μm or less, and the thickness of the structure 10 was 1 μm or less.

The measurement of the refractive index was carried out by using a spectroscopic film thickness meter (OPTM-F2 manufactured by Otsuka Electronics and FE-37S manufactured by Otsuka Electronics), and the refractive index was calculated by reflection spectroscopy. The measurement conditions were a measurement spot size of 10 μm and a measurement wavelength range of 360 to 1100 nm.

The analysis conditions are based on an analysis wavelength range of 360 to 1100 nm, and an optimization method and a least square method are used. The refractive index of each sample for each wavelength is shown in Table 3 and FIG. 19 described later.

As shown in Table 3 and FIG. 19, the refractive index of the structure including the YO compound according to the present invention is larger than that of the conventionally known Y ₂ O ₃ by 1.92, and has a high resistance to particulates.

4. Structural property evaluation 4-1. Plasma resistance evaluation Regarding samples a to n, [reference plasma resistance test] was performed.

Specifically, the test used an inductively coupled plasma reactive ion etching device (Muc-21 Rv-Aps-Se / Sumitomo Precision Industry) as the plasma etching device. The conditions for plasma etching are as follows: 1500W for ICP output as power output system, 750W for bias output, 750W for CHF ₃ gas and 10ccm O ₂ gas as process gas system, pressure 0.5Pa, plasma The etching time is 1 hour.

The state of the surface 10a of the structure 10 after the plasma irradiation was captured by SEM. These are shown in Figure 14. The SEM observation conditions were 5000 times magnification and 3 kV acceleration voltage.

Next, the area of the corrosion marks on the surface after the plasma irradiation was calculated from the obtained SEM image. The results are shown in Table 2. Then, the state of the surface 10a of the structure 10 after the plasma irradiation was photographed with a laser microscope. Specifically, a laser microscope [OLS4500 / Olympus] was used, and the objective lens was MPLAPON100xLEXT (numerical aperture 0.95, working distance 0.35 mm, focal spot diameter 0.52 μm, and measurement area 128 × 128 μm), and the magnification was 100 times. The λc filter from which the moiré component was removed was set to 25 μm. The measurement was performed at arbitrary three places, and the average value was used as the arithmetic mean height Sa. In addition, it is appropriate to refer to the international standard for three-dimensional surface properties ISO25178. Table 4 shows the arithmetic mean height Sa of the surface 10a after the plasma irradiation.

FIG. 15 is a graph of the area of corrosion marks (μm ² ) for each sample. As shown in FIG. 15, the sample formed by the aerosol deposition method generally has a smaller area of corrosion marks than the sample (e) formed by ion plating, and has good results.

FIG. 16 shows the relationship between the brightness Sa and the area of the corrosion mark (μm ² ) for each sample formed by the aerosol deposition method. As shown in FIG. 16, it is understood that the area of the corrosion mark can be significantly reduced by the brightness Sa below a predetermined value.

Furthermore, as shown in FIG. 17 and Table 2, it is understood that the particle resistance can be improved by reducing the amount of hydrogen (the number of hydrogen atoms per unit volume / the concentration of hydrogen atoms) on the surface of the structure.

Table 1

Table 2

table 3

Table 4

10‧‧‧ Structure

10a‧‧‧Surface of structure

10u‧‧‧upper area

10b‧‧‧lower area

10c‧‧‧Microcrystalline

11‧‧‧Gas Supply Department

12‧‧‧Piping

13‧‧‧ Aerosol Supply Department

14‧‧‧ Reaction Room

15‧‧‧ Nozzle

16‧‧‧ platform

17‧‧‧Driver

18‧‧‧Exhaust

19‧‧‧ device

30a‧‧‧ Plasma irradiation area

30b‧‧‧ Plasma non-irradiated area

31‧‧‧hole

40‧‧‧ Obtaining samples for measuring Sa

50‧‧‧carbon layer

60‧‧‧Tungsten layer

70‧‧‧ substrate

70a‧‧‧ substrate surface

81‧‧‧ primary particle

82‧‧‧ secondary particles

90‧‧‧TEM observation sample

90h‧‧‧Sample height

90u‧‧‧Thickness of upper part of sample

90b‧‧‧Thickness of sample

90w‧‧‧sample width

100‧‧‧ composite structure

301, 302‧‧‧ semiconductor manufacturing device components

dL‧‧‧Regional length

dW‧‧‧Regional horizontal length

G‧‧‧Image area

Gl‧‧‧Image length

Gw‧‧‧Image horizontal length

L‧‧‧ vertical

R‧‧‧ Brightness acquisition area

W‧‧‧Horizontal

FIG. 1 is a schematic cross-sectional view of a composite structure 100 according to the present invention. FIG. 2 is a flowchart showing a method for evaluating the brightness Sa according to the present invention. FIG. 3-1, FIG. 3-2, and FIG. 3-3 are schematic diagrams of the TEM observation sample 90. FIG. FIG. 4 is a schematic diagram showing a TEM image G of the structure 10. FIG. 5 is a diagram showing a TEM image G and a brightness value of each pixel. FIG. 6 is a diagram showing brightness correction of a TEM image G. FIG. FIG. 7-1 and FIG. 7-2 are diagrams showing brightness values in the brightness acquisition area R. FIG. FIG. 8 is a schematic diagram showing an example of a case where the composite structure 100 is used as a semiconductor manufacturing device member 301. FIG. 9 is a schematic diagram showing an example of a case where the composite structure 100 is used as a semiconductor manufacturing device member 302. FIG. 10 is a schematic diagram showing an example of a device configuration for an aerosol deposition method. FIG. 11 is a TEM image of the structure 10 at a magnification of 400,000 times. 12-1, 12-2, 12-3, and 12-4 are scanning electron microscope (SEM) images of the structure 10. Figures 13-1, 13-2, 13-3, 13-4, 13-5, 13-6, 13-7, 13-8, 13-9, and 13-10 are A transmission electron microscope (TEM) image G of the structure 10. Fig. 14-1, Fig. 14-2, Fig. 14-3, Fig. 14-4, Fig. 14-5, Fig. 14-6, Fig. 14-7, and Fig. 14-8 show the structure 10 after the benchmark plasma resistance test. Scanning electron microscope (SEM) image of the surface. FIG. 15 is a graph showing the area of corrosion marks on the surface of the structure 10 after the reference plasma resistance test. FIG. 16 is a graph showing the relationship between the brightness Sa on the surface 10 a of the structure 10 and the area of the corrosion mark. FIG. 17 is a graph showing the relationship between the amount of hydrogen on the surface 10a of the structure 10 and the area of the corrosion mark. Fig. 18 is a schematic sectional view for explaining the microstructure of the composite structure. FIG. 19 is a graph showing the relationship between the wavelength and the refractive index of the structure 10. FIG. 20 is a graph showing a wavelength dispersion of a refractive index of a yttrium oxide sintered body related to a conventional technique.

Claims (28)

A composite structure includes a substrate and a structure provided on the substrate and having a surface. The structure includes a polycrystalline ceramic, and a secondary ion mass spectrometry (Dynamic-Secondary Ion Mass Spectrometry_D-SIMS method) is provided. The number of hydrogen atoms per unit volume in any one of measurement depths of 500 nm and 2 μm is 7 * 10 ²¹ atoms / cm ³ or less. For example, the composite structure of item 1 of the patent application scope, wherein the number of hydrogen atoms per unit volume of the structure is 5 * 10 ²¹ atoms / cm ³ or less. A composite structure includes a substrate and a structure provided on the substrate and having a surface. The structure includes a polycrystalline ceramic, and is reversed by hydrogen forward scattering spectroscopy (HFS) -Rasefort. The hydrogen atom concentration measured by scattering spectroscopy (RBS) and proton-hydrogen forward scattering spectroscopy (p-RBS) is 7 atomic% or less. A composite structure includes a base material and a structure provided on the base material and having a surface, characterized in that the structure is made of polycrystalline ceramic, and a brightness Sa value calculated by the following method is Below 19, a method for obtaining the brightness Sa includes: (i) a procedure for preparing a transmission electron microscope (TEM) observation sample of the structure; (ii) preparing a digital black and white image of a bright field image of the TEM observation sample (Iii) a procedure for obtaining the luminance value of the color data of each pixel in the digital black-and-white image by using a hue value; (iv) a procedure for correcting the luminance value; (v) using the corrected luminance value A program for calculating the brightness Sa. In the program (i), at least three TEM observation samples are prepared from the structure, and each of the at least three TEM observation samples is suppressed using a focused ion beam method (FIB method). It is made by processing damage. During the FIB processing, a carbon layer and a tungsten layer for preventing charging and sample protection are provided on the surface of the structure. When the FIB processing direction is a longitudinal direction, the structure is a plane perpendicular to the longitudinal direction. Long in the minor axis direction of the surface The thickness of the upper part of the sample is 100 ± 30nm. In this procedure (ii), the digital black-and-white image is obtained for each of the at least three TEM observation samples, and each of the digital black-and-white image uses a transmission electron microscope. (TEM), at a magnification of 100,000 times, and an acceleration voltage of 200 kV, including the structure, the carbon layer, and the tungsten layer, in each of the digital black and white images, a distance from the surface of the structure is set to 0.5 in the longitudinal direction. μm is the luminance acquisition area of the longitudinal length of the area, and a plurality of the digital black and white images are obtained from each of the at least three TEM observation samples so that the total area of the luminance acquisition area is 6.9 μm ² or more. In this program (iv ), Regarding the brightness value, the brightness value of the carbon layer is 255, and the brightness value of the tungsten layer is relative corrected to obtain a corrected brightness value. In the program (v), the brightness value is corrected. Each area is acquired, and an average of the absolute values of the differences in the luminance values after the correction is calculated for each pixel using the least square method, and the average of these is used as the luminance Sa. For example, the composite structure of item 4 of the patent application scope, wherein the brightness Sa is 13 or less. A composite structure includes a base material and a structure provided on the base material and having a surface, characterized in that the structure is made of polycrystalline ceramic, and a brightness Sa value calculated by the following method is The method for obtaining the brightness Sa below 10 includes: (i) a procedure for preparing a transmission electron microscope (TEM) observation sample of the structure; (ii) obtaining a digital black and white image of a bright field image of the TEM observation sample (Iii) a procedure for obtaining the luminance value of the color data of each pixel in the digital black-and-white image by using a hue value; (iv) a procedure for correcting the luminance value; (v) using the corrected luminance value A program for calculating the brightness Sa. In the program (i), at least three TEM observation samples are prepared from the structure, and each of the at least three TEM observation samples is suppressed using a focused ion beam method (FIB method). It is made by processing damage. During the FIB processing, a carbon layer and a tungsten layer for preventing charging and sample protection are provided on the surface of the structure. When the FIB processing direction is a longitudinal direction, the structure is a plane perpendicular to the longitudinal direction. Long in the minor axis direction of the surface The thickness of the upper part of the sample is 100 ± 30nm. In this procedure (ii), the digital black-and-white image is obtained for each of the at least three TEM observation samples, and each of the digital black-and-white image uses a transmission electron microscope. (TEM), at a magnification of 100,000 times, and an acceleration voltage of 200 kV, including the structure, the carbon layer, and the tungsten layer, in each of the digital black and white images, a distance from the surface of the structure is set to 0.5 in the longitudinal direction. μm is the luminance acquisition area of the longitudinal length of the area, and a plurality of the digital black and white images are obtained from each of the at least three TEM observation samples so that the total area of the luminance acquisition area is 6.9 μm ² or more. In this program (iv ), Regarding the brightness value, the brightness value of the carbon layer is 255, and the brightness value of the tungsten layer is relative corrected to obtain the corrected brightness value. Regarding the digital black-and-white image with the brightness value corrected, To perform noise removal using a low-pass filter, and the cut-off frequency in noise removal using the low-pass filter is 1 / (10 pixels), and in this program (v), for each of the brightness acquisition regions, Calculate using least square method The average of absolute difference in luminance value of the pixel to the corrected each order such as the average luminance Sa. For example, the composite structure of the sixth item of the patent application, wherein the brightness Sa is 5 or less. A composite structure includes a substrate and a structure provided on the substrate and having a surface. The structure includes a polycrystalline ceramic including Y (yttrium element) and O (oxygen element), and a wavelength of 400 nm to The refractive index at 550 nm is greater than 1.92. This refractive index is calculated by reflection spectroscopy using a microspectrographic film thickness meter. As a measurement condition, the measurement spot size is 10 μm, and the average surface of the substrate surface and the composite structure surface is rough. The degree Ra is less than or equal to 0.1 μm, the thickness of the structure is less than or equal to 1 μm, the measurement wavelength range is 360 to 1100 nm, and the analysis wavelength range is 360 to 1100 nm. An optimization method and a least square method are adopted. A composite structure includes a substrate and a structure provided on the substrate and having a surface. The structure includes a polycrystalline ceramic including Y (yttrium element) and O (oxygen element), and a refractive index thereof. Satisfy at least one of 1.99 or more at 400nm, 1.96 or more at 500nm, 1.94 or more at 600nm, 1.93 or more at 700nm, or 1.92 or more at 800nm or more. The refractive index is a microspectrographic film thickness meter. Calculated by reflection spectroscopy, the measurement condition is a measurement spot size of 10 μm, the average surface roughness Ra of the substrate surface and the surface of the composite structure Ra ≦ 0.1 μm, the thickness of the structure ≦ 1 μm, and the measurement wavelength range is 360 to 1100 nm. Analytical conditions are analytical wavelengths ranging from 360 to 1100 nm, and optimization and least square methods are used. An evaluation method is an evaluation method for a microstructure of a structure including a polycrystalline ceramic having a surface, comprising: (i) a procedure for preparing a transmission electron microscope (TEM) observation sample of the structure; (ii) preparing A program for observing a digital black-and-white image of a bright-field image of a sample by the TEM; (iii) a program for obtaining a brightness value of a color data of each pixel in the digital black-and-white image by a hue value; (iv) correcting the brightness value (V) a program for calculating the brightness Sa using the corrected brightness value. In the program (i), at least three TEM observation samples are prepared from the structure, and the at least three TEM observation samples are prepared. Each of the systems is manufactured using a focused ion beam method (FIB method) to suppress processing damage. During this FIB processing, a carbon layer and a tungsten layer are provided on the surface of the structure. When the FIB processing direction is a vertical direction, the vertical direction is perpendicular to the vertical direction. The thickness of the upper part of the sample in the minor axis direction of the surface of the structure on the plane is 100 ± 30 nm. In this procedure (ii), the digital black and white image is obtained for each of the at least 3 TEM observation samples. Departments of Digital Black and White Images Using a transmission electron microscope (TEM), the structure, the carbon layer, and the tungsten layer are included at a magnification of 100,000 times and an acceleration voltage of 200 kV. In each of the digital black and white images, the distance from the surface of the structure is set. In the longitudinal direction, a luminance acquisition area having a longitudinal length of 0.5 μm is taken as an area, and a plurality of the digital black and white images are obtained from each of the at least three TEM observation samples so that the total area of the luminance acquisition area becomes 6.9 μm ² or more. In the program (iv), regarding the brightness value, the brightness value of the carbon layer is 255, and the brightness value of the tungsten layer is 0. Relative correction is performed to obtain the corrected brightness value, and in the program (v ), For each of the luminance acquisition regions, the average of the absolute values of the differences in the corrected luminance values for each of the pixels is calculated using the least square method, and the average of these is used as the luminance Sa. For example, the evaluation method of the tenth item of the patent application, wherein the program (iv) further includes a digital black and white image corrected for the luminance value, and a program for removing noise using a low-pass filter is used. The cut-off frequency in the noise removal of this low-pass filter is 1 / (10 pixels). For example, the evaluation method of the 10th or 11th in the scope of patent application, wherein the method of suppressing the processing damage is at least any of the following: finishing processing at a low voltage of 5kV; removing the processing damage by Ar ions; The surface was cleaned by ion milling before observation. For example, the evaluation method of the 10th or 11th in the scope of patent application, wherein the thickness of the upper part of the sample is obtained by observing the TEM with a secondary electron image using a scanning electron microscope (SEM). The conditions are a magnification of 200,000 times, an acceleration voltage of 2 kV, a scanning time of 40 seconds, and an image number of 2560 * 1920 pixels. The SEM image constitutes a plane perpendicular to the longitudinal direction. For example, the evaluation method of the 10th or 11th in the scope of patent application, wherein the at least 3 TEM observation samples are equally obtained from the surface of the structure. For example, the evaluation method of the 10th or 11th in the scope of the patent application, wherein in the acquisition of the digital black and white image, the digital black and white image is obtained by focusing adjustment at a magnification of 300,000 or more times. For example, the evaluation method of the 10th or 11th in the scope of patent application, wherein in the procedure (iv), the brightness value of the tungsten layer before correction is included in the brightness value of the tungsten layer in the digital black and white image, From the minimum value, the average value of the brightness values of 10,000 pixels is used continuously in small order. For the brightness value of the carbon layer before correction, the brightness value of the carbon layer in the digital black and white image is continuously determined by the maximum value. The average value of the luminance values of 100,000 pixels is used in the largest order. For example, the evaluation method of the 10th or 11th in the scope of patent application, wherein in the brightness acquisition area, the horizontal length of the area perpendicular to the vertical length of the area is set in such a way that it becomes the longest in the digital black and white image. For example, the evaluation method of item 10 or item 11 of the patent application, wherein when the plurality of digital black and white images are obtained from each of the at least 3 TEM observation samples, the plurality of digital black and white images are The black-and-white image is obtained in such a manner that the vertical direction and the horizontal direction are continuous. For example, the evaluation method of item 10 or item 11 of the patent application scope, wherein image analysis software is used in each of the procedures (iii) to (iv). For example, the composite structure according to any one of claims 1 to 9, wherein the average crystallite size of the polycrystalline ceramic calculated from a TEM image with a magnification of 400,000 to 2 million times is 3 nm to 50 nm. . For example, the composite structure according to any one of claims 1 to 9, wherein the crystallite size is 30 nm or less. For example, the composite structure in the scope of patent application No. 21, wherein the crystallite size is 5 nm or more. For example, the composite structure according to any one of claims 1 to 9, wherein the structure is selected from oxides, fluorides and acid fluorides of rare earth elements and mixtures thereof. For example, the composite structure of item 23 of the patent application, wherein the rare earth element is selected from the group consisting of Y, Sc, Yb, Ce, Pr, Eu, La, Nd, Pm, Sm, Gd, Tb, Dy, Ho, Er At least one of the group consisting of Tm, Tm and Lu. For example, the composite structure according to any one of the first to the ninth aspects of the patent application scope, wherein the structure shows an arithmetic average height Sa of 0.060 or less after the reference plasma resistance test. For example, the composite structure according to any one of claims 1 to 9 of the scope of application for a patent, wherein the composite structure is used in an environment requiring resistance to particles. A semiconductor manufacturing device includes a composite structure according to any one of claims 1 to 9, and 21 to 26. A display manufacturing device includes a composite structure according to any one of claims 1 to 9, and 21 to 26.