Wafer placement table

Abstract

A wafer placement table 10 comprises: a ceramic plate 20 having a wafer placement surface 22 on the upper surface thereof; and a cooling plate provided on the lower surface of the ceramic plate 20. A refrigerant flow path 32 built into the cooling plate has a first portion 32x, and a second portion 32y branching from the first portion 32x into two or more paths that run in parallel to each other. The cross-sectional area of the first portion 32x is smaller than the sum of the cross-sectional areas of the paths of the second portion 32y.

Description

Wafer mounting table

[0001] The present invention relates to a wafer stage.

[0002] Conventionally, a wafer mounting table has been known that includes a ceramic plate having a wafer mounting surface on its upper surface, a cooling plate provided on the lower surface of the ceramic plate, and a refrigerant flow path built into the cooling plate. For example, Patent Document 1 (JP-A-2005-102626) discloses a wafer mounting table in which the cooling plate is formed of a material with high thermal conductivity, such as aluminum, and the distance between the upper surface of the refrigerant flow path and the wafer mounting surface is constant from the inlet to the outlet of the refrigerant flow path, and the cross-sectional shape of the refrigerant flow path varies depending on the position of the refrigerant flow path. Patent Document 1 describes that the cross-sectional area of ​​the refrigerant flow path corresponding to the relatively high-temperature portion of the wafer mounting surface is smaller than the cross-sectional area of ​​the refrigerant flow path corresponding to the relatively low-temperature portion of the wafer mounting surface. It also describes that the width of the upper surface of the refrigerant flow path from the inlet to the outlet is constant, and the height of the refrigerant flow path is shorter at the position corresponding to the relatively high-temperature portion of the wafer mounting surface than at the position corresponding to the relatively low-temperature portion of the wafer mounting surface.

[0003] Japanese Patent Application Laid-Open No. 2021-28961

[0004] However, while the configuration of Patent Document 1 suppresses uneven heat dissipation in the refrigerant flow path by devising the cross-sectional shape of the cooling plate, it may not be possible to sufficiently secure a cross-sectional area of ​​the flow path corresponding to the area where the temperature of the wafer mounting surface is relatively low, and thus it may not be possible to sufficiently suppress uneven heat dissipation.

[0005] The present invention has been made to solve the above problems, and a main object of the present invention is to suppress temperature variations on the wafer mounting surface of a wafer mounting table.

[0006] [1] A wafer mounting table of the present invention is a wafer mounting table comprising: a ceramic plate having a wafer mounting surface on an upper surface thereof; a cooling plate provided on a lower surface of the ceramic plate; and a refrigerant flow path built into the cooling plate, wherein the refrigerant flow path has a first portion and a second portion that branches off from the first portion into two or more branches running parallel to each other, and the cross-sectional area of ​​the first portion is smaller than the sum of the cross-sectional areas of the branches of the second portion.

[0007] In this wafer mounting table, the coolant flow path has a first portion and a second portion that branches off from the first portion into two or more branches that run parallel to each other. Therefore, even when the width through which the coolant flow path can be passed is narrow, each branch in the second portion can fit within that width while ensuring a relatively large flow path cross-sectional area for the entire second portion. The cross-sectional area of ​​the first portion is smaller than the sum of the cross-sectional areas of the branches in the second portion. Therefore, the flow velocity in the first portion of the coolant flow path is faster than in the second portion, resulting in higher cooling efficiency. Therefore, by adjusting the arrangement of the first and second portions of the wafer mounting table, such as by positioning the first portion to correspond to an area with high cooling demand, temperature unevenness on the wafer mounting surface can be suppressed.

[0008] [2] In the wafer mounting table of the present invention (the wafer mounting table described in [1] above), the first portion may be disposed in correspondence with an outer periphery region of the wafer mounting surface, and the second portion may be disposed in correspondence with a central region of the wafer mounting surface. Generally, the heat input of plasma into the wafer mounting table is greater in the outer periphery region of the wafer mounting surface than in the central region. In consideration of this, by disposing the first and second portions as described above, the cooling efficiency of the outer periphery region of the wafer mounting surface can be made higher than that of the central region, thereby effectively suppressing temperature variations on the wafer mounting surface.

[0009] [3] In the wafer mounting table of the present invention (the wafer mounting table described in [2] above), the ceramic plate may have an annular focus ring mounting surface around the wafer mounting surface that is one step lower than the wafer mounting surface, and an annular focus ring having an outer diameter larger than the outer diameter of the ceramic plate and the outer diameter of the cooling plate may be mounted on the focus ring mounting surface. In this case, since the focus ring overhangs the wafer mounting table, the peripheral region of the wafer mounting surface is likely to reach a higher temperature. Therefore, the application of the present invention is highly significant.

[0010] [4] In the wafer stage of the present invention (the wafer stage described in any one of [1] to [3] above), the cross-sectional area of ​​each branch of the second portion may be greater than half the cross-sectional area of ​​the first portion. The larger the cross-sectional area of ​​each branch of the second portion, the slower the flow velocity of each branch, and therefore the relatively faster flow velocity of the first portion, thereby further suppressing uneven heat transfer.

[0011] [5] In the wafer stage of the present invention (the wafer stage according to any one of [1] to [4] above), the refrigerant flow path may have a turning portion that reverses the direction of the flow path and branches into two at the turning portion to prevent unevenness in the amount of refrigerant distributed to each branch in the second portion. This prevents unevenness in the heat removal capacity of each branch, and further reduces uneven heat removal.

[0012] [6] In the wafer stage of the present invention (the wafer stage described in any one of [1] to [4] above), the refrigerant flow path may be branched into a branch that continues the curvature before the branching and a branch that temporarily deviates from the curvature, thereby suppressing unevenness in the amount of refrigerant distributed to each branch in the second section. This also suppresses unevenness in the heat removal capacity of each branch, thereby further suppressing uneven heat removal.

[0013] [7] In the wafer stage of the present invention (the wafer stage described in any one of [1] to [6] above), a non-flow path area, which is an area of ​​a portion where the refrigerant flow path is not formed, may be 50% or more in a region where the second portion is arranged in a plan view. The larger the non-flow path area, the greater the degree of freedom in arranging components other than the refrigerant flow path.

[0014] [8] In the wafer mounting table of the present invention (the wafer mounting table according to any one of [1] to [7] above), the wafer mounting surface may have an area where cooling is highly required and an area where cooling is low, and the first portion may be disposed in the area of ​​the wafer mounting surface corresponding to the area where cooling is high, and the second portion may be disposed in the area where cooling is low. For example, the area where cooling is high may be a peripheral area of ​​the wafer mounting surface, and the area where cooling is low may be a central area of ​​the wafer mounting surface.

[0015] [9] In the wafer mounting table of the present invention (the wafer mounting table described in any one of [1] to [8] above), the heat exchange efficiency of a region of the wafer mounting surface corresponding to the first portion may be higher than the heat exchange efficiency of a region of the wafer mounting surface corresponding to the second portion. In this case, the region of the wafer mounting surface corresponding to the first portion may be a peripheral region of the wafer mounting surface, and the region of the wafer mounting surface corresponding to the second portion may be a central region of the wafer mounting surface.

[0016] 6 is a cross-sectional view of the wafer mounting table 10. A plan view of the wafer mounting table 10. A cross-sectional view taken along line A-A in FIG. 1. A partially enlarged view of FIG. 1. An explanatory diagram showing the results of studying the flow of refrigerant in a refrigerant flow path. A cross-sectional view of the wafer mounting table 110. A plan view of the wafer mounting table 110. A cross-sectional view taken along line B-B in FIG. 6. A partially enlarged view of FIG. 6.

[0017] [First Embodiment] A first embodiment of the present invention will be described with reference to the drawings. Fig. 1 is a cross-sectional view of a wafer mounting table 10 (a cross-sectional view of the wafer mounting table 10 cut along a plane including the central axis of the wafer mounting table 10), Fig. 2 is a plan view of the wafer mounting table 10, Fig. 3 is a cross-sectional view taken along the line A-A in Fig. 1, Fig. 4 is an enlarged view of a portion of Fig. 1, and Fig. 5 is an explanatory diagram showing the results of an investigation into the flow of refrigerant within a refrigerant flow channel. Fig. 5A is a vector diagram showing the flow velocity distribution within the refrigerant flow channel (the original diagram is a color diagram in which the flow velocity is represented in descending order from red to orange, yellow, green, blue, indigo, and purple), and Figs. 5B and 5C are explanatory diagrams showing preferred branching configurations.

[0018] The wafer mounting table 10 is used to perform CVD, etching, or the like using plasma on a wafer W. The wafer mounting table 10 includes a ceramic plate 20, a cooling plate 30, and a bonding layer 40.

[0019] Ceramic plate 20 is made of a ceramic material such as alumina or aluminum nitride. Ceramic plate 20 has a wafer mounting surface 22, an electrostatic electrode 23, and a focus ring mounting surface 24. Hereinafter, the focus ring may be abbreviated as "FR."

[0020] The wafer mounting surface 22 is a circular surface provided on the upper surface of the ceramic plate 20. A wafer W is mounted on the wafer mounting surface 22. The wafer mounting surface 22 has a plurality of gas holes 50 (six in this embodiment) that penetrate the wafer mounting table 10 in the vertical direction, through which a thermally conductive gas such as He gas is supplied from a gas supply source (not shown). Although not shown, an annular seal band is formed along the outer edge of the wafer mounting surface 22, and a plurality of small circular protrusions are formed on the entire surface of the area surrounded by the seal band. The seal band and the small circular protrusions have the same height, for example, several μm to several tens of μm. The wafer mounting surface 22 has areas that are prone to high temperatures (areas requiring high cooling) and areas that are less likely to become high temperatures (areas requiring low cooling). In this embodiment, when processing the wafer W with plasma, the heat input of the plasma is greater on the outer periphery side, and therefore, as shown in FIG. 2, the outer periphery region 22a (lightly shaded region) of the wafer mounting surface 22 is an area requiring high cooling, and the central region 22b (darkly shaded region) of the wafer mounting surface 22 is an area requiring low cooling.

[0021] The electrostatic electrode 23 is a planar mesh electrode or plate electrode, and is connected to a DC power supply (not shown) via a power supply terminal 26. When a DC voltage is applied to the electrostatic electrode 23, the wafer W is attracted and fixed to the wafer mounting surface 22 (specifically, the upper surface of the seal band and the upper surfaces of the small circular protrusions) by electrostatic attraction, and when the application of the DC voltage is stopped, the wafer W is released from the attraction and fixation to the wafer mounting surface 22. The power supply terminal 26 is inserted into a terminal hole 56 provided in the wafer mounting table 10 between the lower surface of the electrostatic electrode 23 and the lower surface of the cooling plate 30.

[0022] The FR mounting surface 24 is provided in an annular shape around the wafer mounting surface 22. The height of the FR mounting surface 24 is one step lower than the height of the wafer mounting surface 22. An annular focus ring 60 is mounted on the FR mounting surface 24. The focus ring 60 is made of, for example, Si. A circumferential groove 62 is formed in the upper part of the inner surface of the focus ring 60 to prevent it from coming into contact with the wafer W. The outer diameter of the focus ring 60 is larger than the outer diameter of the ceramic plate 20 and the outer diameter of the cooling plate 30. Therefore, the focus ring 60 is mounted on the FR mounting surface 24 in a state where it protrudes outside the wafer mounting table 10 (in an overhanging state).

[0023] The cooling plate 30 is a disc-shaped plate having a refrigerant flow path 32 therein through which a refrigerant can circulate. As shown in FIG. 3 , the refrigerant flow path 32 is provided across the entire surface of the ceramic plate 20 from one end (inlet 32 ​​in) to the other end (outlet 32 ​​out) in a plan view. In this embodiment, the refrigerant flow path 32 is formed in a spiral shape in a plan view. Such a cooling plate 30 can be fabricated, for example, by diffusion bonding multiple layered members. The refrigerant is supplied to the inlet 32 ​​in of the refrigerant flow path 32 from a refrigerant circulation device (not shown), passes through the refrigerant flow path 32, and is discharged from the outlet 32 ​​out of the refrigerant flow path 32 and returns to the refrigerant circulation device. The refrigerant circulation device can adjust the refrigerant to a desired temperature. The refrigerant is preferably a liquid, preferably electrically insulating. Examples of electrically insulating liquids include fluorine-based inert liquids.

[0024] Materials used for the cooling plate 30 include metal materials and metal-ceramic composite materials. Metal materials include Al, Ti, Mo, and alloys thereof. Metal-ceramic composite materials include metal matrix composites (MMCs) and ceramic matrix composites (CMCs). Specific examples of such composite materials include materials containing Si, SiC, and Ti (also known as SiSiCTi), porous SiC materials impregnated with Al and / or Si, and composite materials of Al2O3 and TiC. To prevent warping of the wafer mounting table 10, materials with a thermal expansion coefficient similar to that of the ceramic plate 20 are preferred for the cooling plate 30. When the ceramic plate 20 is made of alumina, the cooling plate 30 is preferably made of pure Ti or an α-βTi alloy. This is because the thermal expansion coefficients of pure Ti and α-βTi alloys are similar to that of alumina. To enhance heat dissipation performance, materials with high thermal conductivity, such as Al, are preferred for the cooling plate 30. The thermal conductivity of Al is 150 to 200 W / mK. The cooling plate 30 may be formed of a material with a lower thermal conductivity than Al. Examples of such materials include Ti-containing materials. The thermal conductivity of the cooling plate 30 may be 50 W / mK or less, or may be 5 to 20 W / mK. For example, the thermal conductivity of pure Ti is 17 W / mK, and the thermal conductivity of an α-β Ti alloy is 7.5 W / mK. The material used for the cooling plate 30 may also be a conductive material.

[0025] The bonding layer 40 bonds the lower surface of the ceramic plate 20 to the upper surface of the cooling plate 30. The bonding layer 40 may be, for example, a metal layer formed from solder or a metal brazing material, or a resin layer formed from a resin adhesive.

[0026] The refrigerant flow path 32 will be described in detail. As shown in FIG. 2 , the refrigerant flow path 32 has a first portion 32x corresponding to the outer peripheral region 22a (region requiring high cooling) of the wafer mounting surface 22 and a second portion 32y corresponding to the central region 22b (region requiring low cooling). The first portion 32x corresponding to the outer peripheral region 22a of the refrigerant flow path 32 extends from an inlet 32in to an intermediate position 32mid of the refrigerant flow path 32. The second portion 32y corresponding to the central region 22b of the refrigerant flow path 32 extends from an intermediate position 32mid to an outlet 32out of the refrigerant flow path 32. The second portion 32y branches into two branches, 32y1 and 32y2, at a branch point 32div located at the intermediate position 32mid, join at a junction point 32join, and reach the outlet 32out. Although the branch point 32div is located at the midpoint 32mid in FIGS. 2 and 3, the branch point 32div may be located at a position away from the midpoint 32mid.

[0027] The refrigerant flow path 32 has a turn portion 32turn that reverses the flow path direction, and branches into two branches, 32y1 and 32y2, at a branch point 32div located midway through the turn portion 32turn. The turn portion 32turn reverses the counterclockwise direction from the inlet 32in to the turn portion 32turn to a clockwise direction from the turn portion 32turn to the outlet 32out. The refrigerant flow path 32 is formed to branch into two branches, 32y1 and 32y2, at the branch point 32div located midway through the turn portion 32turn, thereby preventing the amount of refrigerant distributed to each branch 32y1, 32y2 from being biased toward one of the branches (this point will be described later using FIG. 5).

[0028] As shown in FIG. 4 , the cross-sectional area S of the refrigerant flow path 32 is smaller than the sum of the cross-sectional areas Sy1 and Sy2 of the branches 32y1 and 32y2 of the second portion 32y corresponding to the central region 22b, i.e., Sx<Sy1+Sy2. The smaller the cross-sectional area S, the faster the flow velocity of the refrigerant through the refrigerant flow path 32, resulting in higher cooling efficiency. The cross-sectional area S is the area of ​​a cross section (flow path cross section) of the refrigerant flow path 32 cut along a plane perpendicular to the longitudinal direction of the refrigerant flow path 32 (each branch 32y1 and 32y2 in the second portion 32y). The cross-sectional areas Sy1 and Sy2 of the branches 32y1 and 32y2 of the second portion 32y may be set as appropriate. However, in this embodiment, the cross-sectional areas Sy1 and Sy2 are the same as the cross-sectional area Sx of the first portion 32x, i.e., Sy1=Sy2=Sx. The flow path cross-sectional area S of the refrigerant flow path 32 is a value obtained by multiplying the horizontal length W by the vertical length H. The horizontal length W and the vertical length H may be set appropriately depending on the flow path cross-sectional area S. In this embodiment, the horizontal lengths Wy1 and Wy2 of the branches 32y1 and 32y2 of the second portion 32y are the same as the horizontal length Wx of the first portion 32x, i.e., Wy1 = Wy2 = Wx. Furthermore, the vertical lengths Hy1 and Hy2 of the branches 32y1 and 32y2 of the second portion 32y are the same as the vertical length Hx of the first portion 32x, i.e., Hy1 = Hy2 = Hx.

[0029] Next, an example of how the wafer mounting table 10 is used will be described. The wafer mounting table 10 is fixed inside a semiconductor process chamber (not shown). A focus ring 60 is placed on the FR mounting surface 24, and a wafer W is placed on the wafer mounting surface 22. In this state, a DC voltage is applied to the electrostatic electrode 23 to attract the wafer W to the wafer mounting surface 22. At the same time, a thermally conductive gas (e.g., He gas) is supplied to a gas hole 50 (a passage leading from the underside of the cooling plate 30 to the wafer mounting surface 22) provided inside the wafer mounting table 10. This fills the space surrounded by the underside of the wafer W and the seal band of the wafer mounting surface 22 with gas, thereby improving thermal conduction between the wafer W and the wafer mounting surface 22. The interior of the chamber is then set to a predetermined vacuum atmosphere (or reduced-pressure atmosphere), and an RF voltage is applied to the cooling plate 30 while a process gas is supplied from a showerhead provided on the ceiling of the chamber. This generates plasma between the wafer W and the showerhead. The plasma is then utilized to perform CVD film formation or etching on the wafer W.

[0030] When processing the wafer W with plasma in this manner, the heat input from the plasma is greater in the outer peripheral region of the wafer W than in the central region, and therefore the outer peripheral region of the wafer W is more likely to become hotter than the central region. Therefore, to uniformly maintain the temperature of the wafer W, it is necessary to cool the outer peripheral region 22a of the wafer mounting surface 22 more efficiently than the central region 22b. Taking this into consideration, in this embodiment, the cross-sectional area S of the coolant flow path 32 is adjusted as described above. As a result, the first portion 32x of the coolant flow path 32, which corresponds to the outer peripheral region 22a, has a higher cooling efficiency than the second portion 32y, which corresponds to the central region 22b.

[0031] In the wafer mounting table 10 described above, the coolant flow path 32 includes a first portion 32x corresponding to an area requiring high cooling and a second portion 32y corresponding to an area requiring low cooling. The second portion 32y branches from the first portion 32x at a branch point 32div into two branches 32y1 and 32y2, which run parallel to each other. Therefore, even when the width through which the coolant flow path 32 can pass is narrow, for example, because the gas hole 50 is close to the terminal hole 56, the branches 32y1 and 32y2 of the second portion can fit within the narrow width, while still ensuring a relatively large flow path cross-sectional area for the entire second portion 32y. Furthermore, the flow path cross-sectional area Sx of the first portion 32x corresponding to an area requiring high cooling of the coolant flow path 32 is smaller than the sum Sy1 + Sy2 of the flow path cross-sectional areas of the branches 32y1 and 32y2 of the second portion 32y corresponding to an area requiring low cooling. Therefore, the first portion 32x of the coolant flow path 32, which corresponds to an area requiring high cooling, has a faster flow rate than the second portion 32y, which corresponds to an area requiring low cooling, resulting in higher cooling efficiency. In this way, the first portion 32x corresponding to an area requiring high cooling has one branch, and the second portion 32y corresponding to an area requiring low cooling has two or more branches, thereby controlling the flow rate. Furthermore, the flow rate can be controlled by changing the cross-sectional areas of the first and second portions 32x and 32y. Therefore, by adjusting the arrangement of the first and second portions 32x and 32y, such as by positioning the first portion 32x so that it corresponds to an area requiring high cooling, temperature unevenness on the wafer mounting surface 22 can be suppressed, thereby improving the thermal uniformity of the wafer W. The cross-sectional area S of the coolant flow path 32 may satisfy 1.5Sx < Sy1 + Sy2 < 2.5Sx or 1.8Sx < Sy1 + Sy2 < 2.2Sx.

[0032] Furthermore, the plasma heat input to the wafer mounting table 10 is generally greater in the outer circumferential region 22a than in the central region 22b of the wafer mounting surface 22. Taking this into consideration, the first portion 32x is disposed at a position corresponding to the outer circumferential region 22a of the wafer mounting surface 22, and the second portion 32y is disposed at a position corresponding to the central region 22b of the wafer mounting surface 22. This makes it possible to increase the cooling efficiency of the outer circumferential region 22a of the wafer mounting surface 22 compared to the central region 22b, thereby effectively suppressing temperature variations in the wafer mounting surface 22.

[0033] Furthermore, ceramic plate 20 has an annular focus ring mounting surface 24 around wafer mounting surface 22, which is one step lower than wafer mounting surface 22. An annular focus ring 60 having an outer diameter larger than the outer diameters of ceramic plate 20 and cooling plate 30 is mounted on focus ring mounting surface 24. In this case, because focus ring 60 overhangs (extends outside) wafer mounting table 10, an outer peripheral region 22a of wafer mounting surface 22 is likely to reach a higher temperature. For this reason, the application of the present invention is highly significant.

[0034] Furthermore, the flow path cross-sectional areas Sy1 and Sy2 of the branches 32y1 and 32y2 of the second portion 32y are both greater than half the flow path cross-sectional area Sx of the first portion 32x. The larger the flow path cross-sectional areas Sy1 and Sy2 of the branches 32y1 and 32y2 of the second portion 32y, the slower the flow velocity of the branches 32y1 and 32y2, thereby further suppressing uneven heat removal. The flow path cross-sectional areas Sy1 and Sy2 of the branches 32y1 and 32y2 may be greater than or equal to two-thirds or even three-quarters of the flow path cross-sectional area Sx of the first portion 32x. The flow path cross-sectional areas Sy1 and Sy2 of the branches 32y1 and 32y2 may be less than or equal to two times or less than 1.5 times the flow path cross-sectional area Sx of the first portion 32x.

[0035] The refrigerant flow path 32 has a turn-back portion 32turn that reverses the flow path direction, and branches into two branches 32y1 and 32y2 at a branch point 32div midway through the turn-back portion 32turn, thereby suppressing bias in the amount of refrigerant distributed to each branch 32y1 and 32y2 of the second portion 32y. This suppresses bias in the heat removal capacity of each branch 32y1 and 32y2, thereby further suppressing uneven heat removal.

[0036] Regarding this point, the results of an investigation into the flow of refrigerant within a refrigerant flow path are described with reference to FIG. 5 . When a refrigerant flow path has a curved shape, such as a spiral, centrifugal force tends to increase the flow velocity toward the outer periphery (outside the curve), as shown in (1) of FIG. 5A . Therefore, simply branching the flow path by providing a branch point near the center of the flow path may result in a bias in the amount of refrigerant distributed to the outer branch. On the other hand, if a turning section with a smaller curvature (e.g., a curvature radius R of 20 mm or less) than the previous curve (e.g., a curvature radius R of 50 mm or more) is provided midway through the curve, the centrifugal force is temporarily canceled in the turning section. As shown in (2) of FIG. 5A , the flow velocity becomes the highest near the center of the flow path, and the difference in flow velocity between the outer periphery (outside the curve) and the inner periphery (inside the curve) becomes smaller. Therefore, for example, if the refrigerant flow path is branched into two at a branch point midway through the turning section, as shown in FIG. 5B , the amount of refrigerant distributed to each branch is prevented from being biased toward one of the branches. When the branch point is provided in the middle of the turning portion, it is preferable to place the branch point near the center of the part where the flow velocity is higher near the center than on either side of the flow path.

[0037] Furthermore, in a plan view of the wafer mounting table 10, a non-flow-path area, which is an area of ​​a portion where the refrigerant flow paths 32 are not formed, may be 50% or more in the region where the second portion 32y is arranged (e.g., the central region 22b where cooling requirements are low). The larger the non-flow-path area, the greater the degree of freedom in arranging components other than the refrigerant flow paths (such as the gas holes 50, terminal holes 56, and lift pin holes described below). In this regard, we compared the case where the second portion 32y of the refrigerant flow path 32 is not branched and the flow path cross-sectional area Sy is twice that of the first portion 32x with the case where the second portion 32y is branched into branches 32y1 and 32y2 and the flow path cross-sectional areas Sy1 and sy2 of each branch are the same as that of the first portion 32x (the total cross-sectional area is twice that of the first portion 32x). When the wafer temperature variation was within a predetermined range (e.g., within 10°C), the non-flow path area was 41.3% in the former case and 54.6% in the latter case, thereby increasing the non-flow path area to over 50%. In this comparison, the cooling plate materials used were a first material (e.g., Ti) with a thermal conductivity of 20 W / mK, a second material (e.g., Al) with a thermal conductivity of 100 W / mK, and a third material (e.g., Al) with a thermal conductivity of 200 W / mK. For the first portion 32x, the flow path cross-sectional area Sx = 84 mm 2 , length Wx = 7 mm, length Hx = 12 mm. When the second portion 32y is not branched, its flow path cross-sectional area Sy = 168 mm 2 When the second portion 32y is branched, the cross-sectional area of ​​the branch 32y1 of the second portion 32y is set to Sy1=84 mm. 2 , length Wy1 = 7 mm, length Hy1 = 12 mm. For the branch 32y2 of the second portion 32y, the flow path cross-sectional area Sy2 = 84 mm 2 , length Wy2 = 7 mm, and length Hy2 = 12 mm. The non-flow path area may be, for example, 70% or less. The region where the second portion 32y is arranged (here, the central region 22b) may be the smallest circumscribed circle that includes all of the branches 32y1 and 32y2.

[0038] Furthermore, with respect to the flow path length L of the refrigerant flow path 32, the ratio Ly1 / Ly2 of the lengths Ly1 and Ly2 (not shown) of the branches 32y1 and 32y2 is preferably 4 / 5 or more and 5 / 4 or less. This reduces the difference in pressure loss between the branches 32y1 and 32y2. Note that, when the ratio Ly1 / Ly2 is large, the difference in pressure loss between the branches 32y1 and 32y2 may be reduced by adjusting the cross-sectional shape of the branches 32y1 and 32y2. The lengths Ly1 and Ly2 of the branches 32y1 and 32y2 may be equal to or greater than half the length Lx of the first portion 32x, or may be equal to or greater than the length Lx of the first portion 32x. Furthermore, the lengths Ly1 and Ly2 of the branches 32y1 and 32y2 may be equal to or less than 10 times the length Lx of the first portion 32x, or may be equal to or less than 5 times the length Lx of the first portion 32x. The length Lx of the first portion 32x and the lengths Ly1 and Ly2 of the branches 32y1 and 32y2 may each be 20 mm or greater. Regarding the cross-sectional area S of the refrigerant flow path 32, the ratio Sy1 / Sy2 of the cross-sectional areas Sy1 and Sy2 of the branches 32y1 and 32y2 is preferably 4 / 5 or greater and 5 / 4 or less. This reduces the difference in pressure loss between the branches 32y1 and 32y2.

[0039] Second Embodiment A second embodiment of the present invention will be described with reference to the drawings. Fig. 6 is a cross-sectional view of the wafer mounting table 110 (a cross-sectional view of the wafer mounting table 110 cut along a plane including the central axis of the wafer mounting table 110), Fig. 7 is a plan view of the wafer mounting table 110, Fig. 8 is a cross-sectional view taken along the line B-B of Fig. 6, and Fig. 9 is an enlarged partial view of Fig. 6. In Figs. 6 to 9, the same components as those in the above-described embodiment are designated by the same reference numerals, and their description will be omitted.

[0040] 8, the refrigerant flow path 132 is provided so as to cover the entire surface of the ceramic plate 20 from one end (inlet 132in) to the other end (outlet 132out) in a plan view. The refrigerant flow path 132 is formed in a spiral shape. The refrigerant, the refrigerant circulation device, etc. may be the same as those in the above-described embodiment.

[0041] 7 , the coolant flow path 132 has a first portion 132x corresponding to the outer circumferential region 22a (region requiring high cooling) of the wafer mounting surface 22 and a second portion 132y corresponding to the central region 22b (region requiring low cooling). The first portion 132x of the coolant flow path 132 corresponding to the outer circumferential region 22a extends from an inlet 132in to a midpoint 132mid of the coolant flow path 132. The second portion 132y of the coolant flow path 132 corresponding to the central region 22b extends from a midpoint 132mid to an outlet 132out of the coolant flow path 132. The second portion 132y branches into two branches, 132y1 and 132y2, at a branch point 132div, join at a junction 132join, and reach the outlet 132out.

[0042] The refrigerant flow path 132 is formed so as to branch at the branch point 132div into a branch 132y1 that continues the curvature tendency before the branch, and a branch 132y2 that deviates from this curvature tendency to the outside, thereby preventing the amount of refrigerant distributed to each branch 132y1, 132y2 from being biased toward one branch.

[0043] Regarding the cross-sectional area S of the refrigerant flow path 132, the cross-sectional area Sx of the first portion 132x corresponding to the outer circumferential region 22a is smaller than the sum of the cross-sectional areas Sy1 and Sy2 of the branches 132y1 and 132y2 of the second portion 132y corresponding to the central region 22b, i.e., Sx < Sy1 + Sy2. Therefore, the first portion 132x corresponding to the region of the refrigerant flow path 132 with high cooling demand has a faster flow rate and higher cooling efficiency than the second portion 132y corresponding to the region with low cooling demand. The cross-sectional area S, horizontal length W, and vertical length H of the refrigerant flow path 132 are equivalent to the cross-sectional area S, horizontal length W, and vertical length H of the refrigerant flow path 32.

[0044] The use of the wafer stage 110 is similar to the manufacturing and use of the wafer stage 10, and therefore, the description thereof will be omitted here.

[0045] In the wafer mounting table 110 described above, similarly to the wafer mounting table 10 described above, the first portion 132x corresponding to the area with high cooling demand is branched into one branch, and the second portion 132y corresponding to the area with low cooling demand is branched into two or more branches to control the flow rate. Furthermore, by controlling the flow rate by changing the flow path cross-sectional area between the first portion 132x and the second portion 132y, it is possible to suppress temperature unevenness on the wafer mounting surface 22 of the wafer mounting table 110, thereby improving the thermal uniformity of the wafer W.

[0046] Furthermore, the refrigerant flow path 132 is formed to branch into a branch 132y1 that continues the curved flow path before the branch at a branch point 132div, and a branch 132y2 that deviates from the curved flow path. This prevents the amount of refrigerant distributed to each branch 132y1, 132y2 from being biased toward one branch. This prevents bias in the heat removal capacity of each branch 132y1, 132y2, and further reduces uneven heat removal. Note that the branch 132y2 may branch at an angle of 30° to 90° relative to the branch 132y1.

[0047] In this regard, as explained using FIG. 5 , the flow velocity tends to increase toward the outer periphery of the flow path (outside the curve). Therefore, simply branching the flow path by providing a branch point near the center of the flow path may result in a bias in the amount of refrigerant distributed to the outer branch. Therefore, when providing a branch point at a portion such as (1) in FIG. 5A , for example, as shown in FIG. 5C , the branch point may be provided closer to the outer periphery of the flow path before the branch, thereby dividing the flow path into a branch that continues the curved flow before the branch and a branch that deviates from the curved flow. This allows the inner branch (inside the curve) to become the main flow path, thereby reducing the amount of refrigerant distributed to the outer branch (outside the curve), and preventing the amount of refrigerant distributed to each branch from being biased toward one branch.

[0048] Furthermore, in a plan view, the area where the second portion 132y is located (e.g., the central region 22b where cooling requirements are low) may have a non-flow-path area, which is the area of ​​the portion where the refrigerant flow path 132 is not formed, of 50% or more. The larger the non-flow-path area, the greater the degree of freedom in arranging components other than the refrigerant flow path (such as the gas holes 50, terminal holes 56, and lift pin holes described below). The non-flow-path area may be, for example, 70% or less.

[0049] Furthermore, the length L and flow path cross-sectional area S of the refrigerant flow path 132 may be made similar to the length L and flow path cross-sectional area S of the refrigerant flow path 32, thereby reducing the pressure loss difference between branch 132y1 and branch 132y2.

[0050] It goes without saying that the present invention is not limited to the above-described embodiment, and can be embodied in various forms as long as they fall within the technical scope of the present invention.

[0051] For example, in the first and second embodiments described above, the refrigerant flow paths 32, 132 are branched into two paths, but may be branched into three or more paths. The arrangement of the branch points 32div, 132div of the refrigerant flow paths 32, 132 is not limited to that described above, and may be appropriately set so that the amount of refrigerant distributed to each branch 32y1, 32y2 and each branch 132y1, 132y2 is a desired amount.

[0052] In the first and second embodiments described above, the refrigerant flow paths 32, 132 are assumed to merge at the junctions 32join, 132join, but they may also lead to outlets provided individually in each branch 32y1, 32y2 or each branch 132y1, 132y2 without merging.

[0053] In the first and second embodiments described above, the area requiring high cooling is the peripheral area 22a of the wafer mounting surface 22, and the area requiring low cooling is the central area 22b of the wafer mounting surface 22, but this is not particularly limited.

[0054] In the first and second embodiments described above, the heat exchange efficiency of the peripheral region 22a of the wafer mounting surface 22 corresponding to the first portions 32x and 132x may be higher than the heat exchange efficiency of the central region 22b corresponding to the second portions 32y and 132y. Using the wafer mounting table 10 as an example, the heat exchange efficiency can be determined as follows. First, a first chiller capable of circulating a refrigerant while controlling the temperature of the refrigerant is connected to the inlet 32in and outlet 32out of the refrigerant flow path 32, and a refrigerant at the same temperature as room temperature (e.g., 25°C) is circulated through the refrigerant flow path 32. At the same time, a refrigerant at a predetermined temperature (e.g., 80 to 100°C) is prepared in a second chiller. Then, a valve is used to switch from the refrigerant at the same temperature as room temperature to the refrigerant at the predetermined temperature, and the refrigerant at the predetermined temperature is circulated through the refrigerant flow path 32. After a predetermined time (e.g., 10 seconds) has elapsed since the refrigerant was switched, the temperature distribution of the wafer mounting surface 22 is measured. The temperature rise rate (temperature rise amount per unit time (°C / sec)) is calculated from the temperature distribution, and this temperature rise rate is used as an index of heat exchange efficiency. For example, when a 25°C refrigerant is switched to an 80°C refrigerant in the wafer mounting table 10, the temperature rise rate of the outer peripheral region 22a of the wafer mounting surface 22 is 5.5°C / sec or more, and the temperature rise rate of the central region 22b is 5°C / sec or less. Therefore, it can be seen that the heat exchange efficiency of the outer peripheral region 22a is higher than that of the central region 22b. Note that the temperature rise rate at the boundary between the outer peripheral region 22a and the central region 22b is an intermediate value.

[0055] In the first and second embodiments described above, the electrostatic electrode 23 is built into the ceramic plate 20 at a position facing the wafer mounting surface 22. In addition, an FR adsorption electrode for electrostatically attracting the focus ring 60 may be provided inside the ceramic plate 20 at a position facing the FR mounting surface 24.

[0056] In the first and second embodiments described above, the ceramic plate 20 has been illustrated as having the wafer mounting surface 22 and the FR mounting surface 24, but is not limited thereto. For example, the ceramic plate 20 may have the wafer mounting surface 22 but not the FR mounting surface 24.

[0057] In the first and second embodiments described above, the outer diameter of the focus ring 60 is larger than the outer diameter of the wafer mounting table 10 (the outer diameter of the ceramic plate 20 and the outer diameter of the cooling plate 30), but this is not limiting. For example, the outer diameter of the focus ring 60 may be the same as the outer diameter of the wafer mounting table 10.

[0058] In the first and second embodiments described above, the refrigerant flow paths 32, 132 are formed in a spiral shape in plan view, but are not limited to this. For example, the refrigerant flow paths 32, 132 may be formed in a zigzag shape in plan view.

[0059] In the first and second embodiments described above, the wafer mounting table 10 has the electrostatic electrode 23 built into the ceramic plate 20, but the present invention is not limited to this. For example, instead of or in addition to the electrostatic electrode 23, the ceramic plate 20 may have a built-in heater electrode (resistive heating element) or a built-in plasma generating electrode (RF electrode).

[0060] In the first and second embodiments described above, the wafer mounting table 10, 110 may have a plurality of lift pin holes that vertically penetrate the wafer mounting table 10, 110. The lift pin holes are holes for inserting lift pins that move the wafer W up and down relative to the wafer mounting surface 22. For example, when the wafer mounting surface 22 is viewed from above, a plurality of lift pin holes are provided at equal intervals along concentric circles of the wafer mounting surface 22.

[0061] The present invention can be used, for example, in an apparatus for plasma processing a wafer.

[0062] 10, 110 wafer mounting table, 20 ceramic plate, 22 wafer mounting surface, 22a outer peripheral region, 22b central region, 23 electrostatic electrode, 24 focus ring mounting surface, 26 power supply terminal, 30 cooling plate, 32, 132 coolant flow path, 32in, 132in inlet, 32mid, 132mid midway position, 32out, 132out outlet, 32x, 132x first portion, 32y, 132y second portion, 32div, 132div branch point, 32join, 132join junction point, 32turn turn-back portion, 40 bonding layer, 50 gas hole, 56 terminal hole, 60 focus ring, 62 circumferential groove, W wafer.

Claims

1. a ceramic plate having a wafer mounting surface on an upper surface thereof; a cooling plate provided on a lower surface of the ceramic plate; A refrigerant flow path built into the cooling plate; A wafer mounting table comprising: The coolant flow path has a first portion and a second portion that branches into two or more branches from the first portion and runs parallel to each other, a cross-sectional area of ​​the first portion is smaller than the sum of the cross-sectional areas of each branch of the second portion; Wafer placement stage.

2. the first portion is disposed in correspondence with an outer periphery region of the wafer mounting surface, and the second portion is disposed in correspondence with a central region of the wafer mounting surface, The wafer stage according to claim 1 .

3. the ceramic plate has an annular focus ring mounting surface around the wafer mounting surface, the focus ring mounting surface being one step lower than the height of the wafer mounting surface, and an annular focus ring having an outer diameter larger than an outer diameter of the ceramic plate and an outer diameter of the cooling plate is mounted on the focus ring mounting surface. The wafer stage according to claim 2 .

4. a cross-sectional area of ​​each branch of the second portion is greater than half the cross-sectional area of ​​the first portion; The wafer mounting table according to any one of claims 1 to 3.

5. The refrigerant flow path has a curved turn-back portion that reverses the direction of the flow path, and branches into two at the midpoint of the curve of the turn-back portion to suppress bias in the amount of refrigerant distributed to each branch in the second portion. The wafer mounting table according to any one of claims 1 to 3.

6. The refrigerant flow path is curved before branching, and is branched into a branch that continues the tendency of curvature before branching and a branch that deviates from the tendency of curvature to an outside of the curvature, thereby suppressing bias in the amount of refrigerant distributed to each branch in the second portion. The wafer mounting table according to any one of claims 1 to 3.

7. In a plan view, in a region in which the second portion is disposed, a non-flow path area, which is an area of ​​a portion in which the refrigerant flow path is not formed, is 50% or more. The wafer mounting table according to any one of claims 1 to 3.