Holding device

Abstract

The invention provides a holding device. The holding device comprises a plate-shaped member mainly composed of high-purity ceramic and a holding substrate having excellent adhesion to an electrode layer disposed inside the plate-shaped member. This holding device is provided with a holding substrate having: a plate-shaped member containing a ceramic as a main component, the plate-shaped member including a first surface for holding an object and a second surface disposed on the opposite side of the first surface; and an electrode layer that is disposed inside the plate-shaped member and contains a conductive material as a main component and a ceramic as an auxiliary component. The plate-like member includes: a first dielectric layer disposed on the first surface side of the electrode layer and having a main component content of 99 mass% or more; and a second dielectric layer disposed on the second surface side of the electrode layer, the second dielectric layer having a main component content of 99 mass% or more. The surface of the electrode layer on the first surface side has a fractal dimension D of 1.18 or more.

Description

Technical Field

[0001] The invention relates to a retaining device. Background Art

[0002] An electrostatic chuck is used as a holding device for holding a wafer (semiconductor wafer) when manufacturing semiconductors. The electrostatic chuck has a holding substrate (ceramic substrate) mainly composed of alumina, which is an insulating ceramic, and holds the wafer on the surface of the holding substrate by electrostatic attraction. The electrostatic attraction is generated by applying a voltage to an electrode layer (chuck electrode) provided inside the holding substrate.

[0003] In this type of electrostatic chuck, in recent years, high-power high-frequency power is required to be applied in plasma processing such as plasma etching. Therefore, as a material constituting the base material (plate-shaped component) of the holding substrate, high-purity aluminum oxide (purity 99.99%) with excellent plasma resistance is used (for example, refer to patent document 1). In addition, if the base material of the holding substrate is composed of high-purity aluminum oxide, the adhesion between the base material and the electrode layer formed therein may be reduced. Therefore, in the case of a holding substrate in which the base material is composed of high-purity aluminum oxide, when forming the electrode layer, in addition to conductive powder (Pd powder) and adhesive, a metallized paste to which aluminum oxide powder is added for the purpose of improving adhesion to the base material is used. In the electrode layer composed of such a metallized paste, a three-dimensional mesh network composed of aluminum oxide is formed that connects one side with the other side in the thickness direction, so that the adhesion between the electrode layer and the base material can be ensured.

[0004] Prior art literature

[0005] Patent Literature

[0006] Patent Document 1: Japanese Patent No. 5441020 Summary of the invention

[0007] Problems to be solved by the invention

[0008] It is required to ensure close contact between the base material (plate-shaped member) and the electrode layer.

[0009] Furthermore, when forming the electrode layer, if the amount of alumina powder used is increased, the adhesion between the electrode layer and the base material of the support substrate is improved, but there is a problem that the thermal conductivity of the support substrate (particularly near the electrode layer) is reduced.

[0010] When one surface (upper surface) of the holding substrate is heated during plasma treatment, heat moves from one surface to the other surface (lower surface) inside the holding substrate. A base member is arranged on the other surface, and when the base member is heated, the base member is cooled by a cooling mechanism (refrigerant flow path, etc.) provided inside. However, as described above, if the amount of aluminum oxide powder used when forming the electrode layer increases, a portion with low thermal conductivity composed of the aluminum oxide powder is formed inside the electrode layer, and this portion hinders the movement of heat in the holding substrate, which causes a decrease in the heat removal performance of the holding substrate.

[0011] An object of the present invention is to provide a holding device including a plate-shaped member mainly composed of high-purity ceramics and a holding substrate having excellent adhesion to an electrode layer arranged inside the plate-shaped member.

[0012] Another object of the present invention is to provide a holding device including a holding substrate having excellent adhesion between a plate-shaped member mainly composed of high-purity ceramics and an electrode layer disposed therein and excellent heat removal properties (thermal conductivity).

[0013] Technical solutions to solve problems

[0014] The means for solving the above-mentioned problems are as follows. That is, <1> a holding device, comprising a holding substrate, the holding substrate having: a plate-like component, including a first surface of a holding object and a second surface arranged on the opposite side of the first surface, and containing ceramic as a main component; and an electrode layer, arranged inside the plate-like component, containing a conductive material as a main component and containing the ceramic as a secondary component, the plate-like component comprising: a first dielectric layer, arranged on the first surface side of the electrode layer, and the content of the main component is 99% by mass or more; and a second dielectric layer, arranged on the second surface side of the electrode layer, and the content of the main component is 99% by mass or more, and the fractal dimension D of the surface on the first surface side of the electrode layer is 1.18 or more.

[0015] <2> The holding device according to <1> above, wherein, in a cross section of the electrode layer along a thickness direction, an area ratio of the ceramic relative to an observation range set on a center side of the electrode layer is 30% or less.

[0016] <3> The holding device according to <1> or <2>, wherein, in the cross section of the electrode layer, the ceramic does not exist in a form connecting the first dielectric layer and the second dielectric layer.

[0017] <4> The holding device according to any one of <1> to <3>, wherein an area ratio of the electrode layer to the first surface is 80% or more in a state where the plate-like member is viewed in plan.

[0018] Effects of the Invention

[0019] According to the present invention, it is possible to provide a holding device including a plate-shaped member mainly composed of high-purity ceramics and a holding substrate having excellent adhesion to an electrode layer disposed inside the plate-shaped member.

[0020] According to the present invention, a holding device can be provided that includes a holding substrate having excellent adhesion between a plate-like member mainly composed of high-purity ceramics and an electrode layer disposed therein and excellent heat removal properties (thermal conductivity). BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is an explanatory diagram schematically showing the cross-sectional structure of the holding device according to the first embodiment.

[0022] Figure 2 This is an explanatory diagram schematically showing a state in which a holding substrate (plate-shaped member) of the holding device according to the first embodiment is viewed from above.

[0023] Figure 3 The figure shows a SEM image of a cross section of an electrode layer in a plate-like member along the thickness direction.

[0024] Figure 4 This is a diagram showing a graph in which the box pixel size (d) is plotted against the number of boxes (N(d)) on the basis of the contour line of the surface extracted from the cross section of the electrode layer using the common logarithm.

[0025] Figure 5 It is an explanatory diagram schematically showing a method for manufacturing a holding substrate.

[0026] Figure 6 This is a diagram showing a SEM image of a cross section along the thickness direction of a plate-like member included in the holding substrate according to the second embodiment.

[0027] Figure 7 It is an explanatory diagram showing an observation range set on the center side of the electrode layer in a cross-sectional SEM image.

[0028] Figure 8 FIG. 1 is an explanatory diagram showing image processing performed when the fractal dimension D is obtained.

[0029] Fig. 9 This is an explanatory diagram schematically showing a test body used for measuring the adhesion strength.

[0030] Fig.10This is a diagram showing a graph summarizing the results of measuring the adhesive strength (N) of the electrode layer of each Example and each Comparative Example.

[0031] Fig.11 This is a diagram showing a graph summarizing the relationship between the fractal dimension D of the electrode layer and the adhesion of the electrode layer in the example and each comparative example.

[0032] Fig.12 This is a diagram showing a graph summarizing the relationship between the surface roughness Ra of the electrode layer and the adhesion of the electrode layer in the example and each comparative example. DETAILED DESCRIPTION

[0033] <Implementation Method 1>

[0034] Below, refer to Figure 1~Figure 5 The holding device 100 according to Embodiment 1 will be described. The holding device 100 is an electrostatic chuck that attracts and holds an object (wafer W) by electrostatic attraction. The electrostatic chuck is used as a stage for mounting the wafer W in a process of etching using plasma in a depressurized chamber, for example.

[0035] Figure 1 1 is an explanatory diagram schematically showing a cross-sectional structure of a holding device 100 according to Embodiment 1. The holding device 100 includes a disk-shaped holding substrate (ceramic substrate) 10 and a disk-shaped base member 20 that is larger than the holding substrate 10. For example, when the holding substrate 10 is in the shape of a disk with a diameter of 300 mm and a thickness of 3 mm, the base member 20 is set to be in the shape of a disk with a diameter of 340 mm and a thickness of 20 mm.

[0036] The holding substrate 10 and the base member 20 overlap each other in the vertical direction with the holding substrate 10 arranged on the upper side and the base member 20 arranged on the lower side. The holding substrate 10 and the base member 20 are bonded to each other with a bonding material 30 interposed therebetween.

[0037] The holding substrate 10 has: a first surface S1 of a substantially circular shape, arranged on the upper side; and a second surface S2 of a substantially circular shape, arranged on the opposite side (i.e., the lower side) of the first surface S1 and facing the base member 20. The base member 20 has: a third surface S3 of a substantially circular shape, arranged on the upper side and facing the second surface S2; and a fourth surface S4 of a substantially circular shape, arranged on the opposite side (i.e., the lower side) of the third surface S3. The above-mentioned bonding material 30 is sandwiched between the second surface S2 of the holding substrate 10 and the third surface S3 of the base member 20, and is in a state of spreading in layers.

[0038] The holding substrate 10 includes a disk-shaped plate-like member (base material) 11 and an electrode layer 12 disposed inside the plate-like member 11. The upper surface of the plate-like member 11 is a first surface S1 of the holding substrate 10, and the lower surface (back surface) of the plate-like member 11 is a second surface S2 of the holding substrate 10.

[0039] The plate-like component 11 is an insulating component in the form of a plate (disk) containing ceramic as a main component. In this specification, "main component" refers to the component with the highest content. In the case of the present embodiment, the content of the main component in the plate-like component 11 is 99% by mass or more (preferably 99.5% by mass or more). In addition, in the case of the present embodiment, the main component of the plate-like component 11 is aluminum oxide (Al2O3). That is, the plate-like component 11 of the present embodiment contains high-purity aluminum oxide as a main component. In addition, as long as it does not impair the purpose of the present invention, in other embodiments, the main component of the plate-like component may also be other ceramics such as aluminum nitride (AlN).

[0040] like Figure 1 As shown, the plate-like component 11 includes: a first dielectric layer 111, which is arranged on the first surface S1 side of the electrode layer 12, and the content of the main component (aluminum oxide) is 99 mass% or more (preferably 99.5 mass% or more); and a second dielectric layer 112, which is arranged on the second surface S2 side of the electrode layer 12, and the content of the main component (aluminum oxide) is 99 mass% or more (preferably 99.5 mass% or more).

[0041] The electrode layer 12 is disposed inside the plate-shaped member 11 , and contains a conductive material as a main component and ceramics as a subcomponent.

[0042] The electrode layer 12 is generally in a planar shape (layered shape) substantially parallel to the first surface S1. Figure 1 As shown, the electrode layer 12 is arranged on the first surface S1 side inside the holding substrate 10 (plate-shaped member 11 ).

[0043] The electrode layer 12 contains a conductive material such as tungsten (W), molybdenum (Mo), platinum (Pt), palladium (Pd) as a main component. In addition, the electrode layer 12 contains a ceramic used as a main component of the plate-like member 11 as a subcomponent. In the case of the present embodiment, the electrode layer 12 contains aluminum oxide (Al2O3) as a subcomponent.

[0044] The electrode layer 12 of this embodiment is a chuck electrode, and is connected to an external power source via a terminal (not shown) etc. When power is supplied to the electrode layer 12 , electrostatic attraction is generated, and the wafer W is attracted and held on the first surface S1 of the holding substrate 10 by the electrostatic attraction.

[0045] Figure 21 is an explanatory diagram schematically showing a state in which the holding device 100 of the first embodiment holds the substrate 10 (plate-shaped member 11) as viewed from above. Figure 2 As shown, in a state where the plate-like member 11 is viewed from above, the area ratio of the electrode layer 12 to the first surface S1 is 80% or more. Figure 2 As shown, the electrode layer 12 has a substantially circular shape in a plan view, and is arranged inside the plate-like member 11 so as to overlap with most of the first surface S1 .

[0046] Figure 3 1 is a diagram showing a SEM image of a cross section of the electrode layer 12 in the plate-like member 11 along the thickness direction. Figure 3 As shown, in the present embodiment, the surface 12a on the first surface S1 side (upper side) of the electrode layer 12 is roughened. That is, the surface 12a of the electrode layer 12 is a concavoconvex surface including a plurality of convex portions and a plurality of concave portions.

[0047] The fractal dimension D of the surface 12 a on the first surface S1 side of the electrode layer 12 is 1.18 or more. The fractal dimension D of the surface 12 a of the electrode layer 12 is obtained by the box counting method.

[0048] The fractal dimension D being 1.18 or more means that the shape of the surface 12a of the electrode layer 12 is complex. The fractal dimension is an index of geometric complexity, and in the present embodiment, the fractal dimension when the electrode layer 12 is observed in cross section (i.e., the two-dimensional fractal dimension) is specified. In addition, in general, it is known that in the two-dimensional fractal dimension, for example, the fractal dimension of a simple shape such as a perfect circle, a square, or a rectangle is approximately 1, and the more complex the shape such as a shape having a plurality of concave and convex shapes, the larger the fractal dimension and the closer it is to 2.

[0049] Here, we briefly explain the analysis method (principle) of fractal dimension based on the box counting method. For example, when a figure existing in a plane is divided into squares with a side length of d, if the figure is covered by N (d) squares, then between N (d) and d, N (d) = ad -D When the relationship (a is a positive integer)…(I) holds true, D is defined as the fractal dimension of the figure.

[0050] Furthermore, taking the logarithm of both sides of the above formula (I), we get log 10 N(d)=-D·log 10 d+log 10 a (a is a positive integer)…(II), if a double logarithmic graph of d and N(d) is plotted, the fractal dimension D can be calculated based on the slope of the straight line.

[0051] Figure 4This is a graph showing a common logarithmic plot of the box pixel size (d) versus the number of boxes (N(d)) based on the contour of the surface 12a extracted from the cross section of the electrode layer 12. The details of the method for obtaining the fractal dimension by the box counting method will be described later.

[0052] In the case of the present embodiment, the surface 12b on the second surface S2 side (lower side) of the electrode layer 12 is not roughened, and has smaller irregularities than the surface 12a, and is a flat surface. That is, the electrode layer 12 of the present embodiment is roughened in such a manner that at least the surface 12a on the first surface S1 side on which the wafer W is mounted becomes an irregular surface represented by a desired fractal dimension D.

[0053] In the present embodiment, in the cross section of the electrode layer 12 along the thickness direction, the area ratio of the ceramic relative to the observation range set on the center side of the electrode layer 12 is 30% or less.

[0054] like Figure 3 As shown in the SEM image of , the conductive material 40, which is the main component of the electrode layer 12, exists in a layered manner. Moreover, there are a plurality of bulk portions of the ceramic (alumina) 50 inside the layered conductive material 40. However, in the case of the present embodiment, in the cross section of the electrode layer 12, the ceramic 50 does not exist in a form connecting the first dielectric layer 111 and the second dielectric layer 112. That is, in the case of the present embodiment, the ceramic 50 does not exist in a form penetrating the electrode layer 12 (conductive material) in the thickness direction. In addition, the upper surface of the layered conductive material 40 becomes the surface 12a of the electrode layer 12, and the lower surface of the conductive material 40 becomes the surface 12b of the electrode layer 12.

[0055] If the area ratio of the ceramic is 30% or less (preferably 25% or less), the electrode layer 12 has excellent thermal conductivity. In addition, the calculation method of the area ratio of the ceramic in the cross section of the electrode layer 12 along the thickness direction will be described later. As the lower limit of the area ratio, from the viewpoint of ensuring the close contact between the electrode layer 12 and the plate-like member 11, it is preferably 3% or more, and more preferably 5% or more.

[0056] In addition, the holding substrate 10 may have other structures such as a gas flow path for supplying an inert gas (for example, helium gas as a heat transfer gas) to the first surface S1 side.

[0057] The base member 20 is mainly composed of, for example, metal (aluminum, aluminum alloy, etc.), a composite of metal and ceramics (Al—SiC), or ceramics (SiC).

[0058] A refrigerant flow path 21 as a cooling mechanism is provided inside the base member 20. The plasma heat is cooled by flowing a refrigerant (e.g., a fluorine-based inert liquid, water, etc.) in the refrigerant flow path 21. In addition, when the refrigerant flows in the refrigerant flow path 21, the base member 20 is cooled, and the substrate 10 is cooled by heat transfer (heat dissipation) between the base member 20 and the substrate 10 via the bonding material 30. As a result, the wafer W held on the first surface S1 of the substrate 10 is cooled. In addition, by appropriately adjusting the refrigerant flow rate in the refrigerant flow path 21, the temperature of the wafer W held on the first surface S1 can be controlled.

[0059] In addition, the base member 20 may include other structures such as a gas flow path for supplying an inert gas.

[0060] The bonding material 30 is composed of, for example, a bonding sheet including a silicone-based organic bonding agent, an inorganic bonding agent, or an Al-based metal bonding agent. As the bonding material, a material having high bonding force to both the substrate 10 and the base member 20 and having high pressure resistance and thermal conductivity is preferred. In addition, a gas flow path for supplying an inert gas may be formed in the bonding material 30 as needed.

[0061] Next, an example of a method for manufacturing the holding device 100 of the present embodiment will be described. Figure 5 A method for manufacturing the holding substrate 10 included in the holding device 100 will be described. Figure 5 1 is an explanatory diagram schematically showing a method for manufacturing the support substrate 10. The method for manufacturing the support substrate 10 employs a sheet lamination method using green sheets (ceramic green sheets).

[0062] First, if Figure 5 As shown in (A), in order to form the first dielectric layer 111 of the plate-like component 11, a plurality of green sheets are stacked to prepare a first stack 111X. The slurry for the green sheet can be obtained by, for example, mixing a mixture including alumina powder, acrylic binder, dispersant, plasticizer, sintering aid (e.g., glass, rare earth oxide), etc., with an organic solvent added thereto using a ball mill. The slurry is formed into a sheet using a casting device, and then the obtained molded product is dried to obtain a plurality of green sheets.

[0063] Then, if Figure 5As shown in (B), the surface 111Xa of the first stack 111X is subjected to a surface roughening treatment. The surface roughening treatment is performed for the purpose of roughening the surface 12a of the electrode layer 12 finally obtained into a concave-convex surface represented by the desired fractal dimension D. The surface roughening treatment can be performed directly on the surface of the metallization paste 12X used to form the electrode layer 12, or it can be performed on the surface 111Xa of the object side (the first stack 111X) in contact with the surface. Here, as an example of surface roughening treatment, the case where it is performed on the surface 111Xa of the first stack 111X on the other side is illustrated.

[0064] The surface roughening treatment is not particularly limited as long as the desired fractal dimension D (1.18 or more) can be obtained, and examples thereof include sandblasting, sputtering, corona (discharge) treatment, flame treatment, ultraviolet irradiation treatment, electron beam irradiation treatment, and the like.

[0065] Afterwards, if Figure 5 As shown in (C), a metallization paste 12X for forming an electrode layer 12 is stacked on the surface 111Xa of the first stack 111X after the surface roughening treatment. The metallization paste 12X is obtained by adding conductive powders (main components) such as tungsten and molybdenum to a mixture of aluminum oxide powder (subcomponent), acrylic adhesive, and organic solvent and kneading them. The metallization paste 12X is formed into a layer on the surface 111Xa of the first stack 111X using a screen printing device, for example. If the metallization paste 12X is formed into a layer in this way, the surface of the first stack 111X that contacts the surface 111Xa eventually becomes a concave-convex surface represented by a desired fractal dimension D (above 1.18).

[0066] In a plan view of the metallizing paste 12X on the first laminate 111X, a frame-shaped green sheet (not shown) surrounding the metallizing paste 12X is arranged around the metallizing paste 12X. The frame-shaped green sheet is arranged at a portion of the surface 111Xa of the first laminate 111X where the metallizing paste 12X is not formed.

[0067] Then, if Figure 5 As shown in (D), a second laminate 112X for forming the second dielectric layer 112 of the plate-like member 11 is prepared, and the second laminate 112X is laminated on the metallization paste 12X on the first laminate 111X.

[0068] The second laminate 112X is formed by laminating a plurality of green sheets similarly to the first laminate 111X. The second laminate 112X thus formed is laminated on the metallizing paste 12X on the first laminate 111X with the metallizing paste 12X sandwiched between the second laminate 112X and the first laminate 111X. These laminates are thermocompression bonded to each other.

[0069] In addition, the outer periphery of the laminate composed of the first laminate 111X, the metallization paste 12X and the second laminate 112X can also be appropriately cut. Then, the laminate is cut by mechanical processing to obtain a disk-shaped molded body. Next, the obtained molded body is degreased and sintered, and then the molded body after degreasing and sintering is sintered (formal sintering) to obtain a sintered body. Thereafter, the surface of the sintered body is appropriately processed, ground, etc. to obtain a Figure 5 The support substrate 10 of the plate-like member 11 as shown in (E) .

[0070] The manufacturing method of the base member 20 is basically the same as the manufacturing method of the conventional product, and therefore, the detailed description thereof is omitted.

[0071] After the holding substrate 10 and the base member 20 are manufactured separately, they are joined using the joining material 30. The joining of the holding substrate 10 and the base member 20 by the joining material 30 is basically the same as that in the existing product. Therefore, detailed description is omitted. In this way, the holding device 100 is manufactured.

[0072] As described above, the holding device 100 of the present embodiment includes the holding substrate 10 having excellent adhesion between the plate-shaped member 11 mainly composed of high-purity alumina (an example of ceramics) and the electrode layer 12 arranged therein and excellent heat dissipation (thermal conductivity).

[0073] The electrode layer 12 includes a predetermined amount of aluminum oxide as a minor component in addition to the conductive material 40 as a main component, so that the surface 12a of the electrode layer 12 can be made into a concave-convex surface with a fractal dimension D of 1.18 or more. Such surface 12a of the electrode layer 12 has excellent adhesion to the surface of the first dielectric layer 111 adjacent thereto, so that the formation of a gap (space) that hinders heat transfer therebetween can be suppressed.

[0074] It should be noted that, since the amount of aluminum oxide in the electrode layer 12 is suppressed to be relatively small, the electrode layer 12 does not include the ceramic (aluminum oxide) 50 of a shape connecting the first dielectric layer 111 and the second dielectric layer 112, which has poor thermal conductivity, and is mainly composed of the conductive material 40 with excellent thermal conductivity. Therefore, the electrode layer 12 of the present embodiment itself has excellent thermal conductivity. Even if the first dielectric layer 111 side stacked on the surface 12a side of such an electrode layer 12 is heated during plasma treatment, the heat can be quickly transferred from the first dielectric layer 111 to the electrode layer 12. The heat transferred to the electrode layer 12 is transferred to the second dielectric layer 112 while further transferring to the base member 20 side. The holding device 100 having such a substrate 10 is excellent in heat removal (thermal conductivity).

[0075] In addition, in the case of the present embodiment, when the plate-like component 11 is viewed from the front side (first surface S1 side), the area ratio of the surface 12a of the electrode layer 12 relative to the first surface S1 is greater than 80%, so it can be said that the electrode layer 12 has a large influence on the heat removal property (thermal conductivity) of the holding substrate 10 (holding device 100).

[0076] <Implementation Method 2>

[0077] Next, refer to Figure 6 A holding substrate 10A included in the holding device according to the second embodiment will be described. Figure 6 1 is a diagram showing a SEM image of a cross section along the thickness direction of the plate-like component 11 provided in the holding substrate 10A of the second embodiment. In the case of the present embodiment, both surfaces of the electrode layer 12A formed in the plate-like component 11 are roughened. The surface 12Aa on the first surface side (upper side) and the surface 12Ab on the second surface side (lower side) of the electrode layer 12A are concave-convex surfaces including a plurality of convex portions and a plurality of concave portions. In addition, the holding device (holding substrate 10A) of the present embodiment has basically the same structure as that of the first embodiment except that both surfaces of the electrode layer 12A are roughened.

[0078] The fractal dimension D of the surface 12Aa of the electrode layer 12A and the fractal dimension D of the surface 12Ab of the electrode layer 12A are both 1.18 or more.

[0079] In the present embodiment, similarly to the first embodiment, in the cross section of the electrode layer 12A along the thickness direction, the area ratio of the ceramic relative to the observation range set on the center side of the electrode layer 12A is 30% or less.

[0080] As a method for roughening the upper surface 12Aa of the electrode layer 12A, for example, the following method can be cited: when manufacturing the retaining substrate 10A, the above-mentioned surface roughening treatment is performed on the surface of the first stack of raw sheets for forming the first dielectric layer 111, and the metallization paste for forming the electrode layer 12A is stacked on the surface of the first stack.

[0081] In addition, as a method for roughening the surface 12Ab on the lower side of the electrode layer 12A, for example, the following method can be cited: when manufacturing the retaining substrate 10A, the above-mentioned surface roughening treatment is performed on the surface of the second stack of raw sheets for forming the second dielectric layer 112, and the second stack is stacked on the metallizing paste in a manner that the surface of the second stack is pressed against the metallizing paste on the first stack.

[0082] As in the present embodiment, not only the upper surface 12Aa of the electrode layer 12A but also the lower surface 12Ab may be roughened so as to have a concave-convex surface represented by a desired fractal dimension D (1.18 or more). By roughening both surfaces of the electrode layer 12A in this way, the heat removal property (thermal conductivity) of the holding substrate 10A (holding device) of the present embodiment becomes more excellent.

[0083] [Example]

[0084] Hereinafter, the present invention will be described in more detail based on examples. However, the present invention is not limited to these examples at all.

[0085] [Example 1]

[0086] (Preparation of Metallization Paste)

[0087] Conductive powder was added to a mixture of alumina powder, acrylic binder, and organic solvent and kneaded to prepare a metallization paste. The amount of alumina powder in the metallization paste was adjusted to 20 volume % relative to 100% of the total volume of alumina powder and conductive powder.

[0088] (Production of green sheets)

[0089] A mixture including alumina powder, acrylic binder, dispersant, plasticizer, sintering aid, etc., to which an organic solvent is further added is mixed using a ball mill to prepare a slurry for forming a green sheet of a high-purity plate-shaped component (alumina content: 99.5% by mass or more).

[0090] (Making of the substrate is maintained)

[0091] A sample of Example 1 imitating the holding substrate was produced by the same method as the method for producing the holding substrate described in Embodiment 1, except that the obtained metallizing paste and the slurry for green sheets were used. In Example 1, only the upper surface of the layered metallizing paste for forming the electrode layer was roughened by the surface roughening treatment.

[0092] [Example 2 to Example 4]

[0093] Each sample of Example 2 to Example 4 was prepared by the same method as Example 1 except that the amount of alumina powder in the metallization paste was changed to the value (volume %) shown in Table 1. In addition, in Example 2 to Example 4, a slurry for forming a high-purity plate-shaped member having an alumina content of 99.5 mass % or more was prepared.

[0094] [Comparative Example 1 to Comparative Example 3]

[0095] Each sample of Comparative Examples 1 to 3 was prepared by the same method as Comparative Example 1, except that the amount of alumina powder in the metallizing paste was changed to the value (volume %) shown in Table 1. In Comparative Examples 1 to 3, a slurry for forming a high-purity plate-shaped member having an alumina content of 99.5 mass % or more was prepared.

[0096] 〔evaluate〕

[0097] For each sample of each example and each comparative example, “area ratio of ceramic in the cross section of the electrode layer”, “fractal dimension D of the surface of the electrode layer”, “adhesion strength”, “thermal conductivity”, etc. were evaluated by the following methods.

[0098] (Area ratio of ceramic in the cross section of the electrode layer)

[0099] The sample of the simulated holding substrate was cut in the thickness direction, and the obtained cross section was polished. Then, the cross section was photographed using a scanning electron microscope (SEM) to obtain a SEM image of the cross section (magnification: 3000 times). Figure 7 1 is an explanatory diagram showing an observation range R set on the center side of the electrode layer 12 in a cross-sectional SEM image. Figure 7 As shown, the observation range R is set between a straight upper reference line L1 set so as to pass through the uppermost portion (top) of the electrode layer 12 and a straight lower reference line L2 set so as to pass through the lowermost portion (bottom). The upper reference line L1 and the lower reference line L2 are arranged parallel to each other, and when the range between such upper reference line L1 and lower reference line L2 is set as 100%, the portion excluding the portion 30% downward from the upper reference line L1 and the portion excluding the portion 30% upward from the lower reference line L2 constitutes the observation range R. That is, the observation range R is composed of the 40% portion on the central side of the range (100%) sandwiched between the upper reference line L1 and the lower reference line L2.

[0100] After that, the SEM image was binarized, and the conductive material 40 constituting the electrode layer 12 and the ceramic portion 50 existing inside the conductive material 40 were distinguished by image processing software based on the obtained binarized image. Then, the area ratio of the ceramic relative to the observation range R (100%) was further calculated by image processing software. With respect to the observation range R, three locations were set for one sample, and the above area ratio was calculated as the average value of these three locations. The results are shown in Table 1.

[0101] (Fractal dimension D)

[0102] The fractal dimension D of the surface of the electrode layer in each sample was determined by the box counting method shown below. Figure 8This is an explanatory diagram showing the image processing performed when obtaining the fractal dimension D. First, the SEM image (magnification: 3000 times) of the cross section of the electrode layer taken when evaluating the above-mentioned area ratio is binarized using image processing software (ImageJ produced by Wayne Rasband of the National Institutes of Health, USA), and then the contour of the electrode layer 12 (conductive material 40) is extracted from the obtained binary image. Figure 8 (A) shows an example of an image from which contour lines are extracted. Figure 8 The contour of the electrode layer 12 shown on the lower side of (A) corresponds to the roughened surface. Figure 8 As shown in (B), the image processing software is used to fill the inside of the contour line, and then, from the obtained filled image, as shown in Figure 8 As shown in (C), the contour line portion corresponding to the roughened surface is extracted. Figure 8 As shown in (C), coordinates are set for the extracted contour line portion.

[0103] Then, for the extracted contour line portion, the box counting tool of the image processing software was used to change the length d of one side of the square box in stages from 2 to 64 pixels (2, 3, 4, 6, 8, 12, 16, 32, 64), and the number of covers N (d) of each d was measured. The length d of one side of each square box and the corresponding number of covers N (d) were plotted according to the above formula (II) using the common logarithm, and the fractal dimension was calculated from the slope of the straight line. The results are shown in Table 1.

[0104] (Surface roughness Ra)

[0105] In addition, as a reference, according to JISB0601-2001, based on Figure 8 The surface roughness (arithmetic mean roughness) Ra of the electrode layer in each sample was determined from the profile portion (roughness curve) shown in (C). The results are shown in Table 1.

[0106] In addition, regarding the surface roughness Ra, only a reference length L is extracted from the contour portion (roughness curve) in the direction of its average line, and the X axis is set in the direction of the average line of the extracted portion and the Y axis is set in the direction of the longitudinal magnification. On this basis, when the roughness curve is expressed by y=f(X), the value obtained by the following formula (1) is expressed in micrometers (μm).

[0107]

[0108] (Adhesion strength)

[0109] Using the metallization paste for the electrode layer prepared in each embodiment, etc., Fig. 9 Test body TP shown. Fig. 9 It is an explanatory diagram schematically showing a test piece TP used for measuring the adhesion strength. Fig. 9 The test body TP shown is composed of a laminated body in which a test electrode layer 12T and a conductive electrode layer (plating attachment electrode layer) 13T are formed on a surface 11Ta of a test plate-shaped member 11T including a recess. A rod-shaped Kovar terminal 80T is fixed to the conductive electrode layer 13T of the laminated body via a nickel plating layer 60T and solder (Ag-Cu). Fig. 9 In the figure, for convenience of explanation, the surface side of the surface-roughened electrode layer 12T is represented by the reference numeral 12Ta.

[0110] The plate-shaped member 11T for the test was produced using the slurry for the green sheet prepared in Example 1 and the like.

[0111] Regarding the test bodies TP of Examples 1 to 4, the electrode layer 12T is formed on the surface 11Ta of the plate-like member 11T roughened by the surface roughening treatment, similarly to the case of producing the sample simulating the above-mentioned holding substrate.

[0112] For the test body TP produced as described above, the adhesion strength of the electrode layer 12T was measured by the method shown below using Autograph (precision universal testing machine) manufactured by Shimadzu Corporation. Specifically, in a state where the laminated body side of the test body TP is fixed in a manner in which a rod-shaped Kovar terminal 80T is arranged in the horizontal direction, a force is applied to the Kovar terminal 80T at a position 15 mm away from the root side by pressing at 5 mm / min using an Autograph. Then, the strength (N) when the test body TP is destroyed by applying such force is measured. In each embodiment and each comparative example, the strength (N) was measured 5 times. The results are as follows. Fig.10 shown. Fig.10 It is a graph showing a graph summarizing the results of measuring the adhesion strength (N) of the electrode layer of each embodiment and each comparative example. It should be noted that the average value of the five measurement results is used as the adhesion strength (N) of each embodiment, etc. In addition, the case where the adhesion strength (N) is 80N or more is defined as "adhesion", and the case where the adhesion strength (N) is less than 80N is defined as "no adhesion". In Table 1, the reference symbol "○" indicates "adhesion", and the symbol "×" indicates "no adhesion".

[0113] (Thermal conductivity)

[0114] The electrode layer for the test was prepared using the metallization paste for the electrode layer prepared in each example, etc. Then, the thermal conductivity was measured by the method shown below using the obtained electrode layer for the test. The results are shown in Table 1.

[0115] The metallization paste for the electrode layer is formed into a predetermined shape after degreasing, and the obtained molded body is fired to obtain a fired body. Then, after appropriately cutting and grinding the fired body, the thermal diffusivity and specific heat of the fired body are measured based on the laser flash method in accordance with JIS R1611. In addition, the bulk density of the fired body is measured in accordance with JIS R1634. The value obtained by multiplying the thermal diffusivity, specific heat and bulk density thus obtained is used as thermal conductivity.

[0116]

[0117] (About Adhesion)

[0118] The results of the adhesion strength test confirmed that the electrode layers of Examples 1 to 4 have excellent adhesion to the plate-like component (high-purity aluminum oxide). In the case of Examples 1 to 3, although the content of aluminum oxide in the electrode layer (5 vol% to 20 vol%) is relatively small, the surface of the electrode layer is roughened in a manner to obtain the desired fractal dimension D (above 1.18). In addition, in the case of Example 4, the surface of the electrode layer is also roughened in a manner to obtain the desired fractal dimension D (above 1.18). In the case of such Examples 1 to 4, when the adhesion strength test was performed using Autograph, no damage occurred at the interface between the electrode layer and the plate-like component of the test body TP, but damage occurred inside the electrode layer.

[0119] In contrast, it was confirmed that the electrode layers of Comparative Examples 1 to 3 did not have sufficient adhesion to the plate-like component (high-purity alumina). In the case of Comparative Examples 1 to 3, the content of alumina in the electrode layer (5 vol% to 20 vol%) was relatively small, so if the surface of the electrode layer was not roughened in a manner to achieve the desired fractal dimension D, it can be said that sufficient adhesion to the plate-like component could not be achieved. It should be noted that in the case of Comparative Examples 1 to 3, when the adhesion strength test using Autograph was performed, damage occurred at the interface between the electrode layer and the plate-like component of the test body TP.

[0120] (About thermal conductivity)

[0121] As shown in Table 1, it was confirmed that the thermal conductivity of the electrode layer decreased as the content of aluminum oxide in the electrode layer increased. In particular, in Examples 1 to 4, the thermal conductivity of the electrode layer of Examples 1 to 3 was 60 W / mK or more, and the thermal conductivity was excellent. It should be noted that, as described above, the electrode layer of Comparative Examples 1 to 3 had excellent thermal conductivity even if the adhesion to the plate-like member was insufficient.

[0122] (Relationship between fractal dimension D and adhesion)

[0123] Here, in Fig.11 In FIG. 1 , the relationship between the fractal dimension D of the electrode layer and the adhesion of the electrode layer in each example and each comparative example is summarized as a graph. Fig.11 The vertical axis of the coordinate graph represents the fractal dimension D, and the horizontal axis represents the content (vol%) of aluminum oxide contained in the electrode layer. Fig.11 In the coordinate diagram of , if the embodiments and comparative examples are distinguished into the case where the adhesion strength of the electrode layer is greater than 80N (adhesion) and the case where the adhesion strength of the electrode layer is less than 80N (no adhesion) based on the fractal dimension D, it can be set as follows Fig.11 The range Z1 (with adhesion) and the range Z2 (without adhesion) shown in FIG. 1 can be distinguished when the fractal dimension D is around 1.18. Fig.11 From the coordinate diagram of , it can be said that if the fractal dimension D of the surface of the electrode layer is 1.18 or more, the adhesion strength of the electrode layer can be ensured.

[0124] In addition, if Fig.11 As shown in the graph of , if the content (vol%) of aluminum oxide in the electrode layer increases, there is a tendency for the fractal dimension D of the surface of the electrode layer to increase. Fig.11 As shown in the coordinate graph of , when the surface of the electrode layer is roughened, the fractal dimension D becomes larger than when the surface of the electrode layer is not roughened.

[0125] (Relationship between surface roughness Ra and adhesion)

[0126] For reference, in Fig.12 In FIG. 1 , the relationship between the surface roughness Ra of the electrode layer and the adhesion of the electrode layer in each example and each comparative example is summarized in a graph. Fig.12 The vertical axis of the coordinate graph represents the surface roughness Ra, and the horizontal axis represents the content (vol%) of aluminum oxide contained in the electrode layer. In such a coordinate graph, when the adhesion strength of the electrode layer is 80N or more (adhesion exists) and when the adhesion strength of the electrode layer is less than 80N (no adhesion), even if one wants to distinguish between the examples and the comparative examples based on the surface roughness Ra, they cannot be distinguished. That is, the adhesion of the electrode layer needs to be represented by the fractal dimension D, not by the surface roughness Ra.

[0127] Description of Reference Numerals

[0128] 10 ... holding substrate, 11 ... plate-shaped member, 12 ... electrode layer, 20 ... base member, 21 ... refrigerant flow path, 30 ... bonding material, 40 ... conductive material, 50 ... ceramic, 100 ... holding device, 111 ... first dielectric layer, 112 ... second dielectric layer, R ... observation range, S1 ... first surface, S2 ... second surface, S3 ... third surface, S4 ... fourth surface

Claims

1. A holding device, comprising a holding substrate, the holding substrate comprising: a plate-shaped member including a first surface for holding an object and a second surface arranged on the opposite side of the first surface, and containing ceramic as a main component; and an electrode layer, which is arranged inside the plate-like member and contains a conductive material as a main component and the ceramic as a secondary component, wherein The plate-shaped member includes: a first dielectric layer, which is arranged on the first surface side of the electrode layer, and the content of the main component is 99 mass % or more; and a second dielectric layer disposed on the second surface side of the electrode layer, wherein the content of the main component is 99 mass % or more, The fractal dimension D of the surface on the first surface side of the electrode layer is 1.18 or more.

2. The holding device according to claim 1, wherein: In a cross section of the electrode layer along a thickness direction, an area ratio of the ceramic relative to an observation range set on a center side of the electrode layer is 30% or less.

3. The holding device according to claim 1 or 2, wherein: In the cross section of the electrode layer, the ceramic does not exist in a form connecting the first dielectric layer and the second dielectric layer.

4. The holding device according to claim 1 or 2, wherein: In a state where the plate-shaped member is viewed from above, an area ratio of the electrode layer to the first surface is 80% or more.

Images