Base for substrate holding member and substrate holding device

Abstract

This technology provides a way to suppress pressure loss and reduce the risk of leakage from the flow path. [Solution] The ceramic base 150 has two main surfaces (upper surface 155 and lower surface 156) and has a flow channel 160 inside. The flow channel 160 formed in the ceramic base 150 is surrounded by a peripheral wall 161. Inside the peripheral wall 161, a first flow control member 164 and a second flow control member 166 are arranged. The first flow control member 164 is positioned to intersect with the straight line connecting the inlet hole 160a and the outlet hole 160b.

Description

Technical Field

[0001] The present invention relates to a base for a substrate holding member and a substrate holding apparatus including the base for a substrate holding member.

Background Art

[0002] Patent Document 1 discloses a mounting table including an electrostatic chuck and a base as a mounting table having a mounting surface for holding a substrate such as a wafer. Inside the base described in Patent Document 1, a spiral flow path is formed as a flow path for a refrigerant. Near a singular point where the flow velocity or the flow direction of the refrigerant in the flow path suddenly changes, protruding members such as a rectifying plate and an elevating screw are provided. Thereby, by correcting the flow of the refrigerant at the singular point, the uniformity of the temperature distribution on the upper surface of the base is enhanced.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In Patent Document 1, by providing a protruding member in the flow path formed inside the base, the flow of the refrigerant is locally changed to enhance the uniformity of the temperature distribution on the upper surface of the base. However, since the refrigerant is configured to flow along a spiral flow path, when the flow velocity and flow rate of the refrigerant are increased to enhance the cooling capacity, the pressure loss between the inlet and outlet of the flow path increases. Therefore, it is necessary to increase the discharge pressure of the refrigerant, and there is a risk that leakage may occur accordingly.

[0005] The present invention has been made in view of such circumstances, and an object thereof is to provide a technique for suppressing pressure loss and reducing the risk of leakage from the flow path.

Means for Solving the Problems

[0006] According to an aspect of the present invention, a plate-shaped member having a first main surface and a second main surface facing the first main surface in the vertical direction, and having a flow path inside, The plate-shaped member is At least two openings that open into the aforementioned flow path, A peripheral wall surrounding the aforementioned flow path and defining the outer edge of the aforementioned flow path, A first flow control member is located inward from the peripheral wall, has a portion that intersects with the straight line connecting the two openings, and defines a part of the flow path, A second flow control member extending from the peripheral wall and defining a part of the flow path, A base for a substrate holding member having the following is provided. [Effects of the Invention]

[0007] According to the above configuration, a first flow control member and a second flow control member are provided inside the flow path. Furthermore, within the flow path, the first flow control member is positioned to intersect with a straight line connecting the two openings. Therefore, when refrigerant flows into the flow path from one opening, the fluid flowing into the flow path strikes the first flow control member and flows through the space between the first and second flow control members towards the other opening. As a result, the fluid flow in the flow path is not a unidirectional flow from one opening to the other, but a two-dimensional flow can be achieved. In this way, the fluid flow in the flow path is two-dimensional, so the pressure loss in the flow path can be reduced compared to when the fluid flow is unidirectional. This makes it possible to suppress the occurrence of leaks from the flow path. [Brief explanation of the drawing]

[0008] [Figure 1] Figure 1 is a perspective view of the electrostatic chuck module 100. [Figure 2] Figure 2 is a schematic diagram illustrating an electrostatic chuck module 100 equipped with a ceramic base 150. [Figure 3] Figure 3 is a schematic diagram illustrating the shape of the electrostatic adsorption electrode 124. [Figure 4] Figures (a) to (d) show the flow of the manufacturing method for the ceramic substrate 110. [Figure 5] Figures (a) to (d) show the flow of another method for manufacturing the ceramic substrate 110. [Figure 6] Figures (a) to (c) show the flow of the manufacturing method for the ceramic base 150. [Figure 7] Figure 7 is a schematic diagram illustrating an electrostatic chuck module 100 comprising a ceramic base 150 having a second flow control member 166. [Figure 8] Figure 8 is a schematic diagram illustrating an electrostatic chuck module 100 comprising a ceramic base 150 having second flow control members 166, 167, and 168. [Figure 9] Figure 9 is a schematic diagram illustrating the ceramic substrate 230 and the porous plate 210. [Figure 10] Figure 10 is a flowchart showing an example of a method for manufacturing a porous plate 210. [Figure 11] Figure 11 is a flowchart showing another example of a method for manufacturing the porous plate 210. [Modes for carrying out the invention]

[0009] <Electrostatic Chuck Module 100> The electrostatic chuck module 100 according to this embodiment will be described with reference to Figures 1 and 2. The electrostatic chuck module 100 according to this embodiment is an example of a substrate holding member for adsorbing and holding a semiconductor wafer such as a silicon wafer (hereinafter simply referred to as wafer 10). In the following description, the vertical direction 5 is defined based on the state in which the electrostatic chuck module 100 is installed for use (the state in Figure 1). As shown in Figure 1, the electrostatic chuck module 100 according to this embodiment mainly comprises a ceramic substrate 110, an electrostatic adsorption electrode 124 (see Figures 2 and 3), and a ceramic base 150. The ceramic substrate 110 is an example of an electrode embedding member of the present invention.

[0010] The ceramic substrate 110 is a circular plate-shaped member with a diameter of 12 inches (approximately 300 mm) and has two main surfaces (upper surface 111 and lower surface 113) facing each other in the vertical direction 5. The upper surface 111 and lower surface 113 of the ceramic substrate 110 correspond to the third main surface and the fourth main surface of the present invention, respectively. The wafer 10 to be held is placed on the upper surface 111 of the ceramic substrate 110. Note that in Figure 1, the wafer 10 and the ceramic substrate 110 are shown separated for clarity. In this embodiment, the ceramic substrate 110 is formed from a ceramic sintered body. For example, the AlN content can be 90% or more. Note that the ceramic substrate 110 does not necessarily have to be formed from AlN ceramics. For example, it may be formed from a ceramic sintered body containing Al2O3 (referred to as Al2O3 ceramics).

[0011] Although not shown in the diagram, the upper surface 111 of the ceramic substrate 110 can be arranged with an annular protrusion on the outer periphery (outer edge) and a plurality of cylindrical protrusions arranged inside the annular protrusion.

[0012] In addition, a gas flow path (not shown) can be formed inside the ceramic substrate 110. The gas flow path can be used as a flow path for supplying gas to the space (gap) defined by the upper surface 111 of the ceramic substrate 110 and the lower surface of the wafer 10. Alternatively, conversely, gas can be exhausted from the space (gap) defined by the upper surface 111 of the ceramic substrate 110 and the lower surface of the wafer 10 through the gas flow path. At this time, by adjusting the exhaust pressure, the differential pressure between the pressure outside the gap and the pressure inside the gap can be adjusted. Thereby, the wafer 10 can be adsorbed toward the upper surface of the ceramic substrate 110.

[0013] As shown in FIG. 2, an electrostatic adsorption electrode 124 is embedded inside the ceramic substrate 110. The electrostatic adsorption electrode 124 can be embedded at a position 0.1 mm to 2.0 mm below from the upper surface 111 of the ceramic substrate 110. That is, the thickness of the ceramic insulating layer from the upper surface 111 of the ceramic substrate 110 to the electrostatic adsorption electrode 124 can be made 0.1 mm or more and 2.0 mm or less.

[0014] As shown in FIG. 3, the electrostatic adsorption electrode 124 is arranged such that two semi-circular electrodes 124a and 124b face each other with a predetermined interval therebetween, and has a substantially circular shape as a whole. In the present embodiment, the outer diameter of the electrostatic adsorption electrode 124 is 292 mm. By applying a predetermined voltage (for example, ±500 V) to the electrodes 124a and 124b respectively, the wafer 10 can be electrostatically adsorbed. In the present embodiment, the ceramic substrate 110 and the electrostatic adsorption electrode 124 constitute an electrostatic chuck (an example of the electrode embedding member of the present invention).

[0015] As shown in FIG. 2, a ceramic base 150 is joined to the lower surface 113 of a ceramic substrate 110 via a joining layer 130. The ceramic base 150 has a circular plate shape with the same diameter as the ceramic substrate 110, and includes two main surfaces (upper surface 155 and lower surface 156) facing each other in the vertical direction 5. Note that the outer diameter of the ceramic base 150 and the outer diameter of the ceramic substrate 110 do not necessarily have to be the same. Note that the upper surface 155 and the lower surface 156 of the ceramic base 150 respectively correspond to the first main surface and the second main surface of the present invention.

[0016] In the present embodiment, the ceramic base 150 is a ceramic sintered body made of a material containing SiC. The thermal conductivity of the material containing SiC is 70 W / mk or more.

[0017] Since SiC ceramics have high resistance to water, water can be flowed as a cooling fluid through a flow path 160 formed inside the ceramic base 150. Water has a high heat transfer rate and is suitable when absorbing a large amount of heat. Also, the cooling fluid does not necessarily have to be water, and may be other refrigerants (liquids or gases). Note that the material containing SiC may be formed by adding sintering aids such as B4C and C in addition to SiC, whereby the sinterability is improved. Also, SiC may be formed by containing metal borides, metal carbides, metal nitrides, etc. For example, as the metal boride, borides of metals in Groups 4 to 6 of the periodic table can be used. Thereby, the linear thermal expansion coefficient can be adjusted, and joining with the ceramic substrate 110 becomes easy. Here, the ceramic base 150 may be made of a material having SiC as a main component. Here, the material having SiC as a main component means a material containing 50 wt% or more of SiC.

[0018] The following are examples of ceramic sintered bodies that constitute the ceramic base 150. For example, a SiC sintered body containing 90 wt% or more of SiC and with added B and C can be used. In this case, the CTE (coefficient of linear thermal expansion) at room temperature can be set to 3.9 ppm / k, and the thermal conductivity at room temperature can be set to 170 W / mk. Alternatively, a SiC sintered body with 47% TiB2 added can be used. In this case, the CTE at room temperature can be set to 4.2 ppm / k, and the thermal conductivity at room temperature can be set to 91 W / mk. Furthermore, a SiC-Ti-Si sintered body can be used. In this case, the CTE at room temperature can be set to 7.6 ppm / k, and the thermal conductivity at room temperature can be set to 75 W / mk.

[0019] As shown in Figure 2, a hollow channel 160 is formed inside the ceramic base 150. As will be described later, the ceramic base 150 is formed by joining an upper plate 151 and a lower plate 152 together (see Figures 6(a) and 6(b)). In Figure 2, for explanatory purposes, the boundary between the upper plate 151 and the lower plate 152 is shown with a solid line.

[0020] The flow path 160 is a closed space formed inside the ceramic base 150 by the joining of the upper plate 151 and the lower plate 152. More specifically, the flow path 160 is a closed space formed by the lower surface 151a of the upper plate 151, the upper surface 152a of the lower plate 152, and the peripheral wall 161. An inlet hole 160a and an outlet hole 160b are opened in the upper surface 152a of the lower plate 152, and the refrigerant that flows into the flow path 160 from the inlet hole 160a flows out from the outlet hole 160b. The peripheral wall 161 is an annular wall provided along the outer edge of the ceramic base 150, and forms the outer edge of the flow path 160 in the circumferential direction. In the following description, the position corresponding to the center of the annular peripheral wall 161 will simply be referred to as the center of the peripheral wall 161. A first flow control member 164 is positioned inside the peripheral wall 161, away from the peripheral wall 161. Furthermore, a second flow control member 166 is positioned in the flow path 160 from the peripheral wall 161.

[0021] The first flow control member 164 and the second flow control member 166 are members that connect the lower surface 151a of the upper plate 151 and the upper surface 152a of the lower plate 152 in the vertical direction. In other words, multiple first flow control members 164 and second flow control members 166 are joined to both the lower surface 151a of the upper plate 151 and the upper surface 152a of the lower plate 152. In this case, since multiple first flow members 164 and second flow control members 166 function as supports, the strength of the ceramic base 150 against vertical loads is improved. Note that the first flow control member 164 and second flow control member 166 do not necessarily have to be joined to both the lower surface 151a of the upper plate 151 and the upper surface 152a of the lower plate 152. At least one of the first flow control member 164 and the second flow control member 166 extends from either the lower surface 151a of the upper plate 151 or the upper surface 152a of the lower plate 152, and a gap may be provided between them and the other.

[0022] In this embodiment, as shown in Figure 2, the first flow control member 164 has eight extending portions 164a that extend radially from the center of the peripheral wall 161 at 45° intervals when viewed from above. The ends of each extending portion 164a that are away from the peripheral wall 161 are in contact with each other. The ends of each extending portion 164a that are closer to the peripheral wall 161 are not in contact with the peripheral wall 161. In other words, the first flow control member 164 is not in contact with the peripheral wall 161.

[0023] In this embodiment, as shown in Figure 2, the eight second flow control members 166 are each positioned between the eight extending portions 164a, and are aligned at approximately 45° intervals in the circumferential direction of the extending peripheral wall 161. One end of each second flow control member 166 is in contact with the peripheral wall 161 and extends from the peripheral wall 161 toward the center of the peripheral wall 161. Note that each second flow control member 166 is in contact with the peripheral wall 161 but not with the first flow control member 164.

[0024] In this embodiment, as shown in Figure 2, there is one inlet hole 160a and one outlet hole 160b. Both the inlet hole 160a and the outlet hole 160b are located between the two extended portions 164a and opposite one end of the second flow control member 166 (the end furthest from the peripheral wall 161). The inlet hole 160a and the outlet hole 160b are located 180° symmetrically with respect to the center of the peripheral wall 161. The number of inlet holes 160a and outlet holes 160b is not limited to one each. For example, by arranging more inlet holes 160a in areas that tend to heat up, such as having two inlet holes 160a and one outlet hole 160b, it is possible to easily design a uniform cooling system for the ceramic base 150.

[0025] In this embodiment, the first flow control member 164 is positioned to intersect with the straight line connecting the inlet hole 160a and the outlet hole 160b. Therefore, the refrigerant flowing into the flow path 160 from the inlet hole 160a is prevented from flowing linearly to the outlet hole 160b. Furthermore, in this embodiment, the inlet hole 160a and the outlet hole 160b are positioned opposite one end of the second flow control member 166 (the end furthest from the peripheral wall 161). Therefore, the refrigerant flowing into the flow path 160 from the inlet hole 160a is divided into two directions upon contact with the end of the second flow control member 166. As a result, the flow of refrigerant in the flow path 160 is not a unidirectional flow from the inlet hole 160a to the outlet hole 160b, but rather a flow in at least two directions can be achieved. Furthermore, as described above, since the first flow control member 164, which is not in contact with the peripheral wall 161, and the second flow control member 166, which extends from the peripheral wall 161, are arranged in an alternating, interwoven manner, the fluid flow can be adjusted so that the fluid flows throughout the entire flow path 160. This makes it easy to design a uniform cooling system for the ceramic base 150.

[0026] In this embodiment, the number of extended portions 164a of the first flow control member 164 and the number of second flow control members 166 are both eight. However, the present invention is not limited to this embodiment, and the number of extended portions 164a of the first flow control member 164 and the number of second flow control members 166 can be set to an appropriate number. For example, the extended portions 164a of the first flow control member 164 and the second flow control members 166 may be arranged at an angle that divides the peripheral wall 161 into n equal parts (where n is a natural number). Furthermore, the total number of inlet holes 160a and outlet holes 160b may be 2k (where k is a natural number less than or equal to n). This makes it possible to suppress pressure loss between the inlet holes 160a and outlet holes 160b in the flow path 160, and to suppress the risk of leakage from the joint surface of the ceramic base 150 (the joint surface between the upper plate 151 and the lower plate 152).

[0027] As described above, the flow path 160, which is an internal space enclosed by the peripheral wall 161, the lower surface 151a of the upper plate 151, and the upper surface 152a of the lower plate 152, is filled with a coolant fluid. Therefore, the flow area S of the flow path 160 is the difference between the area S0 of the upper surface 155, which is the main surface of the ceramic base 150 (hereinafter referred to as the substrate area S0), and the area S1 occupied by the peripheral wall 161, the first flow control member 164, and the second flow control member 166 (S = S0 - S1). By adjusting the number and size of the first flow control member 164 and the second flow control member 166, for example, the ratio of the flow area S to the substrate area S0, S / S0, can be set to a range of 40% to 90%.

[0028] A conductive film 180 is formed on the surface (top surface 155 and side surface 157) of the ceramic base 150. For example, if the conductivity of the ceramic base 150 is insufficient, the ceramic base 150 can be used as a high-frequency electrode by forming a conductive film 180 on its surface.

[0029] Next, the manufacturing method of the electrostatic chuck module 100 will be described. In the following explanation, we will use the case where the ceramic substrate 110 is an AlN ceramic formed from AlN as an example. AlN ceramics are formed from AlN ceramics that contain AlN as the main component. Here, AlN ceramics that contain AlN as the main component refers to a ceramic sintered body containing 50 wt% or more of AlN. Also, for the sake of simplicity, we will assume that only the electrostatic adsorption electrode 124 is embedded in the ceramic substrate 110.

[0030] First, the method for manufacturing the ceramic substrate 110 will be described. As shown in Figure 4(a), granulated powder P, mainly composed of AlN powder, is placed in a carbon bed mold 601 and pre-pressed with a punch 602. Preferably, the granulated powder P contains 7 wt% or less of a sintering aid (for example, Y2O3). Next, as shown in Figure 4(b), electrodes 124 cut to a predetermined shape are placed on top of the pre-pressed granulated powder P. The electrodes 124 are positioned parallel to the surface perpendicular to the pressing direction (the bottom surface of the bed mold 601). At this time, W pellets or Mo pellets may be embedded at the terminal 124T (see Figure 3) of the electrode 124.

[0031] As shown in Figure 4(c), granulated powder P is further added to the bed mold 601 so as to cover the electrostatic adsorption electrode 124, and then pressed and molded with a punch 602. At this time, the amount of granulated powder P covering the electrostatic adsorption electrode 124 can be adjusted so that the electrostatic adsorption electrode 124 is embedded at a depth of 0.3 mm or more. Next, as shown in Figure 4(d), the granulated powder P with the embedded electrostatic adsorption electrode 124 is fired while pressed. The pressure applied during firing is preferably 1 MPa or more. It is also preferable to fire at a temperature of 1800°C or higher. Note that firing can be performed multiple times. Next, blind holes are machined up to the electrostatic adsorption electrode 124 in order to form the terminal 124T (see Figure 3). Note that if pellets are embedded, blind holes should be machined up to the pellets. Furthermore, if necessary, through holes may be formed to form gas flow channels. In this case, a ceramic substrate 110 with gas flow channels formed inside can be manufactured.

[0032] The upper surface 111 and lower surface 113 of the ceramic substrate 110 formed in this manner are ground, and further polishing is performed as needed. At this time, the vertical distance 5 from the upper surface 111 of the ceramic substrate 110 to the electrostatic adsorption electrode 124 can be adjusted. Furthermore, by sandblasting the upper surface 111, multiple protrusions and annular protrusions can be formed on the upper surface 111. Sandblasting is preferred as the processing method for forming multiple protrusions and annular protrusions, but other processing methods can also be used.

[0033] The ceramic substrate 110 can also be manufactured by the following method. A sintering aid (Y2O3) of 7 wt% or less is added to the AlN raw material powder as needed. Metal carbides, metal nitrides, and metal borides can also be added to the AlN raw material powder. Next, a binder is added to the AlN raw material powder, mixed in ethanol, dried, and then granulated to produce a granulated powder containing AlN as a component. Next, the granulated powder is filled into a rubber mold and isotropically pressure-molded (CIP molding) using water pressure to produce two plate-shaped CIP molded bodies 610 (see Figure 5(a)). For example, CIP molding can be performed at a pressure of 130 MPa. Next, the CIP molded bodies 610 are degreased to remove the binder (see Figure 5(b)). Next, as shown in Figure 5(c), a recess 611 for embedding an electrostatic adsorption electrode 124 is formed in one of the degreased CIP molded bodies 610. The recess 611 may be formed in the CIP molded body 610 before the degreasing treatment. After placing the electrostatic adsorption electrode 124 in the recess 611 of the CIP molded body 610, another CIP molded body 610 is stacked on top. At this time, the thickness of the other CIP molded body 610 can be adjusted so that the electrostatic adsorption electrode 124 is embedded at a depth of 0.3 mm or more. Next, as shown in Figure 5(d), the CIP molded bodies 610 stacked with the electrostatic adsorption electrode 124 sandwiched in between are fired in a pressed state to produce a fired body. The pressure applied during firing is preferably 1 MPa or more. It is also preferable to fire at a temperature of 1800°C or higher. Multiple firings can also be performed. The process after producing the fired body is the same as the process described above, so the explanation is omitted.

[0034] Next, the manufacturing method for the ceramic base 150 will be described. As shown in Figure 6(a), the case in which two CIP molded bodies 150A and 150B are stacked will be explained as an example. In the same manner as the manufacturing method for the ceramic substrate 110 described above, multiple plate-shaped CIP molded bodies 150A and 150B are prepared and degreased.

[0035] Next, the degreased CIP molded body 150A is subjected to external shaping to form grooves 260 that will later become flow channels 160. Specifically, as shown in Figure 6(a), grooves 260 are formed on the upper surface of the CIP molded body 150A. This forms protrusions on the upper surface of the CIP molded body 150A that will later become the first flow control member 164, the second flow control member 166, and the peripheral wall 161. Then, by stacking another plate-shaped CIP molded body 150B on top of the upper surface of the CIP molded body 150A with grooves 260 formed thereon, flow channels 160 are formed inside the laminate of CIP molded bodies 150A and 150B (see Figure 6(b)). Alternatively, grooves 260 may be formed on the lower surface of the CIP molded body 150B, and the CIP molded body 150A may be stacked on top from below to form flow channels 160 inside the laminate of CIP molded bodies 150A and 150B. Alternatively, grooves 260 may be formed on the upper surface of CIP molded body 150A and the lower surface of CIP molded body 150B, respectively, and by stacking them, a flow channel 160 may be formed inside the laminate of CIP molded bodies 150A and 150B.

[0036] Next, the laminate of CIP molded bodies 150A and 150B is fired while pressed to produce a fired body. The pressure applied during firing is preferably 1 MPa or more. It is also preferable to fire at a temperature of 1800°C or higher. The firing time is preferably 2 hours or more. It is also possible to perform firing multiple times. When firing the laminate of CIP molded bodies 150A and 150B while pressed in this manner, the CIP molded bodies 150A and 150B can be joined simultaneously by diffusion bonding. As shown in Figure 6(c), a conductive film 180 can be formed on the upper surface 155 and side surface 157 of the ceramic base 150.

[0037] Furthermore, the CIP molded bodies 150A and 150B do not necessarily have to be joined by diffusion bonding at the same time during firing. For example, after firing the CIP molded bodies 150A and 150B, the ceramic sintered bodies can be mirror-polished, the polished surfaces can be joined together, and then the joint surface can be heated to near the firing temperature of the ceramic sintered body while applying pressure to the joint surface for heat treatment. For example, in the case of SiC ceramics, diffusion bonding is possible with a surface roughness of Ra 0.2 μm or less, a joining pressure of 1 MPa or more, and a joining temperature of 1800°C or higher. Also, after firing the CIP molded bodies 150A and 150B, it is not always necessary to join them by diffusion bonding. For example, after firing the CIP molded bodies 150A and 150B, the ceramic sintered bodies can be joined by hard brazing. Hard brazing is a brazing material having a melting point (including liquidus temperature and glass transition temperature) of 450°C or higher. Two ceramic sintered bodies can be joined by heating them to a temperature of 450°C or higher with a hard brazing material interposed between them, using either fluidized or molten hard brazing material. At this time, the joining is performed under a load of 0.001 MPa or higher. The atmosphere can be appropriately selected from air, inert gas, or vacuum. Examples of hard brazing materials include silver brazing (BAg-8, melting point 780°C), aluminum brazing (A4047, eutectic point 577°C), gold brazing (BAu-4, liquidus temperature 950°C), and nickel brazing (BNi-2, liquidus temperature 1000°C). These hard brazing materials may also contain active metals such as Ti, Hf, and Zr. Metal foils such as Al foil can also be used as hard brazing material. When joining ceramic sintered bodies by hard brazing, it is preferable to have a centerline average roughness Ra of the joining surface before joining at 1.6 μm or less.

[0038] Alternatively, two ceramic sintered bodies can be joined by soft soldering. Soft solder is a brazing material having a melting point (including liquidus temperature and glass transition temperature) of less than 450°C. By heating the two ceramic sintered bodies to a temperature of less than 450°C with the soft solder interposed between them, the two ceramic sintered bodies can be joined by the fluidized or molten soft solder. At this time, the joining is performed under a load of surface pressure of 0.001 MPa or more. The atmosphere can be appropriately selected from air, inert gas, or vacuum. In addition, for example, In, Sn, Pb, and their alloys (solder) can be used as the soft solder. When joining ceramic sintered bodies by soft soldering, it is preferable to make the centerline average roughness Ra of the joining surface 1.6 μm or less before joining.

[0039] The joining method for ceramic fired bodies (or CIP molded bodies 150A, 150B) is not limited to the diffusion bonding, hard brazing, and soft brazing described above, but can also be used, as is known, such as anodic bonding. When ceramic fired bodies are joined using joining materials such as hard brazing or soft brazing materials, joining can be done at a lower temperature compared to joining by diffusion bonding, thus making manufacturing easier.

[0040] Furthermore, after firing the CIP molded bodies 150A and 150B, the two fired ceramic bodies can be joined using the following joining material. For example, if the two fired ceramic bodies are SiC ceramics, a joining material containing 20-80 wt% SiC and the remainder being an Al alloy (1-15 wt% Si, 1-10 wt% Mg) can be used. If the two fired ceramic bodies are AlN ceramics, a joining material containing 20-80 wt% AlN and the remainder being an Al alloy (1-15 wt% Si, 1-10 wt% Mg) can be used. If the two fired ceramic bodies are Al2O3 ceramics, a joining material containing 20-80 wt% Al2O3 and the remainder being an Al alloy (1-15 wt% Si, 1-10 wt% Mg) can be used. The joining material can be used in the form of a paste, foil, or thin sheet. The bonding paste can be formed by adding a solvent (e.g., alcohol, water, organic solvent) and a thickener (e.g., carboxymethylcellulose, polyvinyl alcohol, xanthan gum, etc.) to filler and metal powder, and then kneading the mixture. Foils and sheets can be formed by dissolving the bonding material and rolling it.

[0041] It should be noted that the groove 260 does not necessarily need to be formed in the CIP molded body. For example, before firing the CIP molded body, a pre-fired body may be prepared by pre-firing it at a temperature lower than the firing temperature (for example, 500°C to 100°C lower than the firing temperature). Then, the pre-fired body can be processed to form the groove 260. Alternatively, the groove 260 can be formed in the CIP molded body, and after pre-firing the CIP molded body with the groove 260 formed in it, the pre-fired body can be further processed so that the groove 260 reaches the desired dimensions.

[0042] The ceramic substrate 110 and the ceramic base 150, thus prepared, can be joined via a bonding layer 130. For example, the bonding layer 130 can be joined using hard brazing with the hard brazing material described above. Alternatively, the bonding layer 130 can be joined using soft brazing with the soft brazing material described above. As described above, when the ceramic substrate 110 and the ceramic base 150 are joined by hard brazing, it is preferable that the centerline average roughness Ra of the bonding surface before joining be 1.6 μm or less. Also, when the ceramic substrate 110 and the ceramic base 150 are joined by the soft brazing described above, it is preferable that the centerline average roughness Ra of the bonding surface before joining be 1.6 μm or less. Note that the ceramic substrate 110 and the ceramic base 150 do not necessarily have to be joined via a bonding layer 130; for example, they may be joined by diffusion bonding.

[0043] <Effects of the Embodiment> In the above embodiment, the electrostatic chuck module 100 comprises a plate-shaped ceramic substrate 110 having electrodes 124 embedded inside, a plate-shaped ceramic base 150 joined to the lower surface 113 of the ceramic substrate 110, and a bonding layer 130 that joins the ceramic substrate 110 and the ceramic base 150.

[0044] The ceramic base 150 has two main surfaces (upper surface 155 and lower surface 156) and has a flow channel 160 inside. The flow channel 160 formed in the ceramic base 150 is surrounded by a peripheral wall 161. In other words, the peripheral wall 161 forms the outer edge of the flow channel 160 in the circumferential direction. A first flow control member 164 is positioned inside the peripheral wall 161, at a distance from the peripheral wall 161.

[0045] The flow path 160 in this embodiment is not a flow path with a predetermined width extending in one direction, as used in the prior art, but is configured so that the refrigerant fluid flows through an internal space enclosed by the peripheral wall 161, the lower surface 151a of the upper plate 151, and the upper surface 152a of the lower plate 152. Furthermore, within the flow path 160, the first flow control member 164 is positioned to intersect with the straight line connecting the inlet hole 160a and the outlet hole 160b. Therefore, the refrigerant flowing into the flow path 160 from the inlet hole 160a strikes the first flow control member 164 and flows towards the outlet hole 160b, passing between the first flow control member 164 and the second flow control member 166. As a result, the flow of refrigerant in the flow path 160 is not a unidirectional flow from the inlet hole 160a to the outlet hole 160b, but a two-dimensional flow can be realized. Thus, since the fluid flow within the channel 160 is two-dimensional, the pressure loss in the channel 160 can be reduced compared to the case where the fluid flow is limited to one direction. This suppresses the occurrence of leaks from the channel 160.

[0046] Furthermore, as described above, by adjusting the number and size of the multiple first flow control members 164, the ratio of the flow area S to the substrate area S0, for example, S / S0, can be set to a range of 40% to 90%. In this way, the ratio of the flow area S to the substrate area S0, S / S0, can be increased, enabling the transfer of a large amount of heat (heat dissipation).

[0047] In the above embodiment, the ceramic base 150 was formed from a material containing SiC. As mentioned above, the thermal conductivity of the SiC-containing material is 70 W / mk or higher, and the SiC-containing material has high resistance to water. Therefore, when the ceramic base 150 is formed from a material containing SiC, water can be flowed through the channel 160 formed inside as a cooling fluid. Water has a high heat transfer coefficient and is suitable for absorbing large amounts of heat.

[0048] In the above embodiment, the first flow control member 164 has a shape that is rotationally symmetric with respect to the center of the peripheral wall 161, that is, the center of the upper surface 155 of the ceramic base 150. This makes it possible to improve the uniformity of the radial temperature distribution on the upper surface 155 of the ceramic base 150.

[0049] <Change form> The embodiments described above are merely illustrative and may be modified as appropriate. In the above embodiments and modified forms, the ceramic substrate 110 does not necessarily have to be formed of AlN ceramics, but may be another ceramic sintered body (e.g., alumina) or a composite material of ceramics and metal. Similarly, the ceramic base 150 does not necessarily have to be formed of a material containing SiC, but may be another ceramic sintered body or a composite material of ceramics and metal. For example, the ceramic base 150 may be formed of a ceramic sintered body containing 90 wt% or more of Al2O3. For example, a ceramic sintered body containing 99.5 wt% or more of Al2O3 and the remainder being sintering aid components such as SiO2, CaO, and MgO, or a ceramic sintered body containing 92 wt% or more of Al2O3 and the remainder being sintering aid components such as SiO2, CaO, and MgO can be used.

[0050] The electrostatic chuck module 100 does not necessarily have a bonding layer 130 that joins the ceramic substrate 110 and the ceramic base 150. The ceramic substrate 110 and the ceramic base 150 may be directly joined without the bonding layer 130. Alternatively, the ceramic substrate 110 and the ceramic base 150 may be mechanically joined using screws or the like. In this case, it is preferable to form a groove in one of the ceramic substrate 110 or the ceramic base 150 and fasten them together by sandwiching an airtight seal such as an O-ring in the groove.

[0051] Furthermore, the shape and thickness of the ceramic substrate 110 and the ceramic base 150 can be changed as appropriate. Also, the shape of the channel 160 formed inside the ceramic base 150 can be changed as appropriate.

[0052] The number, position, and shape of the first flow control members 164 and second flow control members 166 provided in the flow path 160 of the ceramic base 150 can be changed as appropriate. For example, in the above embodiment, the first flow control member 164 had eight linear extensions 164a that extended radially from the center of the peripheral wall 161 at 45° intervals when viewed from above. The eight second flow control members 166 were each positioned between the eight extensions 164a in the circumferential direction of the peripheral wall 161, and were arranged at approximately 45° intervals. However, the present invention is not limited to such embodiments. For example, as shown in Figure 7, the first flow control member 264 may have eight linear extensions 264a that extend radially from the center of the peripheral wall 161 at 45° intervals when viewed from above, and branching portions 264b that branch in two directions at the end of each extension 264a closer to the peripheral wall 161. Furthermore, the second flow control member 266 may comprise a first portion 266a extending from the peripheral wall 161 toward the center of the peripheral wall 161, and a second portion 266b extending in a direction perpendicular to the first portion 266a at approximately the center of the first portion 266a. In the example shown in Figure 7, eight second flow control members 266 are arranged between eight extending portions 264a, and are aligned at approximately 45° intervals in the circumferential direction of the peripheral wall 161.

[0053] In the above embodiment, the first flow control member 164 and the second flow control member 166 were formed by linearly extending plate-shaped walls. The present invention is not limited to such embodiments. For example, as shown in Figure 8, the flow path 160 may have a first flow control member 364 and a second flow control member 366 formed by curvedly extending plate-shaped walls. In Figure 8, for the sake of simplicity, the first flow control member 364 is represented by a solid line and the second flow control member 366 is represented by a dotted line. In the example shown in Figure 8, the first flow control member 364 has six spiral-shaped extensions 364a that, when viewed from above, curve circumferentially as they extend radially outward from the center of the peripheral wall 161. The ends of each extension 364a on the center side are in contact with each other. Furthermore, six second flow control members 366 are arranged between the six extensions 364a. One end of each second flow control member 366 is in contact with the peripheral wall 161 and extends spirally from the peripheral wall 161 toward the center of the peripheral wall 161. In this way, the first flow control member 364, which is not in contact with the peripheral wall 161, and the second flow control members 366, which extend from the peripheral wall 161, are arranged alternately in an interwoven manner, so that the fluid flow can be adjusted so that the fluid flows throughout the entire flow path 160. This makes it easy to design a uniform cooling system for the ceramic base 150. Furthermore, since the first flow control member 364 and the second flow control member 366 extend spirally, the uniformity of the circumferential temperature distribution on the upper surface 155 of the ceramic base 150 can be improved by flowing a refrigerant through the spiral-shaped flow path formed by the first flow control member 364 and the second flow control member 366.

[0054] Furthermore, in the above embodiments and modified examples, the inlet hole 160a and the outlet hole 160b were both provided at positions facing the ends of the second flow control members 166, 266, and 366. However, the present invention is not limited to such embodiments, and for example, they can also be arranged at positions facing the ends of the first flow control members 164, 264, and 364. The inlet hole 160a and the outlet hole 160b can be arranged at any appropriate position as long as the straight line connecting them intersects with the first flow control members 166, 266, and 366.

[0055] In the above embodiments and modifications, a ceramic substrate 110, which functions as an electrostatic chuck, was bonded to the upper surface 151 of the ceramic base 150. The present invention is not limited to such embodiments. Among the electrostatic chuck modules 100 described above, the ceramic base 150 can be used as a ceramic susceptor (ceramic base) for mounting a substrate holding member (electrostatic chuck) for semiconductor manufacturing.

[0056] As shown in Figure 9, a ceramic substrate 230 and a porous plate 210, which function as a vacuum chuck, can be placed on the upper surface 155 of the ceramic base 150. The ceramic substrate 230 and the porous plate 210 correspond to the substrate mounting members of the present invention.

[0057] The porous plate 210 is a porous material having a circular, plate-like shape. The porous plate 210 is an example of the "porous ceramic material" of the present invention. In this specification, a porous material means that the porosity is 20 volume% or more. A wafer 10 (see Figure 1) is placed on the upper surface 211 of the porous plate 210. There are no particular restrictions on the size of the porous plate 210, but for use as a substrate holding member 100 for semiconductor manufacturing equipment, it is preferable that the diameter of the porous plate 210 is about the same as the diameter of the wafer 10. For example, the diameter of the porous plate 210 can be 150 mm or more. Also, the length (hereinafter referred to as thickness) of the porous plate 210 in the vertical direction 5 can be 5 mm to 50 mm.

[0058] The porous plate 210 can be a porous ceramic material containing SiC as the main component. Containing SiC as the main component means containing 50 wt% or more of SiC. The porous ceramic material includes porous ceramic sintered bodies in which ceramic particles are bonded together by necking, and porous materials formed by ceramic particles bonded together by a binder such as glass. The porosity of the porous plate 210 is 20 to 60 volume%. In the porous plate 210, the SiC ceramic particles constituting it are bonded in contact with each other. The thermal conductivity of the porous plate 210 is preferably 10 W / (m·K) or higher, more preferably 30 W / (m·K) or higher, and more preferably 50 W / (m·K) or higher. For example, the thermal conductivity of the porous plate 210 can be 30 to 100 W / (m·K).

[0059] Furthermore, the porous plate 210 is not necessarily limited to a porous ceramic material mainly composed of SiC, but may be a porous ceramic material other than SiC. Also, the porous plate 210 is not necessarily a porous ceramic sintered body in which ceramic particles are bonded together by necking, but may be a porous material formed by ceramic particles bonded together with a binder such as glass, as described above.

[0060] <Ceramic base material 230> The ceramic substrate 230 is a dense material having a circular, plate-like shape. In this specification, a dense material means that its porosity is 5 volume% or less. The ceramic substrate 230 is an example of a support member of the present invention. The ceramic substrate 230 can be formed from ceramics such as SiC ceramics or AlN ceramics. As shown in Figure 9, the ceramic substrate 230 is joined to the upper surface of the base 150. The ceramic substrate 230 has a plate-like base portion 238 and an outer peripheral wall portion 239 that protrudes upward from the outer periphery of the base portion 238. The base portion 238 and the outer peripheral wall portion 239 form a circular recess 233 with approximately the same diameter as the porous material plate 210 approximately in the center of the upper surface 231 of the ceramic substrate 230. The porous material plate 210 is placed on the bottom surface 233b of the recess 233. The side surface of the recess 233 is formed by the outer peripheral wall portion 239, and the bottom surface 233b of the recess 233 is formed by the base portion 238. Since the depth of the recess 233 is approximately the same as the thickness of the porous plate 210, the top surface 231 of the ceramic substrate 230 and the top surface 211 of the porous plate 210 are flush. When the recess 233 is formed on the top surface 231 of the ceramic substrate 230, the bottom surface 233b of the recess 233 corresponds to the main surface of the ceramic substrate 230. As shown in Figure 9, a sealing material 240 such as glass, resin, or a ceramic layer can be filled between the outer peripheral wall portion 239 and the side surface of the porous plate 210. This improves the airtightness of the side surface of the porous plate 210 and suppresses the suction of gas from between the outer peripheral wall portion 239 and the side surface of the porous plate 210 when the wafer 10 is vacuum adsorbed. This improves the adsorption force of the wafer 10.

[0061] As shown in Figure 9, multiple grooves 234 are formed in the recess 233 of the ceramic substrate 230. Although not shown in the figure, the multiple grooves 234 are formed to form concentric circles when viewed from above. However, the number and shape of the grooves 234 are not limited to this configuration and can be arbitrary. For example, one groove 234 may be arranged to form a roughly spiral shape when viewed from above.

[0062] A connecting passage 256 is located below each groove 234 of the ceramic substrate 230. The connecting passage 256 is a substantially cylindrical void formed inside the ceramic substrate 230 and has a size that overlaps with the multiple grooves 234 when viewed from above. An opening is formed in the lower wall of each groove 234 that connects to the connecting passage 256. The multiple grooves 234 communicate with each other via the connecting passage 256. Furthermore, an exhaust passage 258 extending downward is located at the lower end of the connecting passage 256. A depressurizing pump (not shown) is provided in the exhaust passage 258. Since the porous plate 210 is made of a porous material, it has a porosity of 20 volume% or more. Therefore, gas can be drawn in through the exhaust passage 258, the connecting passage 256, the grooves 234, and the porous plate 210. This allows the wafer 10 placed on the upper surface 211 of the porous plate 210 to be vacuum-adsorbed.

[0063] Next, the manufacturing method for the porous plate 210 will be described. Note that the manufacturing method for the ceramic substrate 230 is the same as that for the ceramic base 150, so the explanation will be omitted.

[0064] The porous plate 210 can be manufactured as follows (see Figure 10). A SiC material powder mainly composed of SiC, a pore-forming material, and a binder are mixed in a dry process (S101). As the SiC material powder, for example, #100 material powder, #600 material powder, a mixture of #100 material powder and #600 material powder, or a mixture of #180 material powder and #800 pore-forming material powder can be used. The average particle size of the #100 material powder is approximately 125 μm to 149 μm, the average particle size of the #180 material powder is approximately 53 μm to 90 μm, the average particle size of the #600 material powder is approximately 20 μm to 30 μm, and the average particle size of the #800 material powder is approximately 12 μm to 15 μm. As the pore-forming material, polymers, carbon, and other materials that sublimate when fired can be used. The size of the pore-forming material is approximately 5 μm to 100 μm, and various shapes such as spherical and rectangular can be used. The amount of pore-forming material can be adjusted so that the porosity of the porous body of the SiC ceramic sintered body after firing is 20 to 60 volume percent. Next, after debinding treatment, firing is performed at a temperature of 1700°C to 2200°C in a non-oxidizing atmosphere such as an Ar atmosphere (S102). This forms a porous body of SiC ceramic sintered body with SiC as the main component. Next, the porous body of the SiC ceramic sintered body is subjected to external processing such as surface grinding, machining, and lapping to process it to a predetermined size and shape (S103).

[0065] Alternatively, the porous plate 210 can be formed as follows (see Figure 11). Mix several types of SiC material powder with different particle sizes (S201). For example, mix SiC material powder containing coarse particles (particle size 10-200 μm) with SiC material powder containing fine particles (particle size 1-50 μm). Alternatively, prepare one type of SiC material powder (S201). When mixing several types of SiC material powder with different particle sizes, the process may be dry or wet. Similarly, when using one type of SiC material powder, the process may be dry or wet. Next, fire at a temperature at which necking between particles progresses (S202). In this case as well, it is preferable to fire in a non-oxidizing atmosphere such as an Ar atmosphere. Next, perform external shaping such as surface grinding, machining, and lapping on the porous body of the SiC ceramic sintered body to process it to a predetermined size and shape (S203). In this way, a porous SiC ceramic sintered body can be formed by mixing multiple types of SiC material powders with different particle sizes, or by using one type of SiC material powder and firing it at a temperature that promotes necking between particles. When multiple types of SiC material powders with different particle sizes are mixed, the porosity of the porous SiC ceramic sintered body after firing can be adjusted to 20-60 volume% by adjusting the particle size and mixing ratio of the multiple types of SiC material powders, firing time, and firing temperature. Similarly, when using one type of SiC material powder, the porosity of the porous SiC ceramic sintered body after firing can be adjusted to 20-60 volume% by adjusting the particle size, firing time, and firing temperature of the SiC material powder.

[0066] The thermal conductivity of the porous plate 210 manufactured using these manufacturing methods is 10 to 100 W / (m·K). The thermal conductivity can be calculated from the specific heat and thermal diffusivity. Preferably, the thermal conductivity of the porous plate 210 is 50 W / (m·K) or higher. As described above, by manufacturing the porous plate 210 using a SiC ceramic sintered body, a porous plate 210 having a thermal conductivity of 50 W / (m·K) or higher can be easily manufactured.

[0067] Next, a method for joining the ceramic substrate 230 and the porous plate 210 will be described. The ceramic substrate 230 and the porous plate 210 can be joined by diffusion bonding at a temperature of 1500°C or higher and a pressure of 1 MPa or higher while they are stacked. In this case, it is not necessary to interpose a bonding material between the ceramic substrate 230 and the porous plate 210. Note that the absence of a bonding material means that the bonding layer cannot be seen at the optical microscope level (magnification of about 500x). As a result, there is no risk of heat conduction being hindered at the interface between the ceramic substrate 230 and the porous plate 210.

[0068] When the ceramic substrate 230 and the porous plate 210 are joined by diffusion bonding, a gap may occur between the outer peripheral wall portion 239 and the side surface of the porous plate 210. This gap can be filled with powder such as alumina or SiC and fired during diffusion bonding to form a ceramic layer. Alternatively, after joining the ceramic substrate 230 and the porous plate 210, a glass sealant, resin sealant, etc., can be filled in.

[0069] Furthermore, the ceramic substrate 230 and the porous plate 210 can be joined using a metal foil such as Al foil. For example, a metal foil mesh can be used as the metal foil. In this case as well, there is no risk of heat conduction being hindered at the interface between the ceramic substrate 230 and the porous plate 210. The ceramic substrate 230 and the porous plate 210 can be joined by heating with a metal foil mesh interposed between them. At this time, the heat joining is performed under a load of surface pressure of 0.001 MPa or more. The atmosphere can be appropriately selected from air, inert gas, or vacuum. When joining using a metal foil such as Al foil, joining can be performed at a lower temperature compared to direct joining (diffusion joining) without using metal foil. This makes it possible to suppress the stress remaining after joining and suppress warping of the substrate holding member 100 after joining. In addition, by using metal foil as the joining layer, a stress buffering effect due to the plastic deformation of the metal foil at the joining surface can also be expected. This further suppresses the stress remaining after joining, and thus suppresses warping of the substrate holding member 100 after joining.

[0070] Although embodiments and modified versions of the invention have been described above, the technical scope of the present invention is not limited to the scope described above. It will be obvious to those skilled in the art that various modifications or improvements can be made to the above embodiments. It is also clear from the claims that such modified or improved forms may be included in the technical scope of the present invention.

[0071] The order in which each process in the manufacturing method shown in the specification and drawings is executed is not specifically defined, and unless the output of a previous process is used in a later process, the processes can be executed in any order. Even if phrases such as "first," and "next," are used for convenience, this does not mean that the processes must be performed in that order. [Explanation of Symbols]

[0072] 100 Electrostatic Chuck Modules 110 Ceramic substrate 124 Electrodes for electrostatic adsorption 130 Bonding layer 150 Ceramic Base

Claims

1. The plate-shaped member has a first main surface and a second main surface that faces the first main surface in the vertical direction, and has a flow path inside, The aforementioned plate-shaped member is At least two openings that open into the aforementioned flow path, A peripheral wall surrounding the aforementioned flow path and defining the outer edge of the aforementioned flow path, A first flow control member is located inward from the peripheral wall, has a portion that intersects with the straight line connecting the two openings, and defines a part of the flow path, A second flow control member extending from the peripheral wall and defining a part of the flow path, A base for a substrate holding member having the following features.

2. The at least two openings include a first opening and a second opening, The base for a substrate holding member according to claim 1, wherein the end of the first flow control member or the end of the second flow control member is arranged facing the first opening.

3. The base for a substrate holding member according to claim 1, wherein the first flow control member has a shape that is rotationally symmetric with respect to the center of the first main surface.

4. The base for a substrate holding member according to claim 1, wherein the first flow control member has a spiral shape when viewed from above.

5. The base for a substrate holding member according to claim 1, wherein the plate-shaped member is a ceramic base formed of ceramics.

6. The first flow control member and the second flow control member are arranged at angles (where n is a natural number) that divide the peripheral wall into 2n equal parts. The base for a substrate holding member according to claim 1, wherein the number of openings is 2k (where k is a natural number less than or equal to n).

7. A base for a substrate holding member according to any one of claims 1 to 6, A substrate holding device comprising: a substrate holding member joined to the first main surface, the substrate holding member having a mounting surface on which a substrate is placed.

8. The substrate holding member is A ceramic substrate formed from ceramics, The substrate holding device according to claim 7, further comprising an electrode embedded inside the ceramic substrate.

9. The substrate holding device according to claim 7, wherein the substrate holding member comprises a porous body having a porosity of 20 to 60 volume percent.

10. The substrate holding device according to claim 9, wherein the porous body is a porous body of a ceramic sintered body in which ceramic particles are bonded together by necking.

11. The substrate holding device according to claim 9, wherein the porous material has a thermal conductivity of 50 W / (m·K) or more and is mainly composed of SiC.

12. The substrate holding member is A support member made of a dense ceramic material with a recess formed therein, The substrate holding device according to claim 9, comprising the porous body disposed in the recess and constituting the aforementioned mounting surface.

Images