Wafer holding table

Abstract

To provide a wafer holding table that can cool a wafer more efficiently.SOLUTION: The wafer holding table includes a plate having an upper surface on which a wafer is placed. The plate includes a refrigerant flow passage through which a refrigerant flows inside the plate. The refrigerant flow passage has an inlet through which the refrigerant is introduced, an outlet through which the refrigerant is discharged, and a parallel flow passage connecting the inlet and the outlet. The parallel flow passage has a plurality of flow passages arranged in parallel within a surface parallel to the upper surface.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present disclosure relates to a wafer chucking stage.

Background Art

[0002] Patent Document 1 discloses a wafer chucking stage. The wafer chucking stage is used, for example, in a prober. A prober is an inspection apparatus that measures the electrical performance of each chip before cutting a semiconductor wafer on which a plurality of circuits are formed into individual chips. The wafer chucking stage disclosed in Patent Document 1 includes a substrate that is a metal disk. This substrate has an upper surface on which the wafer is placed and a refrigerant flow path provided inside the substrate. When refrigerant flows through this flow path, the wafer placed on the upper surface of the substrate is cooled.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] The wafer chucking stage of Patent Document 1 efficiently cools the wafer by setting the shape and dimensions of the cross-section of the flow path within a specific range. Considering shortening the time required for the inspection process and the like, more efficient cooling of the wafer is desired.

[0005] One object of the present disclosure is to provide a wafer chucking stage capable of cooling a wafer more efficiently.

Means for Solving the Problems

[0006] The wafer holding stage of the present disclosure includes a plate having an upper surface on which a wafer is placed. The plate includes a refrigerant flow path through which refrigerant flows inside the plate. The refrigerant flow path has an inlet into which the refrigerant is introduced, an outlet from which the refrigerant is discharged, and parallel flow paths connecting between the inlet and the outlet. The parallel flow paths have a plurality of flow paths arranged in parallel in a plane parallel to the upper surface.

Advantages of the Invention

[0007] The wafer holding stage of the present disclosure can cool the wafer more efficiently.

Brief Description of the Drawings

[0008]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

DETAILED DESCRIPTION OF THE INVENTION

[0009] [Description of Embodiments of the Present Disclosure] First, embodiments of the present disclosure will be listed and described.

[0010] (1) The wafer holding stage of the present disclosure includes a plate having an upper surface on which the wafer is placed. The plate includes a refrigerant flow path through which refrigerant flows inside the plate. The refrigerant flow path has an inlet through which the refrigerant is introduced, an outlet through which the refrigerant is discharged, and parallel flow paths connecting between the inlet and the outlet. The parallel flow paths have a plurality of flow paths arranged in parallel in a plane parallel to the upper surface.

[0011] The wafer holding stage of the present disclosure can cool the wafer more efficiently. The refrigerant flow path having parallel flow paths has a larger contact area than a refrigerant flow path composed of a single flow path. Here, the single flow path means a virtual refrigerant flow path described later. The larger the contact area of the parallel flow paths, the easier the heat of the plate is transferred to the refrigerant. Since the cooling efficiency by the refrigerant is increased, the cooling capacity of the plate is improved.

[0012] (2) In the wafer holding stage of (1) above, when the hydraulic diameter of the virtual refrigerant flow path is 1, the ratio of the hydraulic diameter of the parallel flow path to the hydraulic diameter of the virtual refrigerant flow path may be 0.2 or more. The virtual refrigerant flow path has a cross section equivalent to a cross section including the entire parallel flow path.

[0013] The wafer holding stage of (2) above can effectively suppress an increase in the pressure loss of the parallel flow paths.

[0014] (3) In the wafer holding stage of the above (1) or (2), the number of the plurality of flow paths may be two or more.

[0015] The wafer holding stage of the above (3) is likely to suppress an increase in the pressure loss of the parallel flow path with respect to the pressure loss of the virtual refrigerant flow path described later.

[0016] (4) In the wafer holding stage of any one of the above (1) to (3), the plate has a first pipe connected to the inlet of the refrigerant flow path, and the upper surface of the inner peripheral surface of the first pipe and the upper surface of the parallel flow path are flush, or the upper surface of the parallel flow path may be lower than the upper surface of the inner peripheral surface of the first pipe.

[0017] The wafer holding stage of the above (4) has high cooling efficiency by the refrigerant. Since the upper surface of the parallel flow path is close to the upper surface of the plate, it greatly contributes to heat transfer from the wafer to the refrigerant. When the upper surface of the parallel flow path is flush with the upper surface of the inner peripheral surface of the first pipe, or lower than the upper surface of the inner peripheral surface of the first pipe, bubbles are less likely to remain in the flow path compared to the case where the upper surface of the parallel flow path is higher than the upper surface of the inner peripheral surface of the first pipe. Therefore, the cooling efficiency by the refrigerant is increased. Among them, when the upper surface of the parallel flow path is flush with the upper surface of the inner peripheral surface of the first pipe, the refrigerant easily flows smoothly near the upper surface of the flow path that contributes to heat transfer to the refrigerant. Therefore, the cooling efficiency by the refrigerant becomes higher.

[0018] (5) In the wafer holding stage of any one of the above (1) to (4), the plate has a first pipe connected to the inlet of the refrigerant flow path, and the cross-sectional area of the parallel flow path may be equal to or larger than the cross-sectional area inside the first pipe.

[0019] The wafer holding stage of the above (5) allows the refrigerant to flow smoothly in the parallel flow path. Since the refrigerant flows smoothly throughout the parallel flow path, the heat of the plate is easily transferred to the refrigerant, and the cooling efficiency by the refrigerant is increased.

[0020] (6) In any of the wafer holders described in (1) to (5) above, the cross-section of at least one of the plurality of flow paths may be rectangular, and the upper corners of the cross-section may be formed into curved surfaces.

[0021] The wafer holder described in (6) above allows the refrigerant to flow smoothly near the corners of the flow paths. Therefore, the cooling efficiency by the refrigerant is increased.

[0022] (7) In any of the wafer holders described in (1) to (6) above, the upper surface of at least one of the plurality of flow paths may have at least one of a convex portion and a concave portion. The convex portion has a shape in which the inner wall constituting the upper surface bulges toward the inside of the flow path, and the concave portion has a shape in which the inner wall constituting the upper surface is concave toward the outside of the flow path.

[0023] The wafer holder described in (7) above increases the contact area of the parallel flow paths. In particular, by increasing the contact area of the upper surface of the flow path close to the wafer, the cooling efficiency by the refrigerant is increased.

[0024] (8) Any of the wafer holders described in (1) to (7) above may include a support provided at the lower part of the plate. The support is a disk made of ceramics or a composite of ceramics. The wafer holder includes an adsorption mechanism for vacuum-adsorbing the plate and the support.

[0025] The wafer holder described in (8) above can suppress warping of the plate due to heat during wafer inspection by vacuum-adsorbing the plate and the support. Also, the wafer holder can increase the rigidity of the entire wafer holder by bringing the plate and the support into close contact with each other.

[0026] [Details of Embodiments of the Present Disclosure] A specific example of a wafer chuck according to an embodiment of the present disclosure will be described with reference to the drawings. The same reference numerals in the drawings denote the same or corresponding parts. The shapes, sizes, positional relationships, etc. shown in each figure are presented for the purpose of clarifying the description and do not necessarily represent the actual shapes, sizes, and positional relationships, etc. Note that the present invention is not limited to these examples, but is defined by the claims, and is intended to include all modifications within the meaning and scope equivalent to the claims.

[0027] The wafer chuck of the present disclosure is used as part of a semiconductor manufacturing apparatus. Here, a wafer chuck used in a wafer prober will be described as an example. A wafer prober is an inspection apparatus that measures the electrical performance of each chip of a semiconductor wafer on which a plurality of circuits are formed while controlling the wafer at a predetermined temperature before cutting the wafer into individual chips. The wafer chuck of the present disclosure is not limited to a wafer prober, and can also be used for wafer chucks for other applications having a similar structure.

[0028] [Embodiment 1] <Wafer Chuck> Referring to FIGS. 1 to 4, the wafer chuck 1 according to Embodiment 1 will be described. As shown in FIG. 1, the wafer chuck 1 according to Embodiment 1 includes a plate 2. The plate 2 has an upper surface 2u on which a wafer is placed. FIG. 1 schematically shows a cross-section of the wafer chuck 1 cut in a plane orthogonal to the upper surface 2u of the plate 2. Here, the side where the upper surface 2u is located is referred to as "upper", and the opposite side is referred to as "lower". The plate 2 includes a refrigerant flow path 3 through which refrigerant flows. The wafer chuck 1 according to Embodiment 1 further includes a support structure 5 that supports the plate 2.

[0029] For example, in the inspection of a wafer, power is supplied from a probe pin of a probe card to a chip while the probe pin is pressed against the chip formed on the wafer from above, and the electrical performance of the chip is measured. When power is supplied to the chip, the chip generates heat and the temperature of the wafer rises. Also, when the chip is pressed by the probe pin, a downward load is applied to the upper surface 2u of the plate 2.

[0030] One of the features of the wafer holding stage 1 of Embodiment 1 is that the refrigerant flow path 3 has a parallel flow path 3p composed of a plurality of flow paths 30. Hereinafter, the configuration of the wafer holding stage 1 will be described in detail.

[0031] (Plate) The plate 2 is a member on which a wafer (not shown) is placed. The wafer is placed on the upper surface 2u of the plate 2. Inside the plate 2, a refrigerant flow path 3 is provided. When refrigerant flows through the refrigerant flow path 3, the plate 2 is cooled, and the wafer can be cooled. The material of the wafer is, for example, silicon or a compound semiconductor. The plate 2 may be provided with a chuck mechanism (not shown) for adsorbing the wafer on the upper surface 2u. By adsorbing the wafer on the upper surface 2u, the warp of the wafer can be corrected, and the entire wafer can be brought into close contact with the upper surface 2u.

[0032] The shape of the plate 2 is, for example, a disc shape or a rectangular plate shape. The shape of the plate 2 is a shape adapted to the shape of the wafer, and is usually a disc shape. The size of the plate 2 is slightly larger than the size of the wafer. The size of the plate 2 is designed according to the size of the wafer. The diameter of the wafer is, for example, from 2 inches (about 50 mm) to 12 inches (about 300 mm). The diameter of the plate 2 is, for example, 100 mm or more and 400 mm or less.

[0033] The thickness of the plate 2 is, for example, 15 mm or more and 30 mm or less. The thickness of the plate 2 is a dimension along the direction perpendicular to the upper surface 2u. If the thickness of the plate 2 is 15 mm or more, it is easy to ensure the rigidity of the plate 2 and easy to form the refrigerant flow path 3. If the thickness of the plate 2 is 30 mm or less, the wafer holding stage 1 can be made compact. Also, the thinner the plate 2, the easier it is for the heat capacity of the plate 2 to be small. Therefore, the cooling efficiency by the refrigerant flowing through the refrigerant flow path 3 and the responsiveness of heating by a heater (described later) are likely to be high. The thickness of the plate 2 may further be 15 mm or more and 20 mm or less, or 16 mm or more and 18 mm or less.

[0034] Plate 2 is composed of a material with high thermal conductivity and high rigidity. The material of Plate 2 is, for example, metal, ceramics or a composite of ceramics. The metal constituting Plate 2 is, for example, copper, a copper alloy, aluminum or an aluminum alloy. The ceramics constituting Plate 2 is, for example, silicon carbide or aluminum nitride. The composite of ceramics constituting Plate 2 is, for example, a composite of silicon and silicon carbide, a composite of aluminum and silicon carbide, or a composite of aluminum, silicon and silicon carbide. The material of Plate 2 in this example is oxygen-free copper.

[0035] The thermal conductivity of Plate 2 is, for example, 100 W / m·K or more. The thermal conductivity of Plate 2 may further be 200 W / m·K or more, 300 W / m·K or more, 400 W / m·K or more. This thermal conductivity is the value at 20°C. The thermal conductivity of copper is about 400 W / m·K, and the thermal conductivity of aluminum is about 230 W / m·K. The thermal conductivity of silicon carbide is about 200 W / m·K, and the thermal conductivity of aluminum nitride is about 150 W / m·K. The thermal conductivity of a composite of silicon and silicon carbide (Si-SiC) is about 170 W / m·K.

[0036] The surface of Plate 2 may be subjected to surface treatment. The surface treatment is, for example, plating. Specific examples of plating are nickel plating or nickel-phosphorus plating. The method of plating is electroplating or electroless plating. When the material of Plate 2 is copper or a copper alloy, it is preferable that nickel plating or nickel-phosphorus plating is applied to the upper surface 2u on which the wafer is placed. Copper easily diffuses into silicon, which is a constituent material of the wafer. By arranging the above plating on the upper surface 2u of Plate 2, the phenomenon that copper contained in Plate 2 diffuses into the wafer can be suppressed.

[0037] As shown in FIG. 1, Plate 2 in this example is composed of an upper plate 21 and a lower plate 22 laminated vertically. The lower surface of the upper plate 21 and the upper surface of the lower plate 22 are joined to each other.

[0038] <Refrigerant Flow Path> Referring to FIGS. 2 to 4, the configuration of the refrigerant flow path 3 in this example will be described. Refrigerant flows through the refrigerant flow path 3. The refrigerant flow path 3 is formed inside the plate 2. FIG. 2 shows a cross-section of the portion of the plate 2 where the refrigerant flow path 3 is formed, as viewed from a direction orthogonal to the upper surface 2u, in a plane parallel to the upper surface 2u. In FIG. 2, the upper plate 21 is viewed from below. The refrigerant flow path 3 is arranged in a plane parallel to the upper surface 2u. The refrigerant flow path 3 in this example has an outward path that goes from the inlet 3i arranged on the outer periphery of the upper plate 21 toward the central portion of the upper plate 21, and a return path that goes from the central portion toward the outlet 3e arranged on the outer periphery. The inlet 3i and the outlet 3e are arranged adjacent to each other on the outer periphery of the upper plate 21. The outward path extends from the outer periphery of the upper plate 21 toward the central portion while the arc-shaped flow paths connected to the inlet 3i alternately fold back counterclockwise and clockwise. The return path is connected to the outward path, extends from the central portion of the upper plate 21 through between the folding points of the outward path toward the outer periphery of the upper plate 21, and reaches the outlet 3e. A wall is provided between each of the flow paths constituting the outward path and the return path. The shape of the refrigerant flow path 3 shown in FIG. 2 is an example, and the shape of the refrigerant flow path 3 is not limited. For example, the shape of the refrigerant flow path 3 may be such that a spiral outward path from the outer periphery toward the central portion and a spiral return path from the central portion toward the outer periphery are folded back at the central portion and arranged alternately in the radial direction. The number of the refrigerant flow paths 3 is one. The refrigerant is, for example, a liquid such as water, oil, ethylene glycol aqueous solution, or fluorine-based liquid, or a gas such as nitrogen gas or dry air.

[0039] As shown in FIG. 1, the refrigerant flow path 3 in this example is constituted by a groove formed on the lower surface of the upper plate 21 and the flat upper surface of the lower plate 22. Specifically, a groove corresponding to the shape of the refrigerant flow path 3 is formed on the lower surface of the upper plate 21, and the refrigerant flow path 3 is constituted by joining the lower surface of the upper plate 21 and the upper surface of the lower plate 22. The above groove only needs to be formed on at least one of the lower surface of the upper plate 21 and the upper surface of the lower plate 22. For example, the above groove may be formed on the upper surface of the lower plate 22, and the lower surface of the upper plate 21 may be flat. In this case, the refrigerant flow path 3 is constituted by the groove formed on the upper surface of the lower plate 22 and the flat lower surface of the upper plate 21. Alternatively, the above groove may be formed on both the lower surface of the upper plate 21 and the upper surface of the lower plate 22. In this case, the refrigerant flow path 3 is constituted by the above grooves formed on the lower surface of the upper plate 21 and the upper surface of the lower plate 22 overlapping vertically.

[0040] (Inlet / Outlet) As shown in FIG. 2, the refrigerant flow path 3 has an inlet 3i, an outlet 3e, and a parallel flow path 3p. The inlet 3i is the part where the refrigerant is introduced. The outlet 3e is the part where the refrigerant is discharged. The inlet 3i and the outlet 3e each open to the outer peripheral surface of the plate 2. There is one inlet 3i and one outlet 3e respectively. A first pipe 41 described later is connected to the inlet 3i. A second pipe 42 described later is connected to the outlet 3e. In this example, the inlet 3i and the outlet 3e are arranged adjacent to each other, but the inlet 3i and the outlet 3e may be provided at different positions. For example, the inlet 3i and the outlet 3e may be provided at opposite positions on the outer peripheral surface of the plate 2.

[0041] (Parallel Flow Path) The parallel flow path 3p is a flow path connecting between the inlet 3i and the outlet 3e. The parallel flow path 3p has a plurality of flow paths 30. As shown in FIG. 3, the plurality of flow paths 30 are arranged in parallel in a plane parallel to the upper surface 2u. FIG. 3 is a cross section cut by a plane orthogonal to the upper surface 2u, and shows an enlarged cross section of the parallel flow path 3p orthogonal to the direction in which the refrigerant flows. As shown in FIG. 2, the parallel flow path 3p is configured to branch from the inlet 3i into a plurality of flow paths 30 and the plurality of flow paths 30 merge at the outlet 3e. The refrigerant is introduced from the inlet 3i into the refrigerant flow path 3, passes through each flow path 30 of the parallel flow path 3p, and is discharged from the outlet 3e. As also shown in FIG. 4, the adjacent flow paths 30 are separated by a wall 35. The wall 35 is arranged over the entire length of the parallel flow path 3p. In this example, as shown in FIGS. 2 to 4, the number of flow paths 30 constituting one parallel flow path 3p is 2, and the number of walls 35 is 1. The end face of the wall 35 facing the inlet 3i may be formed into a curved surface. By forming the end face of the wall 35 facing the inlet 3i into a curved surface, the flow resistance can be reduced. The end face of the wall 35 facing the outlet 3e may also be formed into a curved surface.

[0042] The length of the parallel flow path 3p is appropriately set according to the size of the plate 2. The length of the parallel flow path 3p is the distance from the first end connected to the inlet 3i to the second end connected to the outlet 3e. The longer the parallel flow path 3p is, the larger the total contact area of the parallel flow path 3p becomes. The larger the contact area is, the higher the cooling efficiency by the refrigerant becomes. The contact area will be described later. The shorter the parallel flow path 3p is, the smaller the total pressure loss of the parallel flow path 3p becomes. The smaller the pressure loss is, the easier it is for the refrigerant to flow. When the diameter of the plate 2 is 310 mm, for example, the length of the parallel flow path 3p is 1500 mm or more and 10000 mm or less. If the length of the parallel flow path 3p is 1500 mm or more, it is easy to secure the contact area of the parallel flow path 3p. If the length of the parallel flow path 3p is 10000 mm or less, it is easy to keep the pressure loss of the parallel flow path 3p within an allowable range. The length of the parallel flow path 3p may further be 3000 mm or more and 9000 mm or less.

[0043] As shown in FIG. 3, the cross-sectional shape of each flow path 30 constituting the parallel flow path 3p in this example is rectangular. The cross-section of the flow path 30 is a cross-section orthogonal to the direction in which the refrigerant flows. The rectangular shape includes a square shape. The rectangular shape referred to here is not limited to a geometric rectangle, and includes, for example, a shape in which at least one of the four corners is rounded, etc., with details modified. The cross-section of the flow path 30 is not limited to a rectangular shape. The cross-section of the flow path 30 may be, for example, triangular, trapezoidal, or pentagonal like a home base, or circular or semi-circular. In this example, the cross-sections of the respective flow paths 30 have the same shape, but the cross-sections of the respective flow paths 30 may have different shapes.

[0044] The size of each flow path 30 constituting the parallel flow path 3p is appropriately set. In this example, each flow path 30 has the same size, but each flow path 30 may have different sizes. The width Wa of the flow path 30 is, for example, 1 mm or more and 10 mm or less. The width Wa of the flow path 30 is the dimension along the direction parallel to the upper surface 2u in the cross-section of the flow path 30. The larger the width Wa of the flow path 30, the larger the contact area of the parallel flow path 3p. If the width Wa of the flow path 30 is 1 mm or more, it is easy to secure the contact area. Also, if the width Wa of the flow path 30 is 1 mm or more, it is easy to secure the cross-sectional area of the flow path 30 so that the refrigerant can flow. If the width Wa of the flow path 30 is 5 mm or less, it is easy to suppress a decrease in the rigidity of the plate 2. Since the rigidity of the plate 2 can be ensured, when a downward load is applied to the upper surface 2u, it is possible to suppress the plate 2 from deforming by bending. Also, if the width Wa of the flow path 30 is 10 mm or less, the width Wp of the entire parallel flow path 3p becomes smaller. The width Wp of the parallel flow path 3p is the sum of the width Wa of each flow path 30 and the width Wb of the wall 35 disposed between adjacent flow paths 30. The width Wb of the wall 35 is equal to the interval between adjacent flow paths 30. When the cross-sectional shape and size of each flow path 30 are the same, the width Wp of the parallel flow path 3p is (Wa × n)+(Wb × (n - 1)). n is the number of flow paths. Since the width Wp of the parallel flow path 3p is small, the ratio of the parallel flow path 3p in the cross-section of the plate 2 increases. Since the length of the parallel flow path 3p can be ensured, it is easy to secure the contact area. The width Wa of the flow path 30 may be further 1 mm or more and 4 mm or less.

[0045] The width Wp of the parallel flow path 3p is, for example, 5 mm or more and 25 mm or less. If the width Wp of the parallel flow path 3p is 5 mm or more, it is easy to reduce the pressure loss. If the width Wp of the parallel flow path 3p is 25 mm or less, the ratio of the parallel flow path 3p in the cross section of the plate 2 can be increased, and it is easy to secure the length of the parallel flow path 3p. By making the parallel flow path 3p longer, it is easy to secure the liquid contact area. The width Wp of the parallel flow path 3p may further be 6 mm or more and 12 mm or less.

[0046] The wall 35 plays a role in enhancing the rigidity of the plate 2. When a downward load is applied to the upper surface 2u, the wall 35 supports the upper surface 2u, thereby suppressing deformation of the plate 2. The width Wb of the wall 35 is, for example, 1 mm or more and 5 mm or less. If the width Wb of the wall 35 is 1 mm or more, it is easy to secure the rigidity of the plate 2. If the width Wb of the wall 35 is 5 mm or less, the overall width Wp of the parallel flow path 3p becomes smaller. The width Wb of the wall 35 may further be 1 mm or more and 4 mm or less.

[0047] The height H of the flow path 30 is, for example, 3 mm or more and 10 mm or less. The height H of the flow path 30 is the dimension along the direction orthogonal to the upper surface 2u in the cross section of the flow path 30. The height Hp of the parallel flow path 3p is equal to the height H of each flow path 30. The larger the height H of the flow path 30, the larger the liquid contact area of the parallel flow path 3p. If the height H of the flow path 30 is 3 mm or more, it is easy to secure the liquid contact area. Also, if the height H of the flow path 30 is 3 mm or more, it is easy to secure the cross-sectional area of the flow path 30 so that the refrigerant can flow. If the height H of the flow path 30 is 10 mm or less, the plate 2 can be made thinner. The height H of the flow path 30 may further be 4 mm or more and 8 mm or less.

[0048] The larger the cross-sectional area Ap of the parallel flow path 3p is, the smaller the pressure loss of the parallel flow path 3p becomes, and the easier it is for the refrigerant to flow. By increasing the flow rate of the refrigerant, the cooling efficiency by the refrigerant becomes higher. The cross-sectional area Ap of the parallel flow path 3p is the sum of the cross-sectional areas A of all the flow paths 30. The cross-sectional area A of one flow path 30 is obtained by the product of the width Wa and the height H. When the cross-sectional shape and size of each flow path 30 are the same, the cross-sectional area Ap of the parallel flow path 3p is A×n = (Wa×H)×n. n is the number of flow paths. The cross-sectional area A of the flow path 30 is, for example, 3 mm 2 or more. When the cross-sectional area A of the flow path 30 is 3 mm 2 or more, it is easy to ensure the flow rate of the refrigerant. The cross-sectional area A of the flow path 30 is, for example, 3 mm 2 or more and 50 mm 2 or less, 4 mm 2 or more and 40 mm 2 or less, and further 7 mm 2 or more and 30 mm 2 or less. The upper limit of the cross-sectional area Ap of the parallel flow path 3p is, for example, 80 mm 2 . When the cross-sectional area Ap of the parallel flow path 3p is 80 mm 2 or less, the ratio of the parallel flow path 3p in the cross-section of the plate 2 is less likely to become excessively large. Therefore, it is easy to suppress the decrease in the rigidity of the plate 2 due to the reduction of the cross-section of the plate 2. Since the rigidity of the plate 2 can be ensured, it is easy to suppress the deformation of the plate 2 when a downward load is applied to the upper surface 2u.

[0049] The distance d from the upper surface of the upper 2u to the upper surface of the parallel flow path 3p is, for example, 5 mm or more and 15 mm or less. The upper surface of the parallel flow path 3p is the upper surface 30u of each flow path 30 that constitutes the parallel flow path 3p. The upper surface 30u of the flow path 30 is the surface closest to the upper surface 2u among the inner peripheral surfaces of the flow path 30. The heat of the wafer is transferred from the upper surface 2u of the plate 2 to the refrigerant in the parallel flow path 3p. The heat transferred from the upper surface 2u is first transferred to the refrigerant in contact with the upper surface 30u of the flow path 30. Therefore, the temperature of the refrigerant flowing near the upper surface 30u of the flow path 30 tends to be high. The shorter the distance d, the easier it is for the heat of the wafer to be transferred to the refrigerant in the parallel flow path 3p, so the wafer can be efficiently cooled by the refrigerant. If the distance d is 15 mm or less, the cooling efficiency by the refrigerant is high. If the distance d is 5 mm or more, it is easy to suppress the deformation of the plate 2 when a downward load is applied to the upper surface 2u. The distance d may further be 6 mm or more and 12 mm or less.

[0050] In FIGS. 2 to 4, the case where the number of flow paths 30 constituting the parallel flow path 3p is 2 is illustrated, but the number of flow paths 30 may be 3 or 4 or more. FIGS. 5 and 6 illustrate the case where the number of flow paths 30 constituting one parallel flow path 3p is 3. In this case, the number of walls 35 arranged between adjacent flow paths 30 is 2.

[0051] The larger the number of flow paths 30 constituting the parallel flow path 3p, the larger the contact area of the parallel flow path 3p, so the cooling efficiency by the refrigerant is high. However, when the number of flow paths 30 increases, the number of walls 35 increases. Therefore, when the width Wp of the parallel flow path 3p is constant, the width Wa of each flow path 30 becomes smaller. The cross-sectional area A of each flow path 30 becomes smaller, and the cross-sectional area Ap of the parallel flow path 3p decreases, so the pressure loss of the parallel flow path 3p increases. From the viewpoint of ensuring the cross-sectional area Ap of the parallel flow path 3p, the number of flow paths 30 is preferably 2 or 3. If the number of flow paths 30 is 2 or 3, an increase in the pressure loss of the parallel flow path 3p can be suppressed, and it is easy to ensure the flow rate of the refrigerant.

[0052] 〈Contact area〉 The larger the contact area of the parallel flow path 3p with the liquid surface, the easier it is for the heat of the plate 2 to be transferred to the refrigerant, so the cooling efficiency by the refrigerant increases. Since the cooling capacity of the plate 2 is improved, the wafer can be cooled more efficiently. The contact area of the parallel flow path 3p is the area of the portion in contact with the refrigerant. The contact area of the parallel flow path 3p is equal to the area of the inner peripheral surface of the entire parallel flow path 3p. The contact area Sp of the parallel flow path 3p is obtained by multiplying the perimeter Lp of the cross-section of the parallel flow path 3p by the length X of the parallel flow path 3p. That is, the contact area Sp per unit length of the parallel flow path 3p is proportional to the perimeter Lp. Since there is a wall 35 between adjacent flow paths 30 in the parallel flow path 3p, the perimeter Lp is larger than when there is no wall 35, and the contact area increases. The perimeter Lp of the parallel flow path 3p is the sum of the perimeters L of the cross-sections of all the flow paths 30. When the cross-sectional shapes and sizes of all the flow paths 30 are the same, the perimeter Lp of the parallel flow path 3p is L×n. n is the number of flow paths. The perimeter L of the cross-section of one flow path 30 is (Wa×2)+(H×2). The perimeter Lp of the cross-section of the parallel flow path 3p is L×n = ((Wa×2)+(H×2))×n.

[0053] The contact area Sp of the parallel flow path 3p is larger than the contact area Sv of the virtual refrigerant flow path. The virtual refrigerant flow path is a single flow path having a cross-sectional area equivalent to the cross-sectional area of a closed region circumscribing the entirety of all the flow paths 30 constituting the parallel flow path 3p and all the walls 35 partitioning the respective flow paths 30. For example, the virtual refrigerant flow path in FIG. 3 is a rectangular flow path having a cross-sectional area equivalent to the cross-sectional area of a rectangle circumscribing the entirety of two flow paths 30 and one wall 35. The cross-section of the virtual refrigerant flow path includes the cross-sections of all the flow paths 30 constituting the parallel flow path 3p and the cross-sections of the walls 35 arranged between adjacent flow paths 30. The perimeter Lv of the cross-section in the virtual refrigerant flow path with respect to the parallel flow path 3p is (Wp×2)+(Hp×2).

[0054] For example, when Wa = 4 mm, Wb = 2 mm, Wp = 10 mm, H = Hp = 10 mm, and n = 2, the perimeter Lp of the parallel flow path 3p consisting of two flow paths 30 shown in FIG. 3 is Lp = 56 mm. When Wp = 10 mm and Hp = 10 mm for the perimeter Lv of the virtual refrigerant flow path, Lv = 40 mm. The ratio (Sp / Sv) of the contact area of this parallel flow path 3p to the contact area of the virtual refrigerant flow path becomes Lp / Lv = 1.4. The contact area Sp of the parallel flow path 3p in this case is 1.4 times the contact area Sv of the virtual refrigerant flow path. Also, when Wa = 2 mm, Wb = 2 mm, Wp = 10 mm, H = Hp = 10 mm, and n = 3, the perimeter Lp of the parallel flow path 3p consisting of three flow paths 30 shown in FIG. 5 is Lp = 72 mm. The ratio (Sp / Sv) of the contact area of this parallel flow path 3p to the virtual refrigerant flow path becomes Lp / Lv = 1.8. The contact area Sp of the parallel flow path 3p in this case is 1.8 times the contact area Sv of the virtual refrigerant flow path. Thus, as the number of flow paths 30 increases, the ratio of the contact area of the parallel flow path 3p to the virtual refrigerant flow path increases.

[0055] 〈Hydraulic Diameter〉 Furthermore, the ratio of the hydraulic diameter of the parallel flow path 3p to the hydraulic diameter of the virtual refrigerant flow path is, for example, 0.2 or more. The hydraulic diameter DHp of the parallel flow path 3p is four times the value obtained by dividing the cross-sectional area Ap of the parallel flow path 3p by the perimeter Lp of the cross-section. That is, the hydraulic diameter DHp of the parallel flow path 3p is 4Ap / Lp. The hydraulic diameter DHv of the virtual refrigerant flow path is four times the value obtained by dividing the cross-sectional area Av of the virtual refrigerant flow path by the perimeter Lv of the cross-section. That is, the hydraulic diameter DHv is 4Av / Lv. The cross-sectional area Av of the virtual refrigerant flow path is Wp×Hp.

[0056] For example, when Wa = 4 mm, Wb = 2 mm, Wp = 10 mm, H = Hp = 10 mm, and n = 2 for the parallel flow path 3p consisting of two flow paths 30 shown in Fig. 3, the hydraulic diameter DHp of the parallel flow path 3p is DHp = 4Ap / Lp = 4×((Wa×H)×n) / (((Wa×2)+(H×2))×n) ≒ 5.7 mm. Also, when Wa = 2 mm, Wb = 2 mm, Wp = 10 mm, H = Hp = 10 mm, and n = 3 for the parallel flow path 3p consisting of three flow paths 30 shown in Fig. 5, the hydraulic diameter DHp of the parallel flow path 3p is DHp ≒ 3.3 mm. When the hydraulic diameter DHv of the virtual refrigerant flow path is Wp = 10 mm and Hp = 10 mm, the hydraulic diameter DHv is 10 mm. For the parallel flow path 3p with a hydraulic diameter DHp of 5.7 mm, the hydraulic diameter ratio (DHp / DHv) of the parallel flow path 3p to the virtual refrigerant flow path is 0.57. For the parallel flow path 3p with a hydraulic diameter DHp of 3.3 mm, the hydraulic diameter ratio (DHp / DHv) is 0.33.

[0057] The pressure loss ΔP of the entire parallel flow path 3p can be expressed by the following equation. ΔP=(α×X) / DHp 4.75 α is a proportionality constant. X is the length of the parallel flow path 3p. DHp is the hydraulic diameter of the parallel flow path 3p.

[0058] From the equation of the pressure loss ΔP above, the pressure loss ΔP is proportional to X / DHp 4.75 When the length X of the parallel flow path 3p is constant, the pressure loss ΔP is proportional to the -4.75th power of the hydraulic diameter DHp. The larger the hydraulic diameter DHp, the smaller the pressure loss ΔP.

[0059] When the hydraulic diameter DHv of the virtual refrigerant flow path is 1, the hydraulic diameter DHp of the parallel flow path 3p is less than 1. The larger the ratio (DHp / DHv) of the hydraulic diameter of the parallel flow path 3p to the virtual refrigerant flow path, the more the increase in the pressure loss of the parallel flow path 3p can be suppressed. If the hydraulic diameter ratio is 0.2 or more, the increase in the pressure loss can be effectively suppressed. The hydraulic diameter ratio may be further 0.3 or more. The hydraulic diameter ratio is, for example, 0.2 or more and 0.8 or less.

[0060] Here, the theoretical pressure loss ΔPt per unit length of the parallel flow path 3p is (1 / DHp 4.75 ) × the flow path length. The unit of the flow path length is millimeter (mm). For example, it is preferable that the flow path length is 1500 mm and the theoretical pressure loss ΔPt is 0.3 MPa or less.

[0061] Figures 7 to 9 show modified examples of the cross-sectional shape of the flow path 30 constituting the parallel flow path 3p.

[0062] 〈Convex portion · Concave portion〉 Figure 7 shows an example having a convex portion 31 on the upper surface 30u of the flow path 30. This convex portion 31 is formed to be convex toward the inside of the flow path 30. The convex portion 31 is a ridge extending along the length of the flow path 30. The convex portion 31 may be provided continuously over the entire length of the flow path 30 or intermittently. In the example shown in Figure 7, the convex portion 31 is provided on the upper surface 30u of each flow path 30 constituting the parallel flow path 3p, but the convex portion 31 may be provided on the upper surface 30u of at least one flow path 30. Also, in the example shown in Figure 7, the number of convex portions 31 for one flow path 30 is one, but at least one convex portion 31 is sufficient, and the number of convex portions 31 may be plural.

[0063] Figure 8 shows an example having a concave portion 32 on the upper surface 30u of the flow path 30. This concave portion 32 is formed to be concave toward the outside of the flow path 30. The concave portion 32 is a concave groove extending along the length of the flow path 30. The concave portion 32 may be provided continuously over the entire length of the flow path 30 or intermittently. In the example shown in Figure 8, the concave portion 32 is provided on the upper surface 30u of each flow path 30 constituting the parallel flow path 3p, but the concave portion 32 may be provided on the upper surface 30u of at least one flow path 30. Also, in the example shown in Figure 8, the number of concave portions 32 for one flow path 30 is one, but at least one concave portion 32 is sufficient, and the number of concave portions 32 may be plural. Although not shown, the upper surface 30u of the flow path 30 may have both the convex portion 31 and the concave portion 32.

[0064] When the upper surface 30u of the flow path 30 has at least one of the convex portion 31 and the concave portion 32, the contact area of the parallel flow path 3p increases, so that the cooling efficiency by the refrigerant becomes high. Further, when at least one of the convex portion 31 and the concave portion 32 is provided on the upper surface 30u where the heat of the wafer is first transmitted, the heat transmitted from the upper surface 30u to the refrigerant increases, so that the cooling efficiency by the refrigerant becomes high.

[0065] 〈Corner of the flow path〉 FIG. 9 shows an example in which the upper corner 30c of the flow path 30 is formed as a curved surface. The cross section of the flow path 30 is rectangular. The corner 30c is a corner portion located at both ends of the upper surface 30u of the flow path 30. In the example shown in FIG. 9, the corner 30c is formed in an arc shape in the cross section of the flow path 30. The radius of curvature of the arc is, for example, 1 / 4 or more and 1 / 2 or less of the width Wa of the flow path 30. Further, in the example shown in FIG. 9, the corners 30c of the respective flow paths 30 constituting the parallel flow path 3p are formed as curved surfaces, but the corner 30c of at least one flow path 30 may be formed as a curved surface.

[0066] Since the heat of the wafer is first transmitted to the upper surface 30u, the temperature of the refrigerant flowing near the upper surface 30u tends to be high. When the corner 30c of the flow path 30 is angular, the flow velocity of the refrigerant flowing near the corner 30c decreases. The refrigerant flowing near the corner 30c tends to stay. When the corner 30c of the flow path 30 is formed as a curved surface, the refrigerant flowing near the corner 30c flows smoothly, so that the cooling efficiency by the refrigerant is improved.

[0067] 〈First pipe · Second pipe〉 As shown in FIGS. 2 and 4, the plate 2 in this example has a first pipe 41 and a second pipe 42. The first pipe 41 is connected to the inlet 3i of the refrigerant flow path 3. The first pipe 41 is a pipe that introduces the refrigerant pumped from a pump (not shown) into the inlet 3i. The second pipe 42 is connected to the outlet 3e of the refrigerant flow path 3. The second pipe 42 is a pipe that sends the refrigerant discharged from the outlet 3e to a tank (not shown). The refrigerant recovered in the tank is sent to a cooler and cooled by the cooler. The refrigerant cooled by the cooler is sent to a pump. The refrigerant circulates in the order of tank → cooler → pump → refrigerant flow path 3 → tank.

[0068] The cross-sectional area Ap of the parallel flow path 3p of the refrigerant flow path 3 is, for example, not less than the cross-sectional area A1 inside the first pipe 41. In other words, the cross-sectional area A1 inside the first pipe 41 is not more than the cross-sectional area Ap of the parallel flow path 3p of the refrigerant flow path 3. By satisfying the relationship Ap ≥ A1 between the cross-sectional area Ap of the parallel flow path 3p and the cross-sectional area A1 of the first pipe 41, the refrigerant can flow smoothly in the parallel flow path 3p. When the refrigerant flows smoothly throughout the parallel flow path 3p, the heat of the plate 2 is easily transferred to the refrigerant, and the cooling efficiency by the refrigerant is increased. The cross-sectional area A2 inside the second pipe 42 may also be not more than the cross-sectional area Ap of the parallel flow path 3p. By satisfying the relationship Ap ≥ A2, the refrigerant can flow smoothly in the parallel flow path 3p. In this example, the cross-sectional area A2 inside the second pipe 42 is the same as the cross-sectional area A1 of the first pipe 41. The cross-sectional area A1 of the first pipe 41 and the cross-sectional area A2 of the second pipe 42 may be different.

[0069] FIG. 10 shows a cross section taken along a plane perpendicular to the upper surface 2u of the plate 2 and along the central axis of the first pipe 41. As shown in FIG. 10, the upper surface 41u of the inner peripheral surface of the first pipe 41 and the upper surface of the parallel flow path 3p may be flush. Alternatively, the upper surface of the parallel flow path 3p may be lower than the upper surface 41u of the first pipe 41. The upper surface of the parallel flow path 3p is the upper surface 30u of each flow path 30 that constitutes the parallel flow path 3p. As described above, since the heat of the wafer first reaches the upper surface 30u, the temperature of the refrigerant flowing near the upper surface 30u tends to increase. When the upper surface 30u of the flow path 30 is higher than, i.e., located above, the upper surface 41u of the first pipe 41, the refrigerant flowing near the upper surface 30u adjacent to the upper surface 41u tends to stagnate, and air bubbles tend to remain in the flow path 30. When the upper surface 30u of the flow path 30 and the upper surface 41u of the first pipe 41 are flush, or when the upper surface 30u of the flow path 30 is lower than, i.e., located below, the upper surface 41u of the first pipe 41, air bubbles are less likely to remain in the flow path 30, so the cooling efficiency by the refrigerant is improved. Among them, when the upper surface 30u of the flow path 30 and the upper surface 41u of the first pipe 41 are flush, the refrigerant flowing near the upper surface 30u can flow smoothly, so the cooling efficiency by the refrigerant is further improved. The same applies to the second pipe 42 regarding this improvement in cooling efficiency. That is, the upper surface of the inner peripheral surface of the second pipe 42 may also be flush with the upper surface 30u of the flow path 30, or the upper surface 30u of the flow path 30 may be lower than the upper surface of the inner peripheral surface of the second pipe 42.

[0070] (Support structure) As shown in FIG. 1, the plate 2 is supported by a support structure 5. The support structure 5 includes, in order from above, a support 50, a heater unit 51, a support column 52, and a pedestal 53.

[0071] 〈Support〉 The support 50 is a member for suppressing the warping of the plate 2 due to heat during the inspection of the wafer. The support 50 may be provided as needed and is not an essential member. When there is no support 50, the support column 52 described later supports the plate 2. If the support 50 is provided, when a load is applied to the upper surface 2u, the deformation of the plate 2 can be more effectively suppressed. The support 50 is in close contact with the lower surface 2d of the plate 2. By bringing the plate 2 and the support 50 into close contact, the rigidity of the entire wafer holding stage 1 can be increased. The support 50 and the plate 2 are brought into close contact, for example, by vacuum adsorption. This vacuum adsorption is performed by a vacuum adsorption mechanism. The vacuum adsorption mechanism includes, for example, a vacuum flow path 50a provided on at least one of the lower surface of the plate 2 and the upper surface of the support 50. The vacuum flow path 50a is, for example, at least one of a groove and a hole provided on at least one of the lower surface of the plate 2 and the upper surface of the support 50. The vacuum flow path 50a in this example is a groove provided on the lower surface of the plate 2. A vacuum pump (not shown) is connected to the vacuum flow path 50a. By driving the vacuum pump, the inside of the vacuum flow path 50a becomes a vacuum state, and the plate 2 and the support 50 are adsorbed. By vacuum-adsorbing the plate 2 and the support 50, warping of the plate 2 can be suppressed. Also, by using the vacuum adsorption mechanism, the support 50 and the plate 2 can be easily attached and detached. The shape of the support 50 is a shape that matches the shape of the plate 2. The support 50 in this example is disk-shaped. The material of the support 50 is, for example, ceramics or a composite of ceramics. The ceramics constituting the support 50 are, for example, alumina, mullite alumina, mullite, cordierite, steatite, silicon nitride, silicon carbide. The composite of ceramics constituting the support 50 is, for example, a composite of silicon and silicon carbide, a composite of aluminum and silicon carbide, or a composite of aluminum, silicon, and silicon carbide.

[0072] 〈Heater Unit〉 The heater unit 51 is a device for heating the wafer to a predetermined temperature. The heater unit 51 can adjust the temperature of the wafer. The heater unit 51 is disposed on the lower surface of the support 50. The heater unit 51 includes a disc-shaped heater. The heater is constituted by, for example, a resistance heating element. Alternatively, instead of the heater unit 51, a temperature adjustment unit having a wafer heating function and a wafer cooling function may be provided. The temperature adjustment unit has, for example, a structure in which a heating body and a cooling body are integrally laminated. The heating body includes the heater described above. A refrigerant flow path is formed in the cooling body. The refrigerant flowing through the cooling body can use a refrigerant different from the refrigerant flowing through the parallel flow path 3p, but the same refrigerant may be used. By having both the heating function and the cooling function in this way, the degree of freedom in adjusting the temperature of the wafer can be increased.

[0073] 〈Support post〉 The support post 52 is a member that supports the plate 2 on the pedestal 53. When there is the support 50, the support post 52 supports the support 50 on the pedestal 53. The shape of the support post 52 is, for example, columnar or cylindrical. The support post 52 supports the support 50 or the plate 2 through a through hole formed in the heater unit 51. The outer diameter of the support post 52 is sufficiently smaller than the outer diameter of the support 50 or the outer diameter of the plate 2. The contact area between the support post 52 and the support 50 or the contact area between the support post 52 and the plate 2 is small. Thereby, heat conduction from the plate 2 to the pedestal 53 via the support post 52 is suppressed, and the heat uniformity of the plate 2 is maintained. Further, a space can be formed between the heater unit 51 and the pedestal 53 by the support post 52. This space contributes to heat insulation between the plate 2 and the pedestal 53. The support post 52 in this example is cylindrical. In this example, the number of the support posts 52 is plural so that the plate 2 or the support 50 can be supported in a well-balanced manner. The material of the support post 52 is ceramics such as alumina, mullite alumina, mullite, cordierite, steatite, or silicon nitride.

[0074] 〈Pedestal〉 The pedestal 53 is a member that supports the support 50, the heater unit 51, and the support column 52. The material of the pedestal 53 is ceramics such as alumina, aluminum nitride, and silicon carbide, or a composite of ceramics such as a composite of silicon and silicon carbide, a composite of aluminum and silicon carbide, and a composite of aluminum, silicon, and silicon carbide.

[0075] (Test Example 1) In Test Example 1, a simulation experiment was conducted to evaluate the cooling capacity of the plate 2 provided with the refrigerant flow path 3 having the parallel flow paths 3p.

[0076] The trial calculation models shown in Table 1 were designed. The trial calculation model 1 is a refrigerant flow path composed of a single flow path instead of parallel flow paths. The trial calculation model 1 corresponds to the virtual refrigerant flow path described above. The trial calculation model 2 is a parallel flow path composed of two flow paths. The trial calculation model 3 is a parallel flow path composed of three flow paths. The trial calculation model 4 is a parallel flow path composed of four flow paths.

[0077] For each trial calculation model, the number n of flow paths constituting the parallel flow path, the width Wp and height Hp of the parallel flow path, the width Wa and height H of each flow path constituting the parallel flow path, and the width Wb of the wall are shown in Table 1. In the trial calculation model, the width Wp and height Hp are each set to 1, and the width Wa, width Wb, and height H are each shown as relative values. Since the trial calculation model 1 is composed of a single flow path, the width Wa and width Wb of the trial calculation model 1 are set to zero. The width and height of the single flow path of the trial calculation model 1 are the width Wp and height Hp. Also, in the trial calculation models 2 to 4, the width Wp and width Wb were fixed, and only the width Wa was changed according to the number n of flow paths.

[0078] The contact area of each trial calculation model was calculated, and the contact area ratio of each trial calculation model was obtained. The contact area S and the contact area ratio of each trial calculation model are shown in Table 1. The contact area S of the trial calculation model 1 corresponds to the contact area Sv of the virtual refrigerant flow path described above. The contact area S of each of the trial calculation models 2 to 4 is the contact area Sp of the parallel flow path described above. The contact area ratio of each trial calculation model is the ratio of the contact area of each trial calculation model to the contact area of the trial calculation model 1.

[0079] Table 1 also shows the hydraulic diameter DH and the hydraulic diameter ratio of each trial model. The hydraulic diameter DH of trial model 1 corresponds to the hydraulic diameter DHv of the virtual refrigerant flow path described above. The hydraulic diameter DH of each of trial models 2 to 4 is the hydraulic diameter DHp of the parallel flow path described above. The hydraulic diameter ratio of each trial model is the ratio of the hydraulic diameter of each trial model to the hydraulic diameter of trial model 1. Table 1 also shows the theoretical pressure loss ΔPt of each trial model obtained from the hydraulic diameter DH.

[0080] · Evaluation of cooling capacity The following simulations were performed using each trial model to evaluate the cooling capacity. A heating element was placed on the upper surface 2u of plate 2. The heating element is a rectangular plate having a thermal conductivity equivalent to that of a silicon wafer. The heat generation density of the heating element is 30 W / cm 2 . It is assumed that the side and upper surfaces of the heating element are insulated. While circulating the refrigerant through the refrigerant flow path 3, power was supplied to the heating element with a downward load applied to the upper surface of the heating element. The temperature of the refrigerant introduced into the refrigerant flow path 3 is set to 23°C. The load is 100 kgf = 980 N. Power was supplied until the temperature of the heating element became stable, and the temperature difference ΔT (°C) between the temperature of the lower surface of the heating element when the temperature became stable and the temperature of the lower surface of the heating element before power supply was determined. The smaller the temperature difference ΔT (°C) of the heating element, the better the cooling capacity.

[0081] Table 1 shows the temperature difference ΔTsim of each trial model obtained from the simulation results. The temperature difference ΔTsim is a trial value obtained by simulation.

[0082]

Table 1

[0083] As shown in Table 1, the liquid contact area S of each of the trial calculation models 2 to 4 is larger than the liquid contact area S of the trial calculation model 1, and the liquid contact area ratio of each of the trial calculation models 2 to 4 is larger than 1. Also, from the comparison of the trial calculation models 2 to 4, it can be seen that the larger the number n of flow paths, the larger the liquid contact area S and the larger the liquid contact area ratio.

[0084] The temperature difference ΔTsim of each of the trial calculation models 2 to 4 is smaller than the temperature difference ΔTsim of the trial calculation model 1. From this simulation result, it can be seen that the trial calculation models 2 to 4 have higher cooling capacity than the trial calculation model 1. Also, from the comparison of the trial calculation models 2 to 4, it can be seen that the larger the liquid contact area ratio, the higher the cooling capacity.

[0085] Next, the measured models shown in Table 2 were designed. The measured model A is a refrigerant flow path consisting of a single flow path, not a parallel flow path. The measured model B is a parallel flow path consisting of two flow paths. The measured model C is a parallel flow path consisting of three flow paths.

[0086] For each measured model, the number n of flow paths constituting the parallel flow path, the width Wp and height Hp of the parallel flow path, the width Wa and height H of each flow path constituting the parallel flow path, and the width Wb of the wall are shown in Table 2. In the measured model, the width Wp, height Hp, width Wa, width Wb, and height H are shown as absolute values. Since the measured model A is composed of a single flow path, the width Wa and width Wb of the measured model A are set to zero. Also, the width and height of the single flow path of the measured model A are the width Wp and height Hp.

[0087] Similar to the trial calculation models shown in Table 1, the liquid contact area of each measured model was calculated, and the liquid contact area ratio of each measured model was obtained. The liquid contact area ratio of each measured model is shown in Table 2. The liquid contact area ratio of the measured model A is 1. The liquid contact area ratio of each of the measured models B and C is the ratio of the liquid contact area of the parallel flow path to the liquid contact area of the above-mentioned virtual refrigerant flow path.

[0088] An experiment was conducted to actually evaluate the cooling capacity under the same conditions as the simulation described above using each measured model. The temperature difference ΔT of each measured model obtained from the experimental results is shown in Table 2. The temperature difference ΔT is a measured value obtained by the experiment.

[0089]

Table 2

[0090] As shown in Table 2, the temperature difference ΔT of each of the measured model B and the measured model C is smaller than the temperature difference ΔT of the measured model A. From this experimental result, it can be seen that, similar to the simulation results shown in Table 1, the larger the contact area ratio, the smaller the temperature difference ΔT and the higher the cooling capacity.

[0091] Figure 11 shows a graph of the relationship between the contact area ratio and the temperature difference of the heating element. The horizontal axis of Figure 11 indicates the contact area ratio, and the vertical axis indicates the temperature difference ΔT of the heating element. The ● marks in Figure 11 represent the measured models, where Measured A represents the measured model A, Measured B represents the measured model B, and Measured C represents the measured model C. The □ marks in Figure 11 represent the trial calculation models, where Trial Calculation 1 represents the trial calculation model 1, Trial Calculation 2 represents the trial calculation model 2, Trial Calculation 3 represents the trial calculation model 3, and Trial Calculation 4 represents the trial calculation model 4. The temperature difference ΔT of the trial calculation model is normalized by the proportionality constant β obtained from the temperature difference ΔT of the measured model. The proportionality constant β was obtained from the comparison between the temperature difference ΔT measured using the measured model and the temperature difference ΔTsim of the trial calculation model. The obtained proportionality constant β was 3.1. The value obtained by multiplying the temperature difference ΔTsim of each trial calculation model by the obtained proportionality constant β = 3.1 was taken as the temperature difference ΔT of each trial calculation model.

[0092] The straight line AC1 in Figure 11 represents the linear approximation curve of the plotted trial calculation models. As shown in Figure 11, the measured models are approximately located on the straight line AC1, and the measured models and the trial calculation models are in relatively good agreement.

[0093] (Test Example 2) In Test Example 2, a test was conducted to verify the relationship between the hydraulic diameter ratio and the pressure loss.

[0094] The measured models shown in Table 3 were designed. The measured models D and E are parallel channels composed of three channels. The measured model F is a parallel channel composed of two channels.

[0095] For each measured model, Table 3 shows the number n of channels constituting the parallel channel, the width Wp and height Hp of the parallel channel, the width Wa and height H of each channel constituting the parallel channel, the width Wb of the wall, and the length X of the parallel channel.

[0096] Similar to the trial calculation models shown in Table 1, the hydraulic diameter of each measured model was calculated, and the hydraulic diameter ratio of each measured model was obtained. Table 3 shows the hydraulic diameter ratio of each measured model. The hydraulic diameter ratio of each measured model is the ratio of the hydraulic diameter of the parallel channel to the hydraulic diameter of the virtual refrigerant channel described above.

[0097] An experiment was actually conducted to flow the refrigerant through each measured model. Pressure gauges were installed at the inlet and outlet of the refrigerant. The pressures at the inlet and outlet were measured while the refrigerant was flowing, and the pressure difference between the outlet and the inlet was measured as the pressure loss. Table 3 shows the pressure loss ΔP and the pressure loss Δp per unit length of each measured model obtained from the experimental results. The pressure loss Δp per unit length is the value obtained by dividing the pressure loss ΔP by the length X. In Table 3 and FIG. 12 described later, "E+00", "E-04", "E-05", "E-06" mean "×10 0 ", "×10 -4 ", "×10 -5 ", "×10 -6 ", respectively.

[0098]

Table 3

[0099] Figure 12 shows a graph of the relationship between the hydraulic diameter ratio and the pressure loss per unit length. The horizontal axis of Figure 12 represents the hydraulic diameter ratio, and the vertical axis represents the pressure loss per unit length. The ● marks in Figure 12 represent the measured models, where measured D represents measured model D, measured E represents measured model E, and measured F represents measured model F. The □ marks in Figure 12 represent the trial calculation models, where trial calculation 1 represents trial calculation model 1, trial calculation 2 represents trial calculation model 2, and trial calculation 3 represents trial calculation model 3. Note that the pressure loss per unit length of the trial calculation models is normalized by the proportionality constant α obtained from the pressure loss per unit length of the measured models. The proportionality constant α was obtained from the comparison between the pressure loss per unit length Δp measured using the measured models and the theoretical pressure loss ΔPt of the trial calculation models. The obtained proportionality constant α was 10 -6 It was. The value obtained by multiplying the theoretical pressure loss ΔPt of each trial calculation model by the obtained proportionality constant α = 10 -6 was used as the pressure loss per unit length of each trial calculation model.

[0100] The curve AC2 in Figure 12 shows the power approximation curve of the plotted trial calculation models. As shown in Figure 12, the measured models are approximately located on the power approximation curve AC2 of the trial calculation models, and the measured models and the trial calculation models are in relatively good agreement.

Explanation of symbols

[0101] 1 Wafer holding stage 2 Plate 2u Upper surface, 2d Lower surface 21 Upper plate, 22 Lower plate 3 Refrigerant flow path 3i Inlet, 3e Outlet 3p Parallel flow path 30 Flow path 30u Upper surface, 30c Corner 31 Protrusion, 32 Recess 35 Wall 41 First pipe, 42 Second pipe 41u Upper surface 5 Support structure 50 Support, 50a Vacuum flow path, 51 Heater unit, 52 Support pillar, 53 Pedestal Wa, Wb, Wp Width Height of H and Hp Distance d

Claims

1. A wafer holding stage comprising a plate having an upper surface on which a wafer is placed, wherein the plate has a refrigerant flow path through which refrigerant flows inside the plate, the refrigerant flow path having an inlet through which the refrigerant is introduced, an outlet through which the refrigerant is discharged, and parallel flow paths connecting between the inlet and the outlet, the parallel flow paths having a plurality of flow paths arranged in parallel in a plane parallel to the upper surface, a wafer holding stage.

2. When the hydraulic diameter of a virtual refrigerant flow path is set to 1, the ratio of the hydraulic diameter of the parallel flow paths to the hydraulic diameter of the virtual refrigerant flow path is 0.2 or more, wherein the virtual refrigerant flow path has a cross section equivalent to a cross section encompassing the entirety of the parallel flow paths, the wafer holding stage according to Claim 1.

3. The wafer holding stage according to Claim 1 or Claim 2, wherein the number of the plurality of flow paths is 2 or more.

4. The plurality of flow paths are separated by a wall disposed between adjacent flow paths, wherein the wall is disposed over the entire length of the parallel flow paths, the wafer holding stage according to Claim 1 or Claim 2.

5. The inlet and the outlet each open to an outer peripheral surface of the plate, wherein the inlet, the outlet, and the parallel flow paths are disposed in the same plane parallel to the upper surface, the wafer holding stage according to Claim 1 or Claim 2.

6. The plate has a first pipe connected to the inlet of the refrigerant flow path, wherein an upper surface of an inner peripheral surface of the first pipe and an upper surface of the parallel flow paths are flush, or the upper surface of the parallel flow paths is lower than the upper surface of the inner peripheral surface of the first pipe, the wafer holding stage according to Claim 1 or Claim 2.

7. The plate has a first pipe connected to the inlet of the refrigerant flow path, wherein a cross-sectional area of the parallel flow paths is equal to or larger than a cross-sectional area inside the first pipe, the wafer holding stage according to Claim 1 or Claim 2.

8. A cross section of at least one of the plurality of flow paths is rectangular, wherein an upper corner of the cross section is formed into a curved surface, the wafer holding stage according to Claim 1 or Claim 2.

9. An upper surface of at least one of the plurality of flow paths has at least one of a convex portion and a concave portion, wherein the convex portion has a shape in which an inner wall constituting the upper surface bulges toward the inside of the flow path, wherein the concave portion has a shape in which an inner wall constituting the upper surface is recessed toward the outside of the flow path, the wafer holding stage according to Claim 1 or Claim 2.

10. comprising a support provided at a lower portion of the plate; the support is a disk made of ceramics or a composite of ceramics; The wafer holding stage according to claim 1 or 2, comprising an adsorption mechanism for adsorbing the plate and the support in a vacuum.

Images