Holding device and method for manufacturing a holding device

Abstract

In a holding device for holding an object, the flatness of the surface on which the object is placed is improved. We provide technology to enable this. [Solution] The holding device has a first surface on which the object is placed and a second surface which is the back surface of the first surface. A plate-shaped plate-like portion having a thermal expansion coefficient different from that of the material mainly forming the plate-like portion. It consists of a plate-shaped base portion, which is arranged on the second surface side of the plate-shaped portion, and the plate-shaped portion and the base portion. It comprises a joint portion that is positioned between the plate-shaped portion and the base portion and joins the plate-shaped portion, and the first The surface flatness is 40 μm or less, and the bonding occurs after a thermal cycling test at -70°C to 80°C. There is no peeling on the outer perimeter of the part.

Description

[Technical Field]

[0001] This disclosure relates to a holding device and a method for manufacturing a holding device. [Background technology]

[0002] Conventionally, as a holding device for holding objects, for example, when manufacturing semiconductors, a holding device for wafers etc. Electrostatic chucks are known for holding objects. Generally, electrostatic chucks are used when the object is placed on them. It comprises a plate-like portion, a base portion, and a joint portion that connects the plate-like portion and the base portion. For example, using ceramics as the material for the plate-like part and metal as the material for the base part, The thermal expansion coefficients of the part and the base part are often different. Therefore, during the manufacturing of the electrostatic chuck, Due to temperature changes during use, there is a risk that the surface of the plate-like part (the surface on which the object is placed) may warp. There is.

[0003] In response to this problem, as a method to reduce the warping of substrates used in semiconductor processes, , while expanding the substrate, the substrate having an epoxy layer is subjected to at least the glass transition temperature of the epoxy layer Heating to a certain temperature and applying clamping forces from above and below towards the substrate The process involves restraining the board, applying at least one electrostatic field to the substrate, and rapidly cooling the substrate. A method has been proposed that includes doing (see, for example, Patent Document 1). Also, the plate-like part and Regarding the joint that connects to the base part, by maintaining flexibility even at low temperatures, the plate-like part Techniques have been proposed to suppress this warping (see, for example, Patent Document 2). [Prior art documents] [Patent Documents]

[0004] [Patent Document 1] U.S. Patent Application Publication No. 2021 / 0035795 [Patent Document 2] Japanese Patent Application Laid-Open No. 2020-23088 [Summary of the Invention] [Problems to be Solved by the Invention]

[0005] However, even with the above technology, the flatness of the mounting surface of the plate-like portion may not be sufficient. . Therefore, other technologies for improving the flatness of the mounting surface of the plate-like portion are desired. Note that these problems are not limited to electrostatic chucks, but are common problems in various holding devices such as semiconductor manufacturing devices such as plasma etching devices.

[0006] The present disclosure has been made to solve at least some of the above-described problems, and aims to provide a technology for improving the flatness of the surface on which an object is placed in a holding device that holds an object. [Means for Solving the Problems] The present disclosure can be realized in the following forms. [[ID= {31}]]According to one aspect of the present disclosure, a holding device for holding an object is provided. This holding device

[0007] has a plate-like plate portion having a first surface on which the object is placed and a second surface on the back side of the first surface, a plate-like base portion made of a material having a coefficient of thermal expansion different from that of the material mainly forming the plate portion, and disposed on the second surface side of the plate portion, and a joining portion disposed between the plate portion and the base portion for joining the plate portion and the base portion. The flatness of the first surface of the plate portion is 40 μm or less, and there is no peeling at the outer peripheral portion of the joining portion after a thermal cycle test at -70°C to 80°C. ​​​

[0008] According to this type of holding device, the flatness of the first surface of the plate-like part is 40 μm or less, The temperature distribution of the object held by the holding device can be made uniform, the object can be flattened, and This reduces in-plane variation in the pitch. It also alleviates residual stress in the plate-like portion. can.

[0009] The holding device is used in an environment where the temperature changes from a low temperature of about -70°C to a high temperature of about 80°C. When used, the thermal expansion coefficients of the material mainly forming the plate-like portion and the material forming the base portion are different. Therefore, thermal stress is generated in the holding device. Due to the thermal stress, the interface between the joint and the plate-like part, and the joint and Although there is a risk of peeling occurring at the interface with the base, with this type of holding device, -7 After thermal cycling tests from 0°C to 80°C, there is no peeling on the outer periphery of the joint. Therefore, this shape The device for maintaining the state is subjected to temperature changes from a low temperature of approximately -70°C to a high temperature of approximately 80°C. We provide a retaining device that can be used without the plate-like part and the base part separating at the outer circumference. It can be provided.

[0010] (2) In the above-described holding device, the joint portion is the storage modulus in dynamic viscoelasticity measurement. E' may be 10 MPa or less at 25°C. The lower the storage modulus, the softer the joint. Therefore, with this configuration, the joint is sufficiently flexible and can withstand temperature changes in the holding device. Consequently, the tensile stress generated on the outer circumference of the retaining device can be reduced. As a result, the outer circumference of the joint This can suppress peeling of the surrounding area.

[0011] (3) In the above-described holding device, the joint portion has a loss coefficient t in dynamic viscoelasticity measurement. There may be a temperature range in which anδ is 0.2 or higher. In a viscoelastic material, the loss coefficient tan The closer δ is to 0, the closer it is to an elastic body; the larger δ is, the more it exhibits the properties of a viscous body. At the joint, If there is a temperature range in which the loss coefficient tanδ in the viscoelasticity measurement is 0.2 or higher, then that temperature... In the viscous range, viscous properties will appear. When viscous properties are present, gradually reduce the stress. This allows for the suppression of warping of the plate-like portion.

[0012] (4) In the above-described holding device, the joint is at -70°C in dynamic viscoelasticity measurement. The loss coefficient tanδ may be 0.2 or greater. With such a configuration, extremely low temperatures (e.g.) For example, at -70°C, the joint exhibits viscosity, so when the holding device is used at extremely low temperatures... This also helps to suppress warping of the plate-like portion.

[0013] (5) In the holding device of the above form, when the thickness is 0.2 mm, -70°C and A joining material is used in which the shear strain at 80°C is 250% or more, and the thickness of the joining part is 0.2 It may be more than mm and less than 1.0 mm. With this configuration, the thickness at the joint is When the thickness is 0.2 mm, use a bonding material that exhibits a shear strain of 250% or more at -70°C and 80°C. Therefore, under the harsh temperature conditions of extremely low temperatures of -70°C and high temperatures of 80°C, the strain direction It can deform sufficiently under force and can withstand temperatures in a range of at least -70°C to 80°C. This effectively suppresses warping of the plate-like portion.

[0014] (6) In the holding device of the above form, when the thickness is 0.7 mm, -70°C and A joining material is used in which the shear strain at 80°C is 70% or more, and the thickness of the joint is 0.2 m It may be m or more and 1.0 mm or less. With this configuration, the thickness of the joint is 0 When the thickness is 0.7 mm, use a bonding material that exhibits a shear strain of 250% or more at -70°C and 80°C. Therefore, under the harsh temperature conditions of extremely low temperatures of -70°C and high temperatures of 80°C, the force in the strain direction It can be sufficiently deformed in a temperature range of at least -70°C to 80°C. The warping of the plate-like portion can be sufficiently suppressed.

[0015] (7) According to other embodiments of the present disclosure, a method for manufacturing a holding device for holding an object is provided. According to the manufacturing method of this holding device, the first surface is the surface on which the object is placed and the A plate-like portion having a second surface which is the back surface of the first surface, and the heat of the material that mainly forms the plate-like portion Prepare a plate-shaped base made of a material having a different coefficient of thermal expansion than the coefficient of thermal expansion, and an adhesive. The preparation step involves placing the base portion on the second surface side of the plate-like portion via the adhesive. The adhesive is heated to a temperature of 60°C or higher to cure it, and the bonded part is formed from the cured adhesive. A jointing step in which the plate-like portion and the base portion are joined together to form a jointed body, and At a temperature in which the loss coefficient tanδ in the dynamic viscoelasticity measurement of the cured product of the adhesive is 0.2 or higher The method comprises an annealing step of annealing the joined body for more than one hour.

[0016] According to the manufacturing method of this type of holding device, the loss in the dynamic viscoelasticity measurement of the cured adhesive is Annealing process: Annealing the joint at a temperature where the loss coefficient tanδ is 0.2 or higher for 1 hour or more. It has a certain degree. In viscoelastic materials, the loss coefficient tanδ in dynamic viscoelasticity measurement is 0.2 or greater. In the temperature range where this occurs, viscous properties emerge. When viscous properties are present, gradually reduce the stress. This is possible. Therefore, by performing the annealing process, the adhesive acquires viscous properties. Stress relaxation and creep occur, the adhesive deforms, and the plate-like part and the base part return to their original shape. By returning to a state without warping, the flatness of the first surface of the plate-like part can be improved. can.

[0017] (8) A method for manufacturing the holding device of the above form, wherein the plate prepared in the preparation step The plate-like portion has a concave first surface, and at the end of the joint formation process, the plate-like portion has a concave first surface The first surface is concave, and the plate-like portion at the end of the annealing process has a concave surface, and the depth of the concave The depth may be shallower than the concave surface of the first surface of the plate-like portion in the preparation step. This improves the flatness of the first surface of the plate-like portion.

[0018] (9) A method for manufacturing the holding device of the above form, wherein the plate prepared in the preparation step The plate-like portion has a convex first surface, and at the end of the joint formation process, the plate-like portion has a convex first surface The surface is convex, the height of the convexity is greater than that of the preparation step, and the plate-like portion at the end of the annealing step The first surface is a convex surface, and the height of the convex surface is the height of the plate-like portion at the end of the joint formation process. It may be lower than the convex surface of the first surface. Even in this way, the flatness of the first surface of the plate-like part can be improved. It is possible.

[0019] This disclosure can be implemented in various forms other than those described above, for example, a semiconductor-made device including a holding device. This can be realized in the form of a manufacturing device, a method for forming joints, and so on. [Brief explanation of the drawing]

[0020] [Figure 1] This is an explanatory diagram illustrating the schematic configuration of the electrostatic chuck in the embodiment. [Figure 2] This is an explanatory diagram illustrating the cross-sectional structure of an electrostatic chuck. [Figure 3] This is an explanatory diagram of the lower limit of shear strain at -70°C. [Figure 4] This is an explanatory diagram of the lower limit of shear strain at 80°C. [Figure 5] This is an explanatory diagram conceptually illustrating the shape changes of each part in each step of the manufacturing process for an electrostatic chuck. [Figure 6] This is an explanatory diagram that summarizes the evaluation results. [Figure 7] This is an explanatory diagram showing the storage modulus. [Figure 8] This is an explanatory diagram showing the loss tangent. [Figure 9] Figure 8 is an explanatory diagram showing a magnified view of the loss tangent. [Figure 10] This is a schematic diagram illustrating the method for calculating shear strain. [Figure 11] This is an explanatory diagram summarizing the evaluation results of the manufacturing method for electrostatic chucks. [Modes for carrying out the invention]

[0021] A. Embodiments: (A-1) Overall configuration of the electrostatic chuck: Figure 1 is an explanatory diagram schematically showing the configuration of the electrostatic chuck 10 in the embodiment. Figure 2 This is an explanatory diagram schematically showing the cross-sectional configuration of the electrostatic chuck 10. In Figure 1, the electrostatic chuck Part of the 10 is shown broken off. Also, in the figure, to determine the direction, mutually orthogonal lines are shown. The XYZ axes are shown. For convenience, in this specification, the positive Z-axis direction is referred to as the upward direction, and the Z-axis Although we define the negative direction as downward, the electrostatic chuck 10 is actually oriented differently. They may be installed in any orientation. The X, Y, and Z axes shown in each figure are in the same orientation. The diagrams above schematically represent the arrangement of each part, and the proportions of the dimensions of each part are not accurately represented. It does not represent anything.

[0022] The electrostatic chuck 10 is a device that attracts and holds an object by electrostatic attraction, for example, a semi-electrostatic chuck. Used to fix the wafer W (Figure 2), which is the object to be manufactured, inside the vacuum chamber of a conductor manufacturing apparatus. The electrostatic chuck 10 comprises a plate-shaped portion 20, a base portion 30, and a joint portion 40. These are the plate-shaped part 20, the joint part 40, and the base part 3, oriented in the -Z axis direction (vertically downward). They are stacked in order of 0. The electrostatic chuck 10 in this embodiment is also called a "holding device". Bu.

[0023] The plate-like portion 20 has a first surface 24 on which the object is placed, and a second surface which is the back surface of the first surface 24. A substantially circular plate-shaped member having 26, and made of ceramic (for example, aluminum oxide or nitride). Formed primarily from aluminum, etc. In this specification, a specific component is "main "A component" or "a material that primarily forms the structure" means that the specific component in question constitutes the majority of the composition. This means that it is a material with a large amount of material. The diameter of the plate-like part 20 is, for example, 50 mm to 500 mm. It should be approximately 200mm to 350mm. The thickness of the plate-like part 20 is, for example, For example, it should be about 1 mm to 10 mm. In other embodiments, the plate-like portion 20 is, for example, It may be formed primarily from materials other than ceramics, such as resins like liimide.

[0024] The flatness of the first surface 24 of the plate-like portion 20 is 40 μm or less. Here, flatness is measured using a contact method. The measurements were taken using a standard method and calculated using the following method: From multiple measurement points, the least squares plane (provisional) was used. The plane is determined, and the sum of the absolute values ​​of the maximum positive and maximum negative distances to that plane is the degree of flatness. It was specified. As will be described in detail later, the flexibility of the joint 40 and the manufacturing of the electrostatic chuck 10 By improving the manufacturing method, the flatness of the first surface 24 of the plate-like portion 20 can be improved, that is, This allows us to reduce the value of flatness.

[0025] As shown in Figure 2, an adsorption electrode 22 is arranged inside the plate-shaped portion 20. 22 is formed from a conductive material such as tungsten or molybdenum. When a voltage is applied to the electrode 22 from a power source (not shown), an electrostatic attraction is generated, and this static The wafer W is adsorbed and fixed to the first surface 24 of the plate-shaped portion 20 by electroattraction. The adsorption electrode 22 is It may be a bipolar type or a unipolar type. Also, a conductive material is placed inside the plate-shaped part 20. It is composed of a resistance heating element made of a material (for example, tungsten or molybdenum), A heater electrode (not shown) is provided for heating the wafer W adsorbed and fixed to the first surface 24. That's good too.

[0026] The base portion 30 is positioned on the second surface 26 side of the plate-shaped portion 20, supporting the plate-shaped portion 20 and the cooling unit. It is a plate-shaped member that has the ability to expand and contract in a roughly circular shape. The thermal expansion relationship of the material constituting the base portion 30 The number differs from the thermal expansion coefficient of the material mainly forming the plate-like portion 20. The base portion 30 is, for example, A small amount of aluminum, magnesium, molybdenum, titanium, tungsten, and nickel It may contain at least one type of metal, such as molybdenum, titanium, or tungsten. Because it has a relatively small coefficient of thermal expansion among the metals mentioned above, at least one of these When the base portion 30 is constructed using a different type of metal, the space between the base portion 30 and the plate-shaped portion 20 It is desirable to be able to suppress the difference in thermal expansion coefficients. " refers to the "coefficient of linear expansion." Also, because magnesium has a relatively small Young's modulus, When using magnesium to construct the base portion 30, the thermal stress generated in the base portion 30 is reduced. It is desirable that it can be reduced. Also, aluminum has a relatively high thermal conductivity and is easy to process. It is easy and low-cost. Therefore, when using aluminum to construct the base part 30 The base portion 30 can improve the cooling efficiency of the plate-shaped portion 20 and the wafer W, and electrostatic discharge It is desirable to reduce the manufacturing cost of the jack 10. The cooling efficiency of the base part 30 From the perspective of increasing performance while keeping manufacturing costs down, a higher metal content in the base part 30 is preferable. It is desirable that the base portion 30 be mainly composed of metal. For example, a highly versatile material It must contain 90% or more by mass of aluminum (for example, A6061, A5052, etc.) It is preferable that it be made of an aluminum alloy. However, the base part 30 may be made of ceramic. It may contain non-metallic components such as chlorine. The diameter of the base portion 30 is, for example, 220 mm. A length of approximately m to 550 mm is sufficient, and typically it is 220 mm to 350 mm. Base section 30 The thickness should be, for example, around 20mm to 40mm.

[0027] Multiple refrigerant flow paths 32 are formed inside the base portion 30 so as to follow the XY plane. By flowing a refrigerant such as a fluorine-based inert liquid, water, or liquid nitrogen through the refrigerant flow path 32, The base portion 30 is cooled. Then, the base portion 30 and the plate-shaped portion 20 are connected via the joint portion 40. The heat transfer between them cools the plate-shaped portion 20, and the wafer W held on the first surface 24 of the plate-shaped portion 20 is cooled. The wafer W is cooled. This enables temperature control of the wafer W. In addition to the configuration having a medium flow path 32, the base portion 30 is cooled from the outside. Alternatively, the base portion 30 may be provided with a cooling function.

[0028] The joint portion 40 is positioned between the plate-shaped portion 20 and the base portion 30, and the plate-shaped portion 20 and the base portion 30 is joined together. The joint 40 is made of an adhesive containing a resin material that has adhesive properties. The adhesive is also called the "adhesive composition." The adhesive further relates to the properties of the joint 40 and the joint Various fillers (inorganic fillers) are used to adjust the properties of the paste for forming 40. It may include. That is, the joint 40 is made of an adhesive resin material and an inorganic filler. It can be composed of composite materials including the following. However, the joint 40 does not contain inorganic filler. This is also acceptable. The thickness of the joint 40 may be, for example, 1. It is acceptable to keep it below 0.0 mm, preferably below 0.60 mm, and even more preferably below 0.50 mm. Furthermore, a thickness of 0.35 mm or less is even more desirable. The thickness of the joint 40 is, for example, the thickness of the joint 40 From the standpoint of ensuring flexibility and strength, a thickness of 0.05 mm or more is sufficient. (Thickness of joint 40) From the viewpoint of thermal resistance, flexibility, and strength, a thickness of 0.2 mm to 1.0 mm is preferred. The joint 40 will be described in detail later.

[0029] The electrostatic chuck 10 is further provided with multiple gas supply passages 50. 50 is provided penetrating the plate-shaped portion 20, the joint portion 40, and the base portion 30 in the Z direction. The gas outlet 52 formed on the first surface 24 is open (see Figure 1). The supply line 50 is supplied with an inert gas, such as helium gas, from a gas supply device (not shown). Then, an inert gas is supplied from the gas outlet 52 to the space between the first surface 24 and the wafer W. This supplies heat. This increases the heat transfer between the plate-shaped part 20 and the wafer W, thereby reducing the temperature of the wafer W. The controllability of the fabric is further improved. Note that the gas supply path 50 is not essential, and the electrostatic chuck 1 It is also possible to omit the gas supply line 50 at point 0.

[0030] (A-2) Composition of the joint: The electrostatic chuck 10 of this embodiment undergoes a thermal cycling test at -70°C to 80°C, and the joint 4 There is no peeling on the outer circumference 40cp of 0. Here, the outer circumference 40cp of the joint 40 is the joint This is the range from the periphery 40c of 40 toward the center at a predetermined distance L (Figure 1). The outer periphery 40cp of the joint 40 is shown with diagonal hatching. Also, the peripheral edge 40c A line at a distance L is shown as a dashed line on the first surface 24 of the plate-like portion 20. For example, the distance L is 2 mm.

[0031] The delamination can be confirmed by ultrasonic testing. The longest diameter (maximum length) of the delamination is 2m. A peeling is defined as a peeling of m or more. The conditions for the thermal cycling test are as follows: This process involves holding the temperature at -70°C for 3 hours, then raising it to 80°C for 3 hours, and repeating this cycle 20 times.

[0032] As described above, the plate-shaped part 20 and the base part 30 have different coefficients of thermal expansion, therefore the plate-shaped part 20 and the base part 30 Due to the difference in thermal expansion of the - section 30, delamination is most likely to occur at the outer circumference of the joint section 40. Stress concentration also occurs around the through-holes that constitute the gas supply passage 50 in the electrostatic chuck 10. This makes the joint 40 prone to peeling. Due to the difference, there is a risk that cracks may occur inside the joint 40. In contrast, this implementation In the electrostatic chuck 10, bonding occurs at least after a thermal cycling test at -70°C to 80°C. Since there is no peeling on the outer circumference 40cp of part 40, it can be repeatedly used in a temperature range of -70℃ to 80℃. It can be said that it can withstand use. For example, by etching an object such as a wafer with plasma. When using a low temperature, the aspect ratio (depth / width) can be increased. In other words, deep etching is possible. Therefore, the electrostatic chuck 10 of this embodiment By using this method, it becomes possible to deeply etch the target object.

[0033] The joint 40 has a storage modulus E' of 10 MPa or less at 25°C in dynamic viscoelasticity measurement. It is preferable that the storage modulus is lower, the softer the joint is. Therefore, the movement of the joint 40 If the storage modulus E' in the viscoelasticity measurement is 10 MPa or less at 25°C, then the joint 40 It is sufficiently soft, and the tension generated on the outer circumference of the electrostatic chuck 10 due to temperature changes in the electrostatic chuck 10 This reduces tensile stress. As a result, it suppresses peeling of the outer circumference of the joint 40. It is possible.

[0034] The joint 40 is in a temperature range where the loss coefficient tanδ in dynamic viscoelasticity measurement is 0.2 or higher. It is preferable that it exists. In a viscoelastic material, the closer the loss coefficient tanδ is to 0, the more elastic the material becomes. The closer to 1, the closer it is to a viscous material. At the joint 40, the loss coefficient in dynamic viscoelasticity measurement. If there exists a temperature range where tanδ is 0.2 or greater, then viscous properties will be exhibited in that temperature range. When viscous properties are present, the stress can be gradually reduced, so the plate-like portion 20 It can suppress warping.

[0035] The joint 40 has a loss coefficient tanδ of 0.2 or more at -70°C in dynamic viscoelasticity measurement. It is preferable that the loss factor tanδ at -70°C is 0.2 or higher. For example, at -70°C, the joint 40 exhibits viscosity, so the electrostatic chuck 10 is extremely cold. When in use, warping of the plate-like portion 20 can be suppressed.

[0036] At joint 40, when the thickness is 0.2 mm, the shear strain at -70°C and 80°C is 250%. Using a joining material as described above, the thickness of the joint 40 is 0.2 mm or more and 1.0 mm or less. This is preferable. When the thickness of the joint 40 is 0.2 mm, the shear strain at -70°C and 80°C is Using a bonding material with a bonding strength of 250% or more allows for extreme temperatures of -70°C and 80°C. Under certain temperature conditions, it can deform sufficiently against strain forces, and at least - Warping of the plate-like portion can be sufficiently suppressed in a temperature range of 70°C to 80°C.

[0037] When the thickness of the joint 40 is 0.7 mm, the shear strain at -70°C and 80°C is less than 70%. The upper joining material is used, and the thickness of the joint is preferably 0.2 mm or more and 1.0 mm or less. It seems that when the thickness of the joint 40 is 0.7 mm, the shear strain at -70°C and 80°C is 7. Using a bonding material with a viscosity of 0% or more will result in severe temperature conditions such as extremely low temperatures of -70°C and high temperatures of 80°C. In this case, it can deform sufficiently against strain forces, and at least -70°C Warping of the plate-like portion can be sufficiently suppressed within a temperature range of 80°C.

[0038] Figure 3 is an explanatory diagram of the lower limit of shear strain at -70°C. Figure 4 shows the lower limit of shear strain at 80°C. This is an explanatory diagram of the lower limit of cross-sectional strain. The material (joining material) used in the joint 40 is shown in Figure 3. It is preferable to use a material whose lower limit of shear strain falls within the area of ​​the hatched lines shown in Figure 4. Figure 4 was created using the following method. The thermal expansion coefficient of the plate-shaped part 20 is 6.80 × 10 -6 (1 / ℃) The thermal expansion coefficient of the base part 30 is 2.30 × 10 -5 (1 / ℃) The curing temperature is 100°C. At the curing temperature, the diameter of the plate-shaped part 20 and the base part 30 is 350 mm, and the joint part 40 is bonded. Assuming a diameter of 350 mm and zero stress, The radii of the plate-like portion 20 and the base portion 30 were calculated when the temperature reached 80°C. The difference in radius between the plate-shaped part 20 and the base part 30 is such that the thickness of the joint part 40 is 0.2 mm and 0.7 mm. For each case in mm, the required shear strain for each thickness is calculated by dividing by the thickness. I took it out. Similarly, for Figure 3, the calculations were performed assuming a temperature of -70°C.

[0039] Shear strain is an indicator value that represents the flexibility and stress relaxation performance of the joint 40, and the joint When a shear force is applied to part 40, the shear stress generated at the joint 40 becomes maximum. In other words, when the maximum shear stress occurs at the joint 40, the strain generated at the joint 40 is This is expressed as a percentage (%) of the thickness of the joint 40. The larger the shear strain (%), the greater the shear strain (%). This indicates that the joint 40 has high flexibility. A tensile testing machine is used to measure shear strain. The specific measurement method used will be explained in detail later. The shear strain of the joint 40 is By using the above values, it is possible to ensure sufficient flexibility and stress relaxation performance of the joint 40. Therefore, damage to the joint 40 when shear force is applied to the joint 40 is suppressed. It is possible.

[0040] The shear strain of the joint 40 is modified by the adhesive (resin) contained in the joint 40. This is possible. For example, the amount of shear strain can be changed depending on the type of resin that makes up the joint 40. In addition, even when using the same type of resin, the distance between crosslinking points in the polymer resin is By controlling the separation, the shear strain of the joint 40 can be changed. Specifically, the above resin Reducing the amount of functional groups that act as crosslinking points (functional group equivalent) increases the distance between crosslinking points. As the flexibility of the resin improves, the amount of shear strain can be increased. Even if the content of reactive functional groups in the resin is the same, the curing temperature and curing time of the resin may differ. The distance between crosslinking points can be controlled by adjusting curing conditions such as the curing temperature. If the curing time is increased, the curing of the resin will progress further, and crosslinking will occur. As density increases, the distance between bridge points decreases.

[0041] Furthermore, the higher the proportion of inorganic filler in the joint 40, the greater the shear strength of the joint 40. The amount of filler tends to decrease. This is because the higher the proportion of inorganic filler, the less inorganic filler is present. The degree to which the agent restrains the surrounding resin (resin composition) increases, and the flexibility of the joint 40 decreases. This is thought to be because the joint 40 becomes less susceptible to distortion.

[0042] Examples of adhesives that make up the joint 40 include silicone resin, acrylic resin, and Epoxy resins and the like can be used. Silicone resins, in particular, have relatively good heat resistance and It is desirable because of its excellent flexibility. Among these resins, silicone resin has a relatively high elastic modulus. Because it is low, it has a high function of mitigating thermal stress generated at the joint 40, and the heat resistance temperature is relatively low. Because it is expensive, it is desirable.

[0043] Among silicone resins, the following silicone resins are preferable for maintaining flexibility at low temperatures (e.g., about -70°C). In the following description, the silicone resin is also referred to as "polyorganosiloxane". This polyorganosiloxane has R 3SiO units (hereinafter also referred to as "M units") at both ends, m R 1 2SiO 1 / 2 units (hereinafter also referred to as "the first D units") and n R R 2 SiO 2 / 2 units (hereinafter also referred to as "the second D units") (where m is an integer of 1 or more and n is an integer of 1 or more). Such polyorganosiloxane is represented by the following general formula (1). (R 3 3SiO 4 )2(R 2 / 2 2SiO 1 ) 1 / 2 (R 2 R 2 / 2 SiO m ) 3 ··· General formula (1) (R 4 and R 2 / 2 are each independently an unsubstituted or substituted aliphatic hydrocarbon group having 1 to 12 carbon atoms, R n is an unsubstituted or substituted aliphatic hydrocarbon group having 1 to 12 carbon atoms, or an unsubstituted or substituted aromatic hydrocarbon group having 6 to 10 carbon atoms, R 1 is an unsubstituted or substituted aromatic hydrocarbon group having 6 to 10 carbon atoms, and the polyorganosiloxane contains at least 2 alkenyl groups directly bonded to silicon (Si) atoms in one molecule). (2) Also, from the viewpoint of enhancing the flexibility of the adhesive composition, in the polyorganosiloxane structure, R SiO 2 units (T units) and SiO​​​​​​​​​​​​​​​​​​​​4 / 2 It is preferable that the unit (Q unit) is not included.

[0044] In the above general formula (1), R 1 and R 2 More preferably, an alke having 2 to 8 carbon atoms. One of the following is an unsubstituted or substituted compound that does not contain a nyl group or an aliphatic unsaturated bond with 1 to 12 carbon atoms. It is a valence hydrocarbon group. Examples of alkenyl groups with 2 to 8 carbon atoms include vinyl groups and allyl groups. Group, propenyl group, isopropenyl group, butenyl group, pentenyl group, hexenyl group, cyclo Examples include the lohexenyl group. Unsubstituted groups with 1 to 12 carbon atoms that do not contain aliphatic unsaturated bonds are also available. Examples of substituted monovalent hydrocarbon groups include methyl, ethyl, propyl, and iso groups. Propyl group, butyl group, isobutyl group, tert-butyl group, pentyl group, neopentyl Alky groups such as hexyl group, heptyl group, octyl group, nonyl group, decyl group, dodecyl group, etc. Cycloalkyl groups such as cyclopentyl, cyclohexyl, and cycloheptyl groups; Aryl groups such as phenyl, tolyl, xylyl, naphthyl, and biphenylyl groups; ben Aralkyl groups such as zyl groups, phenylethyl groups, phenylpropyl groups, and methylbenzyl groups; Furthermore, at least one of the hydrogen atoms bonded to the carbon atoms of these groups is fluorine, chlorine, or bromine. Examples include halogen atoms, cyano groups, and other substituted groups.

[0045] In the M unit of the general formula (1) above, R is included in the M unit. 1 At least one of them is More preferably, an alkenyl group having 2 to 8 carbon atoms, and even more preferably, a vinyl group, Low carbon-number groups of 2 to 4, including lyl group, propenyl group, isopropenyl group and butenyl group It is a subalkenyl group, and is particularly preferably a vinyl group. The M unit of the above general formula (1) In this context, R is included in the M unit. 1 The remaining two of them are, more preferably, carbon independently of each other. It is an unsubstituted or substituted monovalent hydrocarbon group that does not contain aliphatic unsaturated bonds number 1 to 12. More preferably, a carbon source containing a methyl group, an ethyl group, a propyl group, and an isopropyl group. It is a lower alkyl group having 1 to 3 members, and is particularly preferably a methyl group. That is, the above In the M unit of general formula (1), R is included in the M unit. 1 One of them is a vinyl group, R 1 The remaining two of these are preferably methyl groups.

[0046] In the above general formula (1), R 3 More preferably, an aliphatic unsaturated carbon atom having 1 to 12 carbon atoms. Unsubstituted or substituted monovalent hydrocarbon groups that do not contain bonds, or unsubstituted groups with 6 to 10 carbon atoms. Alternatively, it is a substituted monovalent aromatic hydrocarbon group. It contains an aliphatic unsaturated bond with 1 to 12 carbon atoms. As for unsubstituted or substituted monovalent hydrocarbon groups, R in the general formula (1) above. 1 Oh biR 2 The same examples given can be used as illustrations. Unsubstituted carbon atoms with 6 to 10 carbon atoms. Alternatively, examples of substituted monovalent aromatic hydrocarbon groups include phenyl, tolyl, and xyl groups. Aryl groups such as aryl groups and naphthyl groups; benzyl groups, 2-phenylethyl groups, 2-phenylpropyl groups. Aralkyl groups such as ropyl groups; and at least one hydrogen atom bonded to the carbon atom of these groups Another example is groups substituted with halogen atoms such as fluorine, chlorine, and bromine, or cyano groups. In the above general formula (1), R 3 More preferably, a phenyl group, a tolyl group, The aryl group is a xylyl group or a naphthyl group, and is particularly preferably a phenyl group.

[0047] In the above general formula (1), R 4 More preferably, an unsubstituted or carbon atom having 6 to 10 carbon atoms. This is a substituted monovalent aromatic hydrocarbon group. It is an unsubstituted or substituted monovalent aromatic group with 6 to 10 carbon atoms. As for the group hydrocarbon group, R in the general formula (1) above is 3 The same as the example given An example can be given. In the above general formula (1), R 4 More preferably, pheny The aryl group includes a tolyl group, a xylyl group, and a naphthyl group, and is particularly preferably a f It is an enyl group. In the above general formula (1), R 3 and R 4 This means that the substituents are identical. Preferred. As the polyorganosiloxane represented by the above general formula (1), particularly preferred It is a vinyl-terminated diphenylsiloxane-dimethylsiloxane copolymer.

[0048] The content of the above aromatic hydrocarbon group in the above polyorganosiloxane is 3 mol% or less. The above is 16 mol% or less. The above aromatic hydrocarbon in the above polyorganosiloxane. If the group content is less than 3 mol%, the introduction of the above aromatic hydrocarbon group results in a galvanic effect. The effect of lowering the transition temperature tends not to be fully realized. In contrast, the above polyorgano When the content of the above aromatic hydrocarbon group in the siloxane exceeds 16 mol%, the above aroma The hydrocarbon groups interact with each other via π-π stacking, leading to crystallization. The lath transition temperature tends to increase. The above aromatic carbon in the above polyorganosiloxane The hydrogenated group content is more preferably 4 mol% or more and 14 mol% or less, and further Preferably, the concentration is 4 mol% or more and 10 mol% or less.

[0049] The polyorganosiloxane having the above structure contained in the above adhesive composition is a single type Preferably, it consists of a single type of polyorganosiloxane. "Consists of the above general" means that all polyorganosiloxanes constituting the adhesive composition are the same as above In equation (1), R 1 ,R 2 ,R 3 ,R 4 A uniquely determined substituent for each substituent This means that it has a substructure, R 1 ,R 2 ,R 3 ,R 4 All or any substituents This means that it is not a mixture of polyorganosiloxanes with different molecular structures. This eliminates the addition of silicone compounds such as silane coupling agents, curing catalysts, and crosslinking agents. There isn't one.

[0050] Silane coupling agents can be added for the purpose of imparting adhesion. Any conventionally known silane coupling agent can be used as the pulling agent. Not particularly limited. Examples of silane coupling agents include vinyltrimethoxysilane, Vinyltriethoxysilane, vinyltrismethoxyethoxysilane, 2-(3,4-epo Xycyclohexyl)ethyltrimethoxysilane, 3-glycidoxypropylmethylmethyl Toxysilane, 3-glycidoxypropyltrimethoxysilane, 3-glycidoxypropyl Methyldiethoxysilane, 3-glycidoxypropyltriethoxysilane, p-Styli Lutrimethoxysilane, 3-methacryloxypropylmethyldimethoxysilane, 3-meth Cryloxypropyltrimethoxysilane, 3-methacryloxypropyltriethoxysilane 3-Acryloxypropyltrimethoxysilane, 3-Aminopropyltrimethoxysilane Lan, 3-aminopropyltriethoxysilane, 3-mercaptopropyltrimethoxysilane Lan, or 3-isocyanatetopropyltriethoxysilane, etc., can be used. Furthermore, the above silane coupling agents can be used individually or in combination of two or more types. This can be done. The amount of silane coupling agent to be added is per 100 parts by weight of the above silicone resin. The amount is 0.1 parts by weight or more and 20 parts by weight or less. If the amount added is less than 0.1 parts by weight As a result, the above adhesive composition tends not to be given sufficient adhesion. In contrast, the amount of the additive When the amount exceeds 20 parts by weight, it tends to inhibit the curing of polyorganosiloxane. The amount added is more preferably 0.5 parts by weight or more and 15 parts by weight or less, and even more preferably The amount is 1 part by weight or more and 10 parts by weight or less. Alternatively, instead of the above silane coupling agent, Titanate-based coupling agents or aluminate-based coupling agents may also be used.

[0051] A curing catalyst can be added to accelerate the curing reaction. Examples of curing catalysts include: Any conventionally known curing catalyst can be used, and is not particularly limited. Examples include organotin, inorganic tin, titanium catalysts, bismuth catalysts, metal complexes, platinum catalysts, and basics. Substances and organophosphorus oxides can be used. The curing catalyst is more preferably a platinum catalyst. The catalyst is rhodium. Platinum catalysts include, for example, chloroplatinic acid and alcohol-modified chloroplatinic acid. or a platinum complex having a chelate structure. Furthermore, the above curing catalyst can be used alone or Two or more types can be used in combination. The amount of curing catalyst added is as described above for polyorgans. The amount of platinum relative to nosiloxane is 5 ppm or more and 100 ppm or less by weight. If the added amount is less than 5 ppm, the curing of the polyorganosiloxane will not proceed sufficiently. There is a tendency for this to happen. However, when the amount of this additive exceeds 100 ppm, the hardening process becomes faster. Therefore, a uniform composition tends not to be obtained. The amount of the additive is more preferably 10p The concentration is between 15 ppm and 70 ppm, and more preferably between 15 ppm and 40 ppm. That is the case.

[0052] The crosslinking agent reacts with the alkenyl group-containing polyorganosiloxane, and the above polyorganosiloxane It can be added for the purpose of forming the main skeleton of the xane. Conventional crosslinking agents are known. Any crosslinking agent can be used, and is not particularly limited. The crosslinking agent contains a small amount per molecule. Both are organohydrogenpolysiloxanes having three hydrosilyl groups. Preferred. Examples of such crosslinking agents include polymethylhydrogensiloxane, Alternatively, poly(dimethylsiloxane-methylhydrogensiloxane), etc., can be used. Yes, it is possible. Furthermore, the above crosslinking agents can be used individually or in combination of two or more types. The amount of crosslinking agent added depends on the [Si-H] group and alkenyl group-containing polyorganosyl group in the crosslinking agent. The molar ratio of [CH2=CH-] groups in xanes is expressed as "[Si-H] / [CH2=CH-]". If so, the value is between 0.5 and 1.5. As a result, the cross-linking of the above polyorganosiloxane tends to be insufficient, and sufficient strength cannot be obtained. In contrast, if the amount of additive exceeds 1.5, crosslinking proceeds excessively, resulting in reduced flexibility. It tends to be lost. The amount added is more preferably 0.7 or more and 1.3 or less. More preferably, it is 0.8 or more and 1.2 or less.

[0053] The joint 40 may contain other components in addition to the silicone resin. Examples include fillers, reaction inhibitors, viscosity modifiers, and so on.

[0054] The filler controls the thermal conductivity and strength of the adhesive composition constituting the joint 40, and adjusts its viscosity. It can be added for the purpose of at least one of the following. As a filler, any of the conventionally known ones Other fillers can also be used, and are not particularly limited. Examples of fillers include silica. For example, wet silica, dry silica, fumed silica, fused silica, or fused spherical silica (e.g., lichen), alumina, aluminum nitride, boron nitride, zirconia, silicon nitride, or Silicon carbide and the like can be used as fillers. More preferably, silica, alumina, Aluminum nitride or boron nitride, more preferably alumina, a It is luminium. Note that the above filler material can be used individually or in combination of two or more types. It is possible.

[0055] The average particle size of the filler is not particularly limited, but it should be between 5 nm and 50 μm. Preferably, if the average particle size is less than 5 nm, the specific surface area of ​​the filler increases, Compared to cases where the specific surface area of ​​the filler is small, there is a larger amount of silicone resin covering the surface of the filler. Therefore, a silicone resin that can flow between fillers in the entire adhesive composition. When the amount of resin is low, the viscosity of the resin increases, which in turn tends to reduce the moldability during sheet molding and other processes. In contrast, when the average particle diameter exceeds 50 μm, the particle diameter is large, so the particles Compared to cases with smaller diameters, it becomes more difficult to control the sheet thickness during sheet molding and other processes. The flatness of the surface of the joint 40 tends to decrease. The average particle size of the filler is Preferably, 50 nm or more and 40 μm or less, more preferably 100 nm or more and 30 μm It is less than or equal to m. Note that the average particle size of the filler is determined by particle size distribution measurement using laser diffraction. It can be determined as the average mass D50 (or median diameter). Also, the filler components The shape is not particularly limited.

[0056] The filler material has a hydrophobic surface, which is intended to improve dispersibility in the polyorganosiloxane mentioned above. It may be surface-treated. For example, silica as a filler may be organosilane or Surface treatment agents such as organosilicon compounds including organosilazanes and diorganopolysiloxanes The surface may be treated. Note that the amount of surface treatment agent added and the surface treatment method are not particularly relevant. Not limited.

[0057] The filler may be added to the adhesive composition in powder form, or the filler may be dissolved in an organic solvent. It may also be added to the adhesive composition in the form of a dispersed slurry.

[0058] The amount of filler to be added is 1 part by weight or more, and 600 parts by weight, per 100 parts by weight of the above silicone resin. It is less than or equal to parts by weight. If the amount of the additive is less than 1 part by weight, it may affect the control of thermal conductivity, strength, or viscosity. The effect of degree adjustment tends to be difficult to obtain. In contrast, when the amount of the additive exceeds 600 parts by weight... This tends to reduce the flexibility of the adhesive composition. The amount of filler added is more , 100 parts by weight or more, 500 parts by weight or less, more preferably 200 parts by weight or more, 400 It is less than the weight of parts.

[0059] Reaction inhibitors can be added to adjust the curing rate of the adhesive composition. Any conventionally known reaction inhibitor can be used as the reaction inhibitor, and is not particularly limited to any other type. No. As reaction inhibitors, for example, 3-methyl-1-buty-3-ol, 3-methyl- 1-Penthin-3-ol, 3,5-dimethyl-1-hexyn-3-ol, 1-ethyn Lucyclohexanol, 3-methyl-3-trimethylsiloxy-1-butyne, 3-methyl -3-trimethylsiloxy-1-pentine, 3,5-dimethyl-3-trimethylsiloxy -1-Hexine, 1-Ethynyl-1-Trimethylsiloxycyclohexane, Bis(2,2 -dimethyl-3-butinoxy)dimethylsilane, 1,3,5,7-tetramethyl-1,3 ,5,7-tetravinylcyclotetrasiloxane, 1,3,5,7-tetramethyl-1, 3,5,7-Tetrahexenylcyclotetrasiloxane, 1,1,3,3-Tetramethyl Examples include -1,3-divinyldisiloxane and triallyl isocyanurate. Reaction suppression The inhibitor is more preferably 1,3,5,7-tetramethyl-1,3,5,7-tetravinyl Lucyclotetrasiloxane, 1,1,3,3-tetramethyl-1,3-divinyldisiloxane The above reaction inhibitors are triaryl isocyanurate. The above can be used in combination. The amount of reaction inhibitor to be added depends on the silicone resin. The amount added is 10 parts by weight or less per 100 parts by weight. If the amount added exceeds 10 parts by weight, It tends to inhibit normal curing and cause curing failure. The amount of reaction inhibitor added is preferable. Alternatively, 0.1 parts by weight or more, 6 parts by weight or less, more preferably 0.2 parts by weight or more, double The amount is less than or equal to parts. Furthermore, the reaction inhibitor controls the catalytic activity of the platinum catalyst, etc., and the adhesive composition To prevent thickening or gelation before heat curing, it may be added as needed. That is the case.

[0060] Viscosity modifiers are used to reduce variations in sheet thickness when manufacturing adhesive sheets. Viscosity modifier. Any conventionally known viscosity modifier can be used, and is not particularly limited. As a degree modifier, for example, fumed silica, colloidal silica, fumed silica, Examples include alumina, fumed alumina, and colloidal alumina. Viscosity modifiers are also used. More preferably, it is atomized silica or atomized alumina. It can be used alone or in combination of two or more types. Furthermore, viscosity modifiers are indicated on their respective tables. The surface may be hydrophobic (surface treated), etc. The amount of viscosity modifier to be added is the same as the above silicone. The amount of additive is 0.05 parts by weight or more and 10 parts by weight or less per 100 parts by weight of resin. When the amount is less than 0.05 parts by weight, the viscosity adjustment effect tends not to be obtained. If the amount of additive exceeds 10 parts by weight, the viscosity increases too much, reducing fluidity and increasing the thickness of the sheet. The amount of viscosity modifier added is more preferably 0.1 parts by weight or less. The amount is 5 parts by weight or less, more preferably 0.5 parts by weight or more and 2 parts by weight or less.

[0061] The glass transition temperature (Tg) of the above adhesive composition is -60°C or lower. When the glass transition temperature exceeds -60°C, flexibility, adhesion, and elongation are maintained in low-temperature environments. The properties related to this tend to decrease. In other words, the flexible elastomer of the joint 40 -The properties as rubber tend to decrease. In contrast, the glass transition temperature of the adhesive composition If the temperature is below -60°C, the flexibility, adhesion, and elongation will remain even in low-temperature environments. The good properties tend to be maintained. The glass transition temperature of the adhesive composition is more The temperature is -70°C or lower, and more preferably -80°C or lower.

[0062] The water content in the above adhesive composition is 10 ppm or more and 5000 ppm or less. The water content in the composition affects the hydrolysis of the silane coupling agent contained in the adhesive composition. This has an effect. Here, the silane coupling agent exerts its adhesive effect by hydrolysis. If the moisture content is less than 10 ppm, the hydrolysis of the silane coupling agent proceeds smoothly. This is insufficient to do so, and consequently, it suppresses the improvement of the adhesion of the joint 40. Also, the amount of moisture When it exceeds 5000 ppm, during the heating / cooling thermal cycle process in semiconductor manufacturing, The moisture contained in the joint 40 aggregates at the joint surface with the plate-like part 20 and the joint surface with the base part 30. Condensation can reduce the adhesion between these joint surfaces. The water content in the composition is more preferably 20 ppm or more and 4000 ppm or less. The moisture content can be measured using a Karl Fischer moisture analyzer.

[0063] In the electrostatic chuck 10, generally, the base portion 30 has a greater thermal expansion coefficient than the plate-shaped portion 20. They are numerous and expand and contract significantly with temperature changes. Therefore, depending on the temperature conditions, When the thermal expansion of the - section 30 is greater than the thermal expansion of the plate-shaped section 20. This is also because the degree of expansion and contraction differs between the plate-like portion 20 and the base portion 30. Furthermore, at the joint 40, a shear force is applied in the X-axis direction (Figure 1), generating shear stress. Therefore, as described above, the magnitude of the shear strain in the joint 40 is increased. Therefore, even when a large shear force is applied to the joint 40, the shear force generated at the joint 40 This reduces stress and minimizes damage to the joint 40.

[0064] (A-3) Method for manufacturing the electrostatic chuck 10: Next, the manufacturing method of the electrostatic chuck 10 in this embodiment will be described. First, the electrostatic charge A plate-like portion 20 is fabricated in which a conductive material layer such as an electrode 22 is arranged inside. The fabrication of the plate-like portion 20 is For example, this can be done by known sheet lamination methods or press forming methods.

[0065] An example of a method for producing a plate-like portion 20 by the sheet lamination method is as follows. First, alumina The raw materials, butyral resin, plasticizer, and solvent are mixed, and the resulting mixture is subjected to the doctor blade method. By forming them into sheets, multiple ceramic green sheets are produced. Furthermore, through-holes and via holes are formed in a predetermined ceramic green sheet. Necessary processing is performed, such as filling with ink and applying electrode ink for forming the adsorption electrode 22. The areas coated with electrode ink become conductive material layers. For example, conductive materials such as tungsten and molybdenum, alumina raw materials, and ethoxycellulose. (Registered Trademark) A metallized ink, which is a slurry formed by mixing resin and solvent, is used. Subsequently, multiple ceramic green sheets are laminated and heat-pressed together, and then processed to the specified size. By doing so, a ceramic molded body is obtained.

[0066] After the obtained ceramic molded body is degreased in nitrogen, a predetermined process is performed in a humidified hydrogen-nitrogen atmosphere. The plate-like portion 20 is produced by firing at atmospheric pressure at a temperature (for example, 1500°C to 1600°C). do.

[0067] Next, the plate-shaped part 20 and the base part 30 are joined together via the joint part 40. Specifically, the plate In addition to the shaped part 20, there is a base part 30 and an adhesive (adhesive sheet) made of the above adhesive composition. Prepare the base (preparation step). The base 30 is formed of, for example, aluminum. The adhesive is an adhesive composition (silicone resin composition) containing polyorganosiloxane. The adhesive sheet is prepared by stirring the above adhesive composition under vacuum to create an adhesive paste. The prepared adhesive paste is then molded into a sheet using a roll coater or the like, as needed. It is then manufactured by heating it at a predetermined temperature for a predetermined time to partially harden it.

[0068] Adhesive is placed between the plate-shaped part 20 and the base part 30, and they are bonded together in a vacuum, and then left as is for 60 Heat to a temperature of 150°C or higher (for example, 60°C to 150°C). This will harden the adhesive. A joint portion 40 is formed, and the plate-shaped portion 20 and the base portion 30 are bonded together by the joint portion 40. The body is formed (joint formation process). After the joint is returned to room temperature (e.g., 25°C), the bonding is performed. The measurement of the dynamic viscoelasticity of the cured agent was performed at a temperature where the loss factor tanδ was 0.2 or higher for 1 hour. The above is the process of annealing the bonded body. The temperature at which the loss coefficient tanδ is 0.2 or higher. The degree was determined by performing DMA (dynamic viscoelasticity) measurements using the adhesive that constitutes the joint, and the measurement results This can be determined from the results. For example, in the case of an annealing process where the temperature is held at -70°C for 24 hours... In total, the temperature was lowered from 20°C to -70°C over 2 hours, held there for 24 hours, and then returned to 20°C over 8 hours. This process can be carried out as an annealing step.

[0069] Afterward, post-processing (polishing of the outer circumference, formation of terminals, etc.) is performed as necessary. Thus, an electrostatic chuck 10 with the above configuration is manufactured.

[0070] Figure 5 shows the shape changes of each part in each step of the manufacturing method of the electrostatic chuck 10 of this embodiment. This is an explanatory diagram that conceptually illustrates the concept. Figure 5(A) shows the shape of each part in the preparation process. In the example shown in Figure 5, the preparation process The first surface 24 of the plate-like portion 20 is a concave surface, and the depth of the concave is . On the other hand, the base portion 30 The surface that is joined to the plate-like portion 20 is a convex surface. Figure 5(B) shows the process of forming the joined body. Joined body shape During the forming process (thermosetting), each part expands and deforms independently due to the rise in temperature, causing stress in each part. The bonding is performed in a state where no such thing occurs. At this time, as shown in the figure, the first surface 24 of the plate-like portion 20 is approximate It is flat and has good flatness. Figure 5(C) shows the joint after it has returned to room temperature. When returned to its original position, the plate-like portion 20 and the base portion 30 contract to return to their original shape. Since the base portion 30 has a greater coefficient of thermal expansion than the plate-shaped portion 20, it shrinks more than the plate-shaped portion 20. Therefore, a large stress is applied to the outer circumference of the joint, and the first surface 24 of the plate-like portion 20 becomes convex. Figure 5(D) shows the annealing process. In the annealing process, stress relief occurs in the joint 40. Due to the properties of the adhesive's viscosity (liquid), the joint 40 itself deforms, resulting in low stress. This occurs, and the plate-shaped part 20 and the base part 30 return to their original shapes, causing the plate-shaped part 2 The flatness of the first surface 24 of 0 is improved. As shown in the figure, at the end of the annealing process, the plate-like portion 2 The first face 24 of 0 is a concave surface, and the depth of the concavity is D2. Here, D2 <D1である。すな Furthermore, the first surface 24 of the plate-like portion 20 at the end of annealing is concave, and the depth D2 of the concave is due to the preparation work. It is shallower than the depth D1 of the recess on the first surface 24 of the plate-like portion 20.

[0071] [Evaluation of adhesives] Using four different types of adhesives, samples of adhesive sheets and the embodiment shown in Figure 1 An electrostatic chuck was fabricated to improve loss tangent, storage modulus, shear strain, and flatness of the plate-like part. The delamination of the joints was examined. Figure 6 is an explanatory diagram summarizing the evaluation results. Figure 7 shows the storage This is an explanatory diagram showing the modulus of elasticity. Figure 8 is an explanatory diagram showing the loss tangent. Figure 9 is shown in Figure 8. This is an explanatory diagram showing a magnified portion of the loss tangent figure.

[0072] In Figure 6, adhesives 1-4 are used to create two different joint thicknesses: 0.2 mm and 0.7 mm. Examples of samples 1-8 with formed joints are shown. The compositions of adhesives 1-4 are as follows: That's right. Adhesive 1: A silicone resin containing phenyl groups is mixed with a filler containing aluminum nitride powder. Filling (44 vol%) Adhesive 2: Silicone resin containing phenyl groups filled with alumina filler (44 vol%) ) Adhesive 3: Silicone resin filled with alumina filler (44 vol%) Adhesive 4: Fill epoxy resin with alumina filler.

[0073] In the example shown in Figure 6, there is a plate-like part mainly composed of alumina and a base part made of aluminum. I used it.

[0074] <Preparation of sheet-like samples> Mix the resin and filler in the proportions specified in the composition of each adhesive to form a paste-like adhesive. It was manufactured. There are no particular restrictions on the mixing method, including mixing with known stirring blades, three-roll mixing, and kneading. - A rotation / revolution mixer, planetary mixer, etc. can be used. Spread the paste onto the polyethylene terephthalate (PET) film. The method used can be a known method, and in this performance evaluation, a doctor blade is used. Next, the adhesive paste spread on the PET film is cut to the desired size, and then The adhesive paste with the cut PET film attached is dried in a dryer for a predetermined time and at a predetermined temperature. By heating it, the adhesive paste is partially cured. This allows the PET film to be attached. An adhesive sheet is formed. During heating, if necessary, such as to prevent dust from adhering, Each adhesive paste may be covered with a cover film.

[0075] <Preparation of electrostatic chuck-type samples> The electrostatic chuck sample has a similar configuration to the electrostatic chuck 10 in the embodiment shown in Figure 1. It has the following characteristics. A sample of the form of the electrostatic chuck 10 is made by the electrostatic chuck manufacturing method described above. The plate-like portion 20 is made of alumina (with a thermal expansion coefficient of 6.8). 0 x 10 -6 Using (°C), the base portion 30 is made of aluminum. The thermal expansion coefficient of the section is 2.30 × 10 -5 The temperature ( / °C) was used. The diameters of the plate-like portion 20 and the joint portion were The thickness was set to 350 mm. Furthermore, all samples underwent an annealing process at -70°C for 2 minutes. The sample was held for 4 hours.

[0076] <Evaluation Method> • Loss factor tanδ and storage modulus E' The measurement was performed using a known automatic dynamic viscoelasticity measuring device under the following conditions: Loss coefficient tanδ = loss It can be calculated as the demodulation modulus G'' / storage modulus G'. Measurement sample (adhesive sheet) size: 4mm x 40mm, thickness 0.35mm Measurement temperature: -150℃~300℃ Heating rate: 2°C / min Preload (load): 1g Load direction: Tension Excitation frequency: 11Hz Excitation amplitude: 16μm

[0077] Shear strain (%) Shear strain was measured using a known tensile testing machine (Shimadzu Autograph AGS-5kNX). It was used and measured by tensile testing. Size of the measurement area sample (adhesive sheet): 12.5 x 12.5 mm

[0078] Figure 10 is a schematic diagram illustrating the method for calculating shear strain. Figure 10(A) shows the shear strain. Figure 10(B) shows the tension test from the front, while Figures 10(C) show it from the side. This represents the child. Also, Figures 10(A) and 10(B) show the situation at the start of the test, and Figure 10 (C) shows the state after the start of the test. Test piece 70 is a semi-cured adhesive sheet of each sample. , two aluminum plates measuring 12.5mm wide x 100mm long x 1mm thick, from one end of 80 Attach two pieces to the 12.5mm x 12.5mm area up to the 2.5mm mark. After bonding the aluminum plates 80 together in a orientation that allows them to be pulled in opposite directions, It was fabricated by heating at 0°C for 10 hours, and then heating at 150°C for 50 hours to bond the parts together. The thickness t of the test specimen 70 at the start of the test was set to the thickness of the joint shown in Figure 6.

[0079] Next, the two aluminum plates 80 were subjected to a tensile test so that a shear force was applied to the test specimen. The parts were gripped with a jig and moved relative to each other. Here, a tensile testing machine was used to measure one aluminum part. While moving the board in one direction parallel to the adhesive surface at a tensile speed of 2 mm / min, the load and movement The strain amount δ as a distance was measured (see Figure 10(C)). The tensile testing machine and fixtures were sufficient. A highly rigid material was used, and the distance traveled by the jig and fixture of the tensile testing machine was defined as the strain amount δ. Figure 10( In B), the relative direction of movement of the two aluminum plates is indicated by white arrows. By dividing the weight by the bonding area of ​​the test piece before movement (12.5 mm × 12.5 mm), the result is... The breaking stress was calculated. The relative movement of these two aluminum plates was used to determine when the test specimen 70 fractured. Continue until the shear stress reaches its maximum, and the maximum shear stress (unit: The strain at maximum shear stress (in mm) was set to MPa. The strain at the start of the test was measured for specimen 7. By dividing by the thickness t (in mm) (Figure 5(B)), the shear strain of the test specimen 70 can be calculated. The percentage (%) was calculated. Figure 10 shows the amount of strain δ when the shear stress is at its maximum. Yes, they are.

[0080] Furthermore, the shear strain of the joint already incorporated into the electrostatic chuck (unit: mm) When measuring ), for example, the following procedure is performed: First, the processing method such as laser cutting. This process cuts out the joint along with the adherend (plate-like part and base part). The shape of the test piece to be cut out. The two adherends, which are joined together, can be held in the fixture of a tensile testing machine, as shown in Figure Any shape that can be pulled in opposite directions as shown in 5 is acceptable. Tensile test Before proceeding, measure the area and thickness of the joint in the cut-out test piece. Subsequently, a tensile test is performed in the same manner as the method described above, and the strain amount at the maximum shear stress is measured. That's all.

[0081] · Improvement in flatness The flatness of the first surface 24 of the plate-like portion 20 in the preparation process of the above-described electrostatic chuck manufacturing method is compared with the flatness of the first surface 24 of the plate-like portion 20 after the annealing process. If the flatness is reduced, it is described as "yes" for the improvement of flatness. Here, the flatness is measured by a contact measurement method and calculated by the following method. A least-squares plane (virtual plane) is obtained from a plurality of measurement points, and the sum of the absolute values of the positive maximum value and the negative maximum value of the distance from that plane is defined as the flatness.

[0082] <Evaluation result> As shown in FIGS. 6 to 9, Samples 1 to 6 satisfy the following requirement [1]. [1] The joint has a storage modulus E' in the dynamic viscoelasticity measurement of 10 MPa or less at 25°C. That is.

[0083] As shown in FIG. 9, Samples 1 to 6 further satisfy the following requirement [2]. [2] The joint has a temperature range in which the loss factor tanδ in the dynamic viscoelasticity measurement is 0.2 or more. Exists.

[0084] As shown in FIG. 6, Samples 1 to 4 further satisfy the following requirement [3]. [3] The joint has a loss factor tanδ at -70°C in the dynamic viscoelasticity measurement of 0.2 or more. That is.

[0085] As shown in FIG. 6, Samples 1, 3, and 5 satisfy the following requirement [4]. [4] A bonding material with a shear strain of 250% or more at -70°C and 80°C when the thickness is 0.2 mm is used for the joint, and the thickness of the joint is 0.2 mm or more and 1.0 mm or less. That is.

[0086] As shown in Figure 6, samples 2, 4, and 6 satisfy the following requirement [5]. [5] At the joint, when the thickness is 0.7 mm, the shear strain at -70°C and 80°C is less than 70%. The upper joining material is used, and the thickness of the joint is 0.2 mm or more and 1.0 mm or less.

[0087] As shown in Figure 6, samples 7 and 8 do not satisfy any of the above requirements [1] to [5]. Furthermore, even after performing the annealing process, the flatness of the first surface 24 of the plate-like portion 20 did not improve. Samples 5 and 6 did not meet the requirements [3]. And in the manufacturing process of the electrostatic chuck Even after performing the annealing process, the flatness of the first surface 24 of the plate-like portion 20 did not improve.

[0088] Samples 1-4 (adhesives 1 and 2) satisfy the above requirements [1]-[3] and requirement [4], [5] either of the following conditions is met. And, in the manufacturing process of the electrostatic chuck, the annealing process After performing this procedure, the flatness of the first surface 24 of the plate-like portion 20 was improved.

[0089] As shown in Figure 9, at -70°C, adhesive 1 exhibits a viscosity that is five times greater than that of adhesive 3. The above is present (adhesive 3: 0.05 or less, adhesive 1: approximately 0.25). Similarly, adhesive 2 is also present. At -70°C, its viscosity is more than 6 times that of adhesive 3 (adhesive 3: 0.05). The following is an adhesive ratio of approximately 2:0.3). Therefore, stress relaxation occurs by annealing at -70°C. As a result, the plate-like portion 20 returns to its original non-warped shape, and the flatness improves.

[0090] As described above, the electrostatic chucks of samples 1-4, which were made using adhesives 1 and 2, were prepared. We were able to improve the flatness of the first surface of the plate-like part in the process.

[0091] [Evaluation of manufacturing methods] The effect of the annealing process in the manufacturing method of electrostatic chucks was evaluated. For details, see section 4 above. Using a type of adhesive, electrostatic chucks for samples S1 to S11 (embodiment shown in Figure 1) were fabricated. The delamination of the joint after the thermal cycling test and the flatness of the first surface of the plate-like portion were evaluated. The thickness of the joint is 0.3 mm.

[0092] Figure 11 is an explanatory diagram summarizing the evaluation results of the manufacturing method for electrostatic chucks. As described above, samples S1, S2, S4, S5, S7, S9, and S10 are as described in the above embodiment. It was manufactured using the manufacturing method (with an annealing step). However, samples S1, S4, and S9 were annealed. The Neil temperature is the temperature at which the loss coefficient tanδ becomes greater than 0.2, as in samples S2 and S5. S7 and S10 are annealing temperatures where the loss coefficient tanδ is 0.2 or less. Lines S3, S6, S8, and S11 are the annealing steps in the manufacturing method described in the above embodiment. It was manufactured without being implemented.

[0093] In Figure 11, the indication "Storage modulus of elasticity 10 MPa or less" refers to the joint at 25°C. This refers to the storage modulus at 25°C. The lower the storage modulus, the softer the joint. If the storage modulus is 10 MPa or less, the joint is sufficiently soft, and the temperature change of the retaining device Because the tensile stress generated on the outer circumference of the retaining device can be reduced, the outer circumference of the joint It can prevent peeling.

[0094] <Evaluation Method> • Peeling Ultrasonic testing was performed on the sample after the thermal cycling test, and the outer circumference of the joint was found to be Check for the presence or absence of peeling. Here, the outer peripheral part is within a range of a predetermined distance from the periphery of the joint part towards the center. In this evaluation, the predetermined distance is 2 mm. In this evaluation, peeling with a major axis (longest length) of 2 mm or more was regarded as peeling.

[0095] The conditions of the thermal cycle test are as follows. One cycle consists of holding at -70°C for 3 hours and then holding at 80°C for 3 hours, and 20 cycles are performed.

[0096] · Flatness (μm) Flatness was measured by a contact measurement method and calculated by the following method. A least-squares plane (virtual plane) was obtained from a plurality of measurement points, and the sum of the absolute values of the positive maximum value and the negative maximum value of the distance from that plane was defined as flatness.

[0097] Evaluation of flatness was performed according to the following criteria. ◎: Less than 20 μm 〇: Between 20 μm and less than 40 μm ×: 40 μm or more

[0098] <Evaluation result> As shown in Fig. 11, there is no peeling after the thermal cycle test for sample S1, and the flatness of the first surface of the plate-like part is also very good. The adhesive 1 used for the joint part of sample S1 has a storage elastic modulus of 10 MPa or less (3.0 MPa) and is soft. And in the production of sample S1, annealing is performed at a temperature at which the loss coefficient tanδ in the dynamic viscoelasticity measurement is 0.2 or more. In the temperature range where the loss coefficient tanδ in the dynamic viscoelasticity measurement is 0.2 or more, viscous properties appear. When viscous properties appear, the stress can be gradually reduced. Therefore, by performing the annealing process, stress relaxation and creep, which are viscous properties, occur in the adhesive. ​​​​​​​​​This occurs, the adhesive deforms, and the plate-like part and the base part return to their original shape, that is, to a state without warping. By doing so, it is believed that the flatness of the first surface of the plate-like portion could be improved.

[0099] Sample S2 showed no peeling after the thermal cycling test, and the flatness of the first surface of the plate-like portion was also good. However, compared to sample S1, the flatness value of the first surface of the plate-like part is larger. In pull S2, the joint is constructed using the same adhesive 1 as in sample S1, and the joint is flexible. Because it was soft, the flatness of the first surface of the plate-like part could be made good. However, the sample Compared to S1, the improvement in flatness is smaller because the annealing conditions are not favorable. ru.

[0100] Sample S3 showed no peeling after the thermal cycling test, but the flatness of the first surface of the plate-like portion was... The values ​​are higher compared to samples S1 and S2, which is not good. In sample S3, Sun The joint is constructed using the same adhesive 1 as Pull S1, and because the joint is soft, the plate-like part The flatness value of the first surface is not particularly large, but it is greater than 40 μm. Sample S1, Compared to S2, the annealing process was not performed, resulting in the same reaction that occurred during the manufacturing of sample S3. This is thought to be because the force cannot be mitigated.

[0101] Sample S4 showed no peeling after the thermal cycling test, and the flatness of the first surface of the plate-like portion was also very high. Good. The adhesive 2 used in the joint of sample S4 has a storage modulus of 10 MPa or less. It is very soft at the lower pressure (1.1 MPa). And in the production of sample S4, dynamic viscoelasticity Because annealing is performed at a temperature where the loss factor tanδ in the performance measurement is 0.2 or higher, As described above, it is believed that we were able to improve the flatness of the first surface of the plate-like portion.

[0102] Sample S5 showed no peeling after the thermal cycling test, and the flatness of the first surface of the plate-like portion was also very high. Although good, the flatness value of the first surface of the plate-like portion is higher compared to sample S4. In sample S5, the joint is constructed using the same adhesive 2 as in sample S4, and the joint is formed. Because the material is soft, it was possible to achieve good flatness on the first surface of the plate-like portion. However, Compared to sample S4, the improvement in flatness is small because the annealing conditions are not suitable. It can be obtained.

[0103] Sample S6 showed no peeling after the thermal cycling test, and the flatness of the first surface of the plate-like portion was also very high. Although good, the flatness value of the first surface of the plate-like part is lower compared to samples S4 and S5. It's large. In sample S6, the joint is constructed using the same adhesive 2 as in sample S4. Because the joint is very soft, although the flatness of the first surface of the plate-like part is very good, Compared to samples S4 and S5, sample S6 was manufactured without the annealing process. This is thought to be because the stress that occurs at times cannot be relieved.

[0104] Although sample S7 showed delamination after the thermal cycling test, the flatness of the first surface of the plate-like portion was good. The results were good. The adhesive 3 used in the joint of sample S7 had a storage modulus of 10 MPa. Since it was soft at a MPa or less (2.1 MPa), it is thought that the flatness of the first surface of the plate-like part was good. However, the annealing process was carried out under the same conditions as sample S1 (-70°C). (Hold for 24 hours), the loss factor tanδ of adhesive 3 is very small at -70°C. (Figure 9) During the annealing process, the joint is close to an elastic body and cannot sufficiently relieve stress. It is thought that good flatness could not be obtained because the stress could not be sufficiently relieved. It is believed that peeling occurred because there was no protective layer.

[0105] Although sample S8 also showed peeling after the thermal cycling test, the flatness of the first surface of the plate-like portion was good. The results were good. In sample S8, the annealing process was not performed, so the stress was not sufficiently relaxed. It is believed that the adhesive could not be bonded properly, resulting in peeling. Sample S8 uses the same adhesive as Sample S7. The joint is constructed using 3, and because the joint is soft, the flatness of the first surface of the plate-like part is improved. I was able to make it work.

[0106] Sample S9 showed delamination after the thermal cycling test. Also, adhesive 3 caused delamination in the plate-like portion. Because the base portion could not be joined, the flatness could not be measured (in Figure 11, "- (Illustrated). The adhesive 4 used in the joint of sample S9 has a storage modulus of 10 MPa. It is very large (1000 MPa) and extremely hard. Therefore, in the manufacturing process of sample S9 Delamination occurred before the annealing process, and the loss factor tanδ in the dynamic viscoelasticity measurement was 0.2 Even when the annealing process was performed at temperatures above the specified level, it was not possible to join the plate-like portion and the base portion. .

[0107] Sample S10, like Sample 9, also showed delamination after the thermal cycling test, and its flatness was also affected. Measurement was not possible. The manufacturing of sample S10 used different annealing conditions than sample S9. Although the annealing process was performed, the plate-like part and the base part were not joined together as in sample S9. I was unable to do that.

[0108] Sample S11, like Sample 9, also showed delamination after the thermal cycling test, and its flatness was also affected. Measurement was not possible. The manufacturing process for sample S11 did not include an annealing step. Similar to the S9 model, it was not possible to join the plate-like portion and the base portion.

[0109] In these examples, the storage modulus E' in the dynamic viscoelasticity measurement of the joint is 10 MPa at 25°C. If it is less than or equal to a, the joint is sufficiently soft, and the interface between the joint and the plate-like part, and the joint and the base part This can suppress peeling at the interface with the material. Furthermore, in the manufacture of electrostatic chucks, By performing the polishing process, the flatness of the first surface of the plate-like portion can be improved. The annealing process is performed at a temperature at which the loss coefficient tanδ in the dynamic viscoelasticity measurement is 0.2 or higher. By doing so, the flatness of the first surface of the plate-like portion can be further improved.

[0110] <Modified form of this embodiment> The present invention is not limited to the embodiments described above, and may be described in part without departing from its spirit. It can be implemented in various forms, and for example, the following modifications are also possible.

[0111] The above embodiment shows an example in which the plate-shaped portion 20 and the base portion 30 are joined by the joint portion 40. However, even if there is another ceramic substrate or inorganic substrate between the plate-shaped portion 20 and the joint portion 40 Good. Also, another ceramic substrate is bonded onto the first surface 24 of the plate-shaped portion 20, The configuration may also include holding the object on top.

[0112] In the manufacturing method of the above embodiment, the plate-like portion prepared in the preparation step has a concave first surface. Although an example has been shown, the plate-like portion prepared in the preparation step may have a convex first surface. In this case, the plate-like portion at the end of the annealing process has a convex first surface, and the height of the convexity is the joint formation. It is preferable that the surface is lower than the convex surface of the first surface of the plate-like portion at the end of the process.

[0113] The method for manufacturing the electrostatic chuck is not limited to the above embodiments, and can be manufactured by various known methods. This can be done. For example, the annealing process does not need to be performed. Also, the conditions of the annealing process (temperature) can be changed. The degree and time may be as described in the above embodiment. The annealing process is preferably performed for 1 hour or more. It's nice.

[0114] • In the above embodiment, an electrostatic chuck was exemplified as a holding device, but the holding device is electrostatic Not limited to chucks. For example, CVD, PVD, PLD (Pulsed Laser Discharge). It is configured as a heater device, susceptor, and mounting base for vacuum equipment such as Deposition. It is possible.

[0115] • In the above embodiment, an example is provided in which the holding device comprises a laminate of plate-shaped members with a substantially circular plane. As shown, the planar shape is not limited to the above embodiment. For example, a rectangular plane, a polygonal plane, etc. It's okay to have it.

[0116] This disclosure is not limited to the embodiments described above, but may extend to the extent that it does not depart from the spirit thereof. It can be realized in various configurations. For example, in each of the embodiments described in the section on the summary of the invention... The technical features in the embodiments corresponding to the technical features solve some or all of the above-mentioned problems. For this purpose, or to achieve some or all of the effects described above, substitutions or combinations may be made as appropriate. It is possible to combine them. Furthermore, the technical features of this specification are not considered essential in this specification. If not explained, it can be deleted as appropriate.

[0117] This disclosure can also be realized in the following application examples. [Application Example 1] A holding device for holding an object, It has a first surface which is the side on which the object is placed, and a second surface which is the back surface of the first surface. A plate-like part, It consists of a material having a different coefficient of thermal expansion from the material that mainly forms the plate-like portion. a plate-shaped base portion is arranged on the second surface side of the plate-shaped portion, It is positioned between the plate-like portion and the base portion to join the plate-like portion and the base portion. The joint and, Equipped with, The flatness of the first surface of the plate-like portion is 40 μm or less, and the temperature is -70°C to 80°C. A characteristic feature is that there is no peeling on the outer circumference of the joint after the cycle test. holding device. [Application Example 2] The holding device described in Application Example 1, The aforementioned joint has a storage modulus E' of 10 MPa or less at 25°C in dynamic viscoelasticity measurements. Characterized by, holding device. [Application Example 3] A holding device as described in Application Example 1 or Application Example 2, The aforementioned joint is in a temperature range where the loss coefficient tanδ in dynamic viscoelasticity measurement is 0.2 or higher. Characterized by existence, holding device. [Application Example 4] A holding device according to any one of Application Examples 1 to 3, The aforementioned joint has a loss factor tanδ of 0.2 or more at -70°C in dynamic viscoelasticity measurement. Characterized by a certain, holding device. [Application Example 5] A holding device according to any one of Application Examples 1 to 4, When the thickness of the joint is 0.2 mm, the shear strain at -70°C and 80°C is 250%. Using the above-mentioned joining materials, The thickness of the joint is 0.2 mm or more and 1.0 mm or less. holding device. [Application Example 6] A holding device according to any one of Application Examples 1 to 4, When the thickness of the joint is 0.7 mm, the shear strain at -70°C and 80°C is less than 70%. Using the upper joining material, The thickness of the joint is 0.2 mm or more and 1.0 mm or less. holding device. [Application Example 7] A method for manufacturing a holding device for holding an object, It has a first surface which is the side on which the object is placed, and a second surface which is the back surface of the first surface. A plate-like portion and a material having a thermal expansion coefficient different from that of the material mainly forming the plate-like portion. A preparation step involves preparing a plate-shaped base made of the material, an adhesive, and The base portion is placed on the second surface side of the plate-like portion via the adhesive, and the temperature is 60°C or higher. The adhesive is heated to a certain temperature to harden, and the bond, which is the hardened product of the adhesive, forms the plate-like shape. A joint formation step in which a joint is formed by joining a part and the base part, The temperature at which the loss factor tanδ in the dynamic viscoelasticity measurement of the cured adhesive is 0.2 or higher An annealing step of annealing the bonded body at a temperature of 1 degree for more than 1 hour, A feature having A method for manufacturing a holding device. [Application Example 8] A method for manufacturing a holding device as described in Application Example 7, The plate-like portion prepared in the above preparation step has a concave first surface. At the end of the joint formation process, the plate-like portion has a convex first surface. At the end of the annealing process, the plate-like portion has a concave first surface, and the depth of the concave is the same as the preparation The process is characterized by being shallower than the concave surface of the first surface of the plate-like portion, A method for manufacturing a holding device. [Application Example 9] A method for manufacturing a holding device as described in Application Example 7, The plate-like portion prepared in the preparation step has a convex first surface, At the end of the joint formation process, the plate-like portion has a convex first surface, and the height of the convex is Higher than the preparation process, At the end of the annealing process, the plate-like portion has a convex first surface, and the height of the convex is the same as the joining. It is characterized by being lower than the convex surface of the first surface of the plate-like portion at the end of the body forming process. A method for manufacturing a holding device. [Explanation of symbols]

[0118] 20...Plate-like part 22...Adsorption electrode 24...Side 1 26…Second side 30...Base section 32… Refrigerant flow path 40…Joint part 40c... Periphery 40cp...Outer circumference 50…Gas supply lines 52...Gas outlet 70…Test piece 80…Aluminum plate W...wafer

Claims

1. A holding device for holding an object, It has a first surface which is the side on which the object is placed, and a second surface which is the back surface of the first surface. A plate-like part, It consists of a material having a different coefficient of thermal expansion from the material that mainly forms the plate-like portion. , a plate-shaped base portion is arranged on the second surface side of the plate-shaped portion, It is positioned between the plate-like portion and the base portion to join the plate-like portion and the base portion. The joint and, Equipped with, The flatness of the first surface of the plate-like portion is 40 μm or less, and the temperature is -70°C to 80°C. A characteristic feature is that there is no peeling on the outer circumference of the joint after the cycle test. holding device.

2. A holding device according to claim 1, The aforementioned joint has a storage modulus E' of 10 MPa or less at 25°C in dynamic viscoelasticity measurements. Characterized by, holding device.

3. A holding device according to claim 1, The aforementioned joint is in a temperature range where the loss coefficient tanδ in dynamic viscoelasticity measurement is 0.2 or higher. Characterized by existence, holding device.

4. A holding device according to claim 1, The aforementioned joint has a loss coefficient tanδ of 0.2 or more at -70°C in dynamic viscoelasticity measurement. Characterized by a certain, holding device.

5. A holding device according to any one of claims 1 to 4, When the thickness of the joint is 0.2 mm, the shear strain at -70°C and 80°C is 250%. Using the above-mentioned joining materials, The thickness of the joint is 0.2 mm or more and 1.0 mm or less, holding device.

6. A holding device according to any one of claims 1 to 4, When the thickness of the joint is 0.7 mm, the shear strain at -70°C and 80°C is less than 70%. Using the upper joining material, The thickness of the joint is 0.2 mm or more and 1.0 mm or less, holding device.

7. A method for manufacturing a holding device for holding an object, It has a first surface which is the side on which the object is placed, and a second surface which is the back surface of the first surface. A plate-like portion and a material having a thermal expansion coefficient different from that of the material mainly forming the plate-like portion. A preparation step involves preparing a plate-shaped base made of the material, an adhesive, and The base portion is placed on the second surface side of the plate-like portion via the adhesive, and the temperature is 60°C or higher. The adhesive is heated to a certain temperature to harden, and the bond, which is the hardened product of the adhesive, forms the plate-like shape. A joint formation step in which a joint is formed by joining a part and the base part, The temperature at which the loss factor tanδ in the dynamic viscoelasticity measurement of the cured adhesive is 0.2 or higher An annealing step of annealing the bonded body at a temperature of 1 degree for more than 1 hour, A feature having A method for manufacturing a holding device.

8. A method for manufacturing a holding device according to claim 7, The plate-like portion prepared in the above preparation step has a concave first surface. At the end of the joint formation process, the plate-like portion has a convex first surface. At the end of the annealing process, the plate-like portion has a concave first surface, and the depth of the concave is the same as the preparation The process is characterized by being shallower than the concave surface of the first surface of the plate-like portion, A method for manufacturing a holding device.

9. A method for manufacturing a holding device according to claim 7, The plate-like portion prepared in the above preparation step has a convex first surface, At the end of the joint formation process, the plate-like portion has a convex first surface, and the height of the convex is Higher than the preparation process, At the end of the annealing process, the plate-like portion has a convex first surface, and the height of the convex is the same as the joining. It is characterized by being lower than the convex surface of the first surface of the plate-like portion at the end of the body forming process. A method for manufacturing a holding device.

Images