Method for manufacturing glass substrate, and glass substrate

Abstract

To suppress dimensional accuracy reduction.SOLUTION: A method for manufacturing a glass substrate is a method for manufacturing a glass substrate supporting a semiconductor device, where a glass base plate is created, the thickness, thickness deviation and warpage of the glass base plate are measured, the glass base plate is selected on the basis of the glass base plate, a selected glass base plate is cut off to create a plurality of glass blanks, a first polishing condition of the glass blank is set on the basis of the thickness, thickness deviation and warpage of the glass base plate, the surface of the glass blank is polished to create glass plate on the basis of the first polishing condition, the thickness, thickness deviation and warpage of the glass plate are measured, a glass plate is selected on the basis of the thickness of the glass plate, a second polishing condition of the glass plate is set on the basis of the thickness, thickness deviation and warpage of the glass plate, and the surface of the selected glass plate is polished on the basis of the second polishing condition to create a rectangular glass substrate having 300 mm or more of the length of a side and 0.5 mm or more of the thickness.SELECTED DRAWING: Figure 6

Description

[Technical Field]

[0001] The present invention relates to a method for manufacturing a glass substrate and a glass substrate. [Background technology]

[0002] During the manufacturing process of semiconductor devices, glass substrates are sometimes used as members for supporting semiconductor devices. For example, Patent Document 1 describes a glass support substrate for fan-out wafer-level packaging.

[0003] [Patent Document 1] Patent No. 6443668 Summary of the Invention [Problem to be solved by the invention]

[0004] Glass substrates that support semiconductor devices require dimensional precision equivalent to that of silicon wafers in terms of thickness and warpage. However, because glass substrates that support semiconductor devices are multi-component and are larger in thickness and area than silicon wafers, it can be difficult to ensure high dimensional precision. Therefore, there is a need to prevent a decrease in dimensional precision.

[0005] The present invention has been made in view of the above-mentioned problems, and an object of the present invention is to provide a glass substrate manufacturing method and a glass substrate that can suppress a decrease in dimensional accuracy. [Means for solving the problem]

[0006] In order to solve the above-mentioned problems and achieve the objective, the manufacturing method of a glass substrate according to the present disclosure is a manufacturing method of a glass substrate for supporting a semiconductor device, which includes producing a glass mother plate, measuring the thickness, thickness deviation, and warpage of the glass mother plate, selecting the glass mother plate based on the thickness of the glass mother plate, cutting the selected glass mother plate to produce a plurality of glass raw plates, setting first polishing conditions for the glass raw plate based on the thickness, thickness deviation, and warpage of the glass mother plate, polishing the surface of the glass raw plate based on the first polishing conditions to produce a glass plate, measuring the thickness, thickness deviation, and warpage of the glass plate, selecting the glass plate based on the thickness of the glass plate, setting second polishing conditions for the glass plate based on the thickness, thickness deviation, and warpage of the glass plate, and polishing the surface of the selected glass plate based on the second polishing conditions to produce a rectangular glass substrate with a side length of 300 mm or more and a thickness of 0.5 mm or more.

[0007] In order to solve the above-mentioned problems and achieve the object, the glass substrate of the present disclosure is a glass substrate for supporting a semiconductor device, having a rectangular shape with a side length of 300 mm or more and a thickness of 0.7 mm or more, a warp of the glass substrate of 1 mm or less, a plate thickness of the glass substrate of 0.5 mm or more and 4.0 mm or less, a thickness deviation of the glass substrate of 5 μm or less, an LTV (Local Thickness Variation) of the glass substrate in a 50 mm x 50 mm area of ​​2 μm or less, a width of the edge surface of the glass substrate of 1 mm or less, and a radius of curvature formed by any three points in the area with the smallest curvature in the boundary area between the edge surface and the end face of the glass substrate of 0.05 mm or more. [Effects of the Invention]

[0008] According to the present invention, it is possible to suppress a decrease in dimensional accuracy. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a schematic diagram of a glass substrate according to this embodiment. [Figure 2A] FIG. 2A is a cross-sectional view taken along line AA in FIG. [Figure 2B] FIG. 2B is an enlarged view of a portion of FIG. 2A. [Figure 3] FIG. 3 is a schematic diagram for explaining the bending of the glass substrate according to this embodiment. [Figure 4] FIG. 4 is a schematic diagram for explaining the bending of the glass substrate according to this embodiment. [Figure 5] FIG. 5 is a schematic diagram illustrating the method for manufacturing a glass substrate according to this embodiment. [Figure 6] FIG. 6 is a flowchart illustrating the method for manufacturing a glass substrate according to this embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0010] Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Note that the present invention is not limited to these embodiments, and when there are multiple embodiments, the present invention also includes configurations that combine the embodiments. Furthermore, numerical values ​​include ranges of rounding and general tolerances.

[0011] (glass substrate) FIG. 1 is a schematic diagram of a glass substrate according to this embodiment, and FIG. 2A is a cross-sectional view taken along line AA in FIG. 1. A glass substrate 10 according to this embodiment is used as a glass substrate for manufacturing semiconductor packages, and can be said to be a glass substrate for supporting semiconductor devices. More specifically, the glass substrate 10 is a supporting glass substrate for manufacturing fan-out panel level packages (FOPLPs). However, the use of the glass substrate 10 is not limited to supporting semiconductor devices or manufacturing FOPLPs, and can be any use, and the glass substrate may be a glass substrate used to support any member.

[0012] As shown in FIG. 1, glass substrate 10 is a plate-like member having surface 10A (one surface) as one of its main surfaces and surface 10B (the other surface) as the main surface opposite surface 10A. Glass substrate 10 is rectangular in plan view, i.e., when viewed from a direction perpendicular to surface 10A. Furthermore, glass substrate 10 is square in plan view, but is not limited to a square shape and may be rectangular in plan view. Glass substrate 10 may have a notch formed in end surface 10C, resulting in a rectangular shape with a partial cutout at the periphery. Hereinafter, the direction perpendicular to surface 10A will be referred to as the Z direction. The Z direction can also be referred to as the thickness direction of glass substrate 10. The direction perpendicular to the Z direction will be referred to as the X direction, and the direction perpendicular to both the Z direction and the X direction will be referred to as the Y direction.

[0013] As shown in FIG. 2A, edge surface 10C of glass substrate 10 is a surface connecting surface 10A and surface 10B, and can also be called a side surface of glass substrate 10. Glass substrate 10 has edge surface 10C chamfered. Specifically, edge surface 10C includes edge surface portion 10C1 and edge surface portion 10C2. Edge surface portion 10C1 corresponds to the non-chamfered portion of edge surface 10C, and edge surface portion 10C2 corresponds to the chamfered portion of edge surface 10C. Note that edge surface 10C has a shape including edge surface portion 10C1 and edge surface portion 10C2 as shown in FIG. 2A over the entire circumferential area of ​​glass substrate 10. In other words, since the glass substrate 10 is rectangular, it can be said that end surface portions 10C1 and edge surface portions 10C2 are formed on the end surfaces 10C on two sides of the glass substrate 10 along the X direction and on the end surfaces 10C on two sides of the glass substrate 10 along the Y direction (i.e., on all four sides).

[0014] The end surface portion 10C1 is a surface of the end surface 10C that includes a portion that protrudes most radially outward from the glass substrate 10. As shown in Fig. 2A, the end surface portion 10C1 is aligned with the Z direction when viewed from a direction perpendicular to the Z direction.

[0015] The edge surface portion 10C1 is connected to the surface of the glass substrate 10 via the edge surface portion 10C2. That is, one end of the edge surface portion 10C2 in the Z direction is connected to the edge surface portion 10C1, and the other end is connected to the surface of the glass substrate 10. The edge surface portions 10C2 are formed on both sides of the edge surface portion 10C1 in the Z direction. That is, the edge surface 10C is formed so that one edge surface portion 10C2, the edge surface portion 10C1, and the other edge surface portion 10C2 are arranged in this order in the Z direction. One end of one edge surface portion 10C2 in the Z direction is connected to the surface 10A of the glass substrate 10, and the other end is connected to the edge surface portion 10C1. Furthermore, the other edge surface portion 10C2 has one end of the other edge surface portion 10C2 in the Z direction connected to the edge surface portion 10C1, and the other end is connected to the surface 10B of the glass substrate 10.

[0016] 2A, the edge surface portion 10C2 is inclined with respect to the Z direction when viewed from a direction perpendicular to the Z direction. The edge surface portion 10C2 is inclined from the end surface portion 10C1 toward the front surface of the glass substrate 10 so as to be inclined radially inward of the glass substrate 10. The edge surface portion 10C2 is linear when viewed from a direction perpendicular to the Z direction, but may be curved (R-shaped).

[0017] As described above, the end face 10C has a chamfered shape including the end face portion 10C1 and the edge face portion 10C2, but the shape of the end face 10C is not limited to this, and for example, the end face 10C may not be chamfered.

[0018] (length of one side of glass substrate) The length L of a side of the glass substrate 10 is preferably 300 mm or more, and more preferably 500 mm or more. The length L of a side of the glass substrate 10 is preferably 1000 mm or less, and more preferably 700 mm or less. By keeping the length L within this range, the size of the area in which the semiconductor device is to be disposed can be maintained sufficiently, and the semiconductor device can be manufactured appropriately. The length L is the length of one side of the glass substrate 10 when viewed from the Z direction, and refers to the length from an end face 10C (more specifically, end face 10C1) of one side of the glass substrate 10 to an end face 10C (more specifically, end face 10C1) of the opposing side.

[0019] (thickness of glass substrate) The thickness D of the glass substrate 10 is preferably 0.5 mm or more, more preferably 0.7 mm or more, and even more preferably 1.0 mm or more. The thickness D of the glass substrate 10 is preferably 4.0 mm or less, and more preferably 2.0 mm or less. Having the thickness D within this range suppresses warpage, allowing semiconductor devices to be manufactured appropriately. The thickness D refers to the length in the Z direction from the surface 10A to the surface 10B.

[0020] Furthermore, when a plurality of glass substrates 10 are manufactured by the manufacturing method according to the present embodiment described below, the difference in thickness D between the glass substrates 10 is preferably 20 μm or less, and more preferably 5 μm or less. When the difference in thickness D between the glass substrates 10 (variation in thickness D between the glass substrates 10) falls within this range, deterioration in dimensional accuracy when manufacturing semiconductor devices can be suppressed.

[0021] Furthermore, the deviation in thickness D of the glass substrate 10 is preferably 10 μm or less, more preferably 5 μm or less, and even more preferably 3 μm or less. When the deviation in thickness D is within this range, the thickness D of the glass substrate 10 approaches uniformity, making it possible to properly manufacture semiconductor devices. The deviation in thickness D refers to the deviation in thickness D for each position (each coordinate) on a plane along the surface of the glass substrate 10. For example, the thickness D at each position (coordinate) on a plane along the surface of the glass substrate 10 may be calculated, and the difference between the maximum and minimum values ​​of the thickness D for each position may be taken as the deviation in thickness D.

[0022] Furthermore, the LTV (Local Thickness Variation) of the glass substrate 10 within a 50 mm × 50 mm area is preferably 2 μm or less, and more preferably 1 μm or less. The LTV within a 50 mm × 50 mm area refers to the difference between the maximum and minimum values ​​of the thickness D within a 50 mm × 50 mm unit area of ​​the glass substrate 10. In other words, while the deviation of the thickness D is the difference between the maximum and minimum values ​​of the thickness D across the entire area of ​​the glass substrate 10, the LTV refers to the difference between the maximum and minimum values ​​of the thickness D within a unit area of ​​the glass substrate 10.

[0023] The outer region is defined as the region of the entire glass substrate 10 between a position 1 mm radially inward from the periphery and a position 5 mm radially inward from the periphery. The central region is defined as the region of the entire glass substrate 10 surrounded by a square with sides of 100 mm and centered at the center point O of the glass substrate 10. In this case, the average thickness D of the glass substrate 10 in the outer region may be larger than the average thickness D of the glass substrate 10 in the central region (i.e., the center may be thicker). Conversely, the average thickness D of the glass substrate 10 in the outer region may be smaller than the average thickness D of the glass substrate 10 in the central region (i.e., the center may be thinner). The deviation of the thickness D in the central region from the thickness D in the outer region is preferably 2 μm or less, more preferably 1 μm or less. To achieve such a shape with a thinner or thicker center, for example, physical processing such as polishing may involve increasing the pressure on the center of the glass substrate 10 or the relative speed of the polishing cloth. For example, in HF (hydrofluoric acid) etching, in order to selectively etch the center of the glass substrate 10, it is possible to mask the outer periphery, heat the center of the glass substrate 10, or adjust the chemical flow path so that fresh chemical always hits the center of the glass substrate 10.

[0024] (deflection of glass substrate) 3 and 4 are schematic diagrams illustrating deflection of the glass substrate according to this embodiment. FIG. 3 illustrates an example of deflection due to the weight of the glass substrate 10 when supported by a support member B at three points: a first position P1A, a second position P2A, and a third position P3A, which are radially outward from the center point O of the glass substrate 10 as viewed from the Z direction. The first position P1A is a position on the surface 10A that is a distance L1A away from the center point O of the glass substrate 10 in the radial direction. The second position P2A is a position on the surface 10A that is a distance L2A away from the center point O of the glass substrate 10 in the radial direction and is circumferentially offset by 120 degrees from the first position P1A when the center point O is the center. The third position P3A is a position on the surface 10A that is a distance L3A away from the center point O of the glass substrate 10 in the radial direction and is circumferentially offset by 120 degrees from the first position P1A and the second position P2A when the center point O is the center. Here, the radial direction refers to the radial direction when the center point O is the center. Distances L1A, L2A, and L3A are all the same length. For example, distance L1A is any length that is equal to or greater than half the length of a straight line from center point O through first position P1A to the peripheral edge of glass substrate 10, distance L2A is any length that is equal to or greater than half the length of a straight line from center point O through second position P2A to the peripheral edge of glass substrate 10, and distance L3A is any length that is equal to or greater than half the length of a straight line from center point O through third position P3A to the peripheral edge of glass substrate 10. 3, the length from the first position P1A to the peripheral edge of the glass substrate 10 along the straight line passing from the center point O to the first position P1A is 10 mm, the length from the second position P2A to the peripheral edge of the glass substrate 10 along the straight line passing from the center point O to the second position P2A is 10 mm, and the length from the third position P3A to the peripheral edge of the glass substrate 10 along the straight line passing from the center point O to the third position P3A is 10 mm. Furthermore, the support member B is a spherical member with a diameter of 1 mm or more and 2 mm or less, and a diameter of 1.6 mm is more preferable.Furthermore, it is preferable that the material of the support member B is softer than glass and is less likely to deform due to temperature, humidity, or the weight of the glass during measurement, such as a resin such as PEEK (Polyetheretherketone) or PTFE (Polytetrafluoroethylene), and PEEK may be used here.

[0025] As shown in FIG. 3, when the surface 10A of the glass substrate 10 faces downward in the vertical direction and is supported by a support member B at a first position P1A, a second position P2A, and a third position P3A on the surface 10A, the lowest point SB1 is defined as the position on the surface 10B on the upper side in the vertical direction where the height in the vertical direction is lowest. In other words, the lowest point SB1 is the position on the surface 10B where the amount of deflection is greatest. In this case, the position of the lowest point SB1 is preferably located within a central region AR when viewed from the vertical direction (Z direction). The central region AR is a region radially inward from the first position P1A, the second position P2A, and the third position P3A. More specifically, the central region AR is a circular region whose center is the center point O and whose diameter Da is 1 / 3 of the diameter W of the glass substrate 10.

[0026] In this way, it is preferable that the lowest point SB of glass substrate 10 be located within central region AR when front surface 10A is facing vertically downward and front surface 10A is supported at three points by support members B at first position P1A, second position P2A, and third position P3A. By locating lowest point SB of glass substrate 10 within central region AR rather than outside central region AR, deformation of glass substrate 10 that shifts the bending position during the manufacture of semiconductor devices, for example, can be suppressed, and deterioration in the manufacturability of semiconductor devices can be suppressed.

[0027] Furthermore, the maximum deflection amount TBmax is the maximum amount of deflection of glass substrate 10 when surface 10A of glass substrate 10 faces vertically downward and a first position P1A, a second position P2A, and a third position P3A on surface 10A are supported by support member B. Maximum deflection amount TBmax can be said to be the maximum amount of deflection due to the weight of glass substrate 10, and can be said to be the distance along the vertical direction from the highest point SB2 to the lowest point SB1 of glass substrate 10. Note that highest point SB2 is the position on surface 10B on the vertically upper side that has the highest vertical height when surface 10A of glass substrate 10 faces vertically downward and a first position P1A, a second position P2A, and a third position P3A on surface 10A are supported by support member B. In this case, when the length L of the glass substrate 10 is 300 mm or more and 1000 mm or less and the thickness D is 4.0 mm or less, the maximum deflection TBmax is preferably 10 mm or less, and more preferably 0 mm or more and 5 mm or less. By keeping the maximum deflection TBmax within this range, the amount of deflection is also reduced, and therefore, deterioration in the manufacturability of semiconductor devices can be more suitably prevented.

[0028] 3 illustrates an example in which the glass substrate 10 is simply warped, but the manner in which the glass substrate 10 is warped is not limited to the simple warp shown in Fig. 3, and the glass substrate 10 may be warped in a manner as shown in Fig. 4, for example. That is, in Fig. 3, the glass substrate 10 is warped to have a concave shape that is concave from the outer circumferential edge toward the lowest point SB1, but as shown in Fig. 4, the glass substrate 10 may be warped so as to be concave from the point indicated by the support member B toward the lowest point SB1 and also to be concave from the point supported by the support member B toward the outer circumferential edge.

[0029] The above explanation has been given regarding the deflection of the glass substrate 10 when the surface 10A faces vertically downward, but below, it is preferable that the same applies to the deflection of the glass substrate 10 when the surface 10B faces vertically downward.

[0030] (Warpage of glass substrate) Here, the amount of warpage of the glass substrate 10 excluding the warpage due to its own weight is defined as the warpage ΔT. In this case, the warpage ΔT is preferably 1.0 mm or less, more preferably 0.5 mm or less, and even more preferably 0.1 mm or less. By keeping the warpage ΔT within this range, it is possible to suppress warpage of semiconductor devices manufactured on the glass substrate 10, thereby more preferably suppressing deterioration in the manufacturability of the semiconductor devices. The warpage ΔT is calculated by dividing the maximum value of the difference in the amount of warpage at each position (each coordinate) on a plane along the surface of the glass substrate 10 between the case where the surface 10B is supported by the support member B as described above with the surface 10B facing vertically downward and the case where the surface 10A is supported by the support member B as described above with the surface 10B facing vertically downward by 2. That is, the amount of warpage at position (coordinate) i on a plane along the surface of the glass substrate 10 when the surface 10B is supported by the support member B as described above with the surface 10B facing vertically downward is defined as TA (i) The amount of deflection at a position (coordinate) i on a plane along the surface of the glass substrate 10 when the surface 10A is supported by the support member B as described above with the surface 10A facing vertically downward is defined as TB (i) Then, the deflection amount TA for each position i is (i) and deflection amount TB (i) The maximum difference between these values ​​is called MAX(TA (i) -TB (i) In this case, the amount of warpage ΔT is calculated by the following formula (1). (i) -TB (i) ) divided by 2.

[0031] ΔT=|MAX(TA (i) -TB (i) )| / 2 ···(1)

[0032] (Glass substrate edge shape) 2B is a partial enlarged view of FIG. 2A. As shown in FIG. 2B, end surface portion 10C1 includes straight portion 10C1a and curved portion 10C1b. More specifically, end surface portion 10C1 has curved portions 10C1b on both sides of straight portion 10C1a in the Z direction, and is connected to edge surface portion 10C2 via curved portions 10C1b. That is, the end surface of glass substrate 10 is configured in the Z direction in the following order: edge surface portion 10C2 on the surface 10B side, curved portion 10C1b on the surface 10B side, straight portion 10C1a on the surface 10A side, curved portion 10C on the surface 10A side, and edge surface portion 10C2 on the surface 10A side.

[0033] The straight line portion 10C1a is a region where the trace of the edge surface 10C1 along the Z direction can be considered a straight line, while the curved line portion 10C1b is a region where the trace of the edge surface 10C1 along the Z direction is considered a curve. The curved line portion 10C1b has an R-shape that slopes radially inward as it moves from the center of the glass substrate 10 toward the outside (surface 10A or surface 10B) in the Z direction. More specifically, the center point of the edge surface 10C in the Z direction is defined as midpoint 10C1P. When the edge surface 10C is traced from midpoint 10C1P toward the Z direction (surface 10A), the position where the traced trace moves 10 μm radially inward of the glass substrate 10 (in the X direction in the example of FIG. 2B ), excluding unevenness in the surface finish, is defined as boundary position 10C1A. When end face 10C is traced from midpoint 10C1P toward the opposite side in the Z direction (toward surface 10B), the position where the traced path moves 10 μm toward the inside in the radial direction of glass substrate 10 (in the X direction in the example of FIG. 2B), excluding unevenness in the surface finish, is defined as boundary position 10C1B. That is, the distance between boundary position 10C1A and midpoint 10C1P in the radial direction of glass substrate 10 and the distance between boundary position 10C1B and midpoint 10C1P in the radial direction of glass substrate 10 are both 10 μm. In this case, the region of end face 10C1 between boundary position 10C1A and boundary position 10C1B is linear portion 10C1a, and the region on the Z direction side of boundary position 10C1A (toward surface 10A) and the region on the opposite side in the Z direction of boundary position 10C1B (toward surface 10B) are curved portion 10C1b. That is, the trajectory of the straight line portion 10C1a in the Z direction can be considered a straight line because the radial displacement of the glass substrate 10 is small, but the trajectory of the curved line portion 10C1b in the Z direction is a curved line that displaces the glass substrate 10 in the radial direction.

[0034] The boundary position between curved portion 10C1b and edge surface portion 10C2 is referred to as boundary position 10C2A. That is, the region from boundary position 10C1A to boundary position 10C2A is curved portion 10C1b, and the region from boundary position 10C2A to boundary position 10C2B between edge surface portion 10C2 and surface 10A (surface 10B) is edge surface portion 10C2. Boundary position 10C2A can be defined as follows: A position on end surface 10C that is 10 μm away from edge surface 10C2 in the direction perpendicular to a line inclined at angle θ with respect to surface 10A (i.e., a line along edge surface 10C2) can be referred to as boundary position 10C2A. That is, the trajectory traced along the edge surface portion 10C2 in the Z direction can be considered a straight line because the radial displacement of the glass substrate 10 is small, but the trajectory traced along the curved portion 10C1b in the Z direction is a curved line that displaces the glass substrate 10 in the radial direction.

[0035] In this way, the glass substrate 10 has a curved portion 10C1b, which is rounded, formed between the straight portion 10C1a and the edge surface portion 10C2. Such a shape can be achieved, for example, by adjusting the shape and grit of the grindstone. However, the glass substrate 10 does not necessarily have to include the curved portion 10C1b. In this case, the straight portion 10C1a and the edge surface portion 10C2 may be directly connected.

[0036] (length of straight section) The length D1 of the straight portion 10C1a in the Z direction is preferably 50% or less of the thickness D of the glass substrate 10, and preferably 5% to 45%. By keeping the length D1 within this range, cracking from the end face 10C can be suppressed, enabling the appropriate manufacture of semiconductor devices. Furthermore, by ensuring that the length D1 is not too small relative to the thickness D, it is possible to prevent the end from becoming sharp and vulnerable to contact with foreign matter. By ensuring that the length D1 is not too large relative to the thickness D, it is possible to prevent the chamfered portion from being almost eliminated or the angle from becoming very small, thereby preventing the boundary on the end face from becoming susceptible to cracking when contact with foreign matter occurs. Note that the length D1 refers to the length in the Z direction from the boundary position 10C1A to the boundary position 10C1B. Note that the length D1 can be measured, for example, by capturing a cross-sectional image of the vicinity of the end face 10C using image capture or contact measurement.

[0037] (Angle of edge surface) As shown in FIG. 2A, the angle θ of the edge surface portion 10C2 is preferably 15° to 80°, and more preferably 20° to 60°. By keeping the angle θ within this range, cracking from the end surface 10C can be suppressed, enabling proper manufacturing of semiconductor devices. The angle θ refers to the angle between a line along the edge surface portion 10C2 and a line along the surface of the glass substrate 10 (surface 10A in the example of FIG. 2A) when viewed from a direction perpendicular to the Z direction. The angle θ of the edge surface portion 10C2 on the Z direction side (surface 10A side) and the angle θ of the edge surface portion 10C2 on the opposite side of the Z direction (surface 10B side) are preferably the same, but are not limited thereto and may be different values ​​for purposes such as front / back identification. The angle θ can be measured, for example, by capturing a cross-sectional image of the area near the end surface 10C using image capture or contact measurement.

[0038] Furthermore, the deviation of the angle θ of the edge surface portion 10C2 is preferably 20 degrees or less, and more preferably 10 degrees or less. By keeping the deviation of the angle θ within this range, cracks from the end surface 10C can be suppressed, enabling proper manufacturing of semiconductor devices. The deviation of the angle θ refers to the deviation of the angle θ at each position on the edge surface portion 10C2, and is the deviation of the difference angle θ between the maximum and minimum values ​​of the angle θ at each position. In this case, for example, the angle θ at each position on the edge surface portion 10C2 is calculated. More specifically, since the glass substrate 10 of this embodiment is rectangular, the angle θ is calculated for each position along the X direction for the edge surface portion 10C2 on the side along the X direction, and the angle θ is calculated for each position along the Y direction for the edge surface portion 10C2 on the side along the Y direction. The difference between the maximum and minimum values ​​of the angle θ at each position may then be used as the deviation of the angle θ.

[0039] (Width of the edge surface) As shown in FIG. 2A, the width W of the edge surface portion 10C2 is preferably 1 mm or less, and more preferably 0.1 mm to 0.5 mm. By keeping the width W within this range, cracks from the end surface 10C can be suppressed, enabling proper manufacturing of semiconductor devices. The width W refers to the length of the edge surface portion 10C2 in a direction perpendicular to the Z direction, from the boundary position 10C2A to the boundary position 10C2B. The width W can be measured, for example, by capturing a cross-sectional image of the vicinity of the end surface 10C using image capture or contact measurement.

[0040] Furthermore, the deviation in width W of the edge surface portion 10C2 is preferably 0.2 mm or less, and more preferably 0.1 mm or less. By keeping the deviation in width W within this range, cracks from the end surface 10C can be suppressed, enabling proper manufacturing of semiconductor devices. The deviation in width W refers to the deviation in width W at each position on the edge surface portion 10C2. The deviation in width W is calculated by dividing the difference between the maximum and minimum values ​​of the width W at each position by the minimum value of width W. In this case, for example, the width W at each position on the edge surface portion 10C2 is calculated. More specifically, since the glass substrate 10 of this embodiment is rectangular, the width W is calculated for each position along the X direction for the edge surface portion 10C2 on the side along the X direction, and the width W is calculated for each position along the Y direction for the edge surface portion 10C2 on the side along the Y direction. The deviation in width W may then be calculated by dividing the difference between the maximum and minimum values ​​of the width W at each position by the minimum value of width W.

[0041] (shape of curved part) Curved portion 10C1b, which is the boundary between edge portion 10C2 and linear portion 10C1a, preferably has an R-shape. The radius of curvature of curved portion 10C1b is preferably 0.01 mm or more, more preferably 0.05 mm or more, and preferably 0.5 mm or less. By keeping the radius of curvature of curved portion 10C1b within this range, cracks from end face 10C can be suppressed, allowing semiconductor devices to be properly manufactured. The radius of curvature of curved portion 10C1b can be measured, for example, by capturing a cross-sectional image of the vicinity of end face 10C using image photography or contact measurement. As an example, in a cross-sectional image of the end face 10C viewed from a direction perpendicular to the Z direction, the radius of curvature of the curved portion 10C1b may be determined as the radius of curvature formed by any three points in the region of the boundary between the edge portion 10C2 and the straight portion 10C1a (i.e., the curved portion 10C1b) with the smallest curvature, excluding the effects of sudden convex or concave points and irregularities that can be considered within the range of surface roughness. Similarly, the boundary between the edge portion 10C2 and the surface of the glass substrate 10 (surfaces 10A and 10B) may also be rounded. In this case, the radius of curvature of the rounded shape of the boundary between the edge portion 10C2 and the surface of the glass substrate 10 is preferably 0.05 mm or more, more preferably 0.1 mm or more and less than half the thickness. Preferably, it is 0.5 mm or less. When the radius of curvature of curved portion 10C1b is within this range, cracks from end surface 10C are suppressed, and semiconductor devices can be appropriately manufactured.

[0042] (surface roughness of glass substrate surface) The arithmetic mean roughness Ra of the surfaces (surfaces 10A and 10B) of the glass substrate 10, as specified in JIS B 0601:2001, is preferably 5 nm or less, and more preferably 0.5 nm to 2.0 nm. A surface roughness within this range allows for proper manufacture of semiconductor devices. The arithmetic mean roughness Ra of the surface of the glass substrate 10 is calculated by sampling a reference length of the surface roughness curve. The reference length may be, for example, 5 μm. The arithmetic mean roughness Ra can be measured using a KEYENCE AFM.

[0043] (surface roughness of the edge of the glass substrate) The arithmetic mean roughness Ra of the edge surface 10C1 of the glass substrate 10, as specified in JIS B 0601:2001, is preferably 5 μm or less, and more preferably 1 μm or less. Having a surface roughness within this range suppresses cracking from the edge surface 10C1, enabling semiconductor devices to be manufactured appropriately. The arithmetic mean roughness Ra of the surface of the edge surface 10C1 is calculated by extracting a reference length from the surface roughness curve. The reference length may be, for example, 128 μm.

[0044] The arithmetic mean roughness Ra of the edge surface 10C2 of the glass substrate 10, as specified in JIS B 0601:2001, is preferably 5 μm or less, and more preferably 1 μm or less. A surface roughness within this range suppresses cracking from the edge surface 10C2, enabling semiconductor devices to be manufactured appropriately. The arithmetic mean roughness Ra of the edge surface 10C2 is calculated by extracting a reference length from the surface roughness curve. The reference length may be, for example, 128 μm.

[0045] (Preferred shape of glass substrate) For example, glass substrate 10 preferably has a thickness D of 1.5 mm, a width W of edge surface 10C2 of 0.2 mm, an angle θ of edge surface 10C2 of 25°, a length D1 of straight portion 10C1a of 1.3 mm, a radius of curvature of curved portion 10C1b of 0.2 mm, an arithmetic mean roughness Ra of edge surface 10C2 of 0.03 μm, and an arithmetic mean roughness Ra of the surface of 0.4 nm. This reduces the surface roughness, improves strength, and prevents yield reduction due to glass breakage and process contamination due to glass fragments being mixed in.

[0046] For example, the glass substrate 10 preferably has a thickness D of 1.5 mm, a width W of the edge surface 10C2 of 0.2 mm, an angle θ of the edge surface 10C2 of 30°, a length D1 of the straight portion 10C1a of 1.2 mm, a radius of curvature of the curved portion 10C1b of 0.2 mm, an arithmetic mean roughness Ra of the edge surface 10C2 of 0.40 μm, and an arithmetic mean roughness Ra of the surface of 0.9 nm. The high surface roughness of the glass main surface can prevent film peeling during the film formation process. This improves the detectability of the edge.

[0047] For example, the glass substrate 10 preferably has a thickness D of 1.5 mm, a width W of the edge surface 10C2 of 0.1 mm, an angle θ of the edge surface 10C2 of 45°, a length D1 of the straight line portion 10C1a of 1.1 mm, a radius of curvature of the curved portion 10C1b of 0.3 mm, an arithmetic mean roughness Ra of the edge surface 10C2 of 0.40 μm, and an arithmetic mean roughness Ra of the surface of 0.9 nm. This increases the area of ​​the main surface and maximizes the number of semiconductor devices mounted on the glass substrate.

[0048] (Young's modulus of glass substrate) The Young's modulus of the glass substrate 10 is preferably 50 GPa or more, more preferably 70 GPa or more, and even more preferably 75 GPa or more. The Young's modulus of the glass substrate 10 is preferably 150 GPa or less, more preferably 130 GPa or less, and even more preferably 120 GPa or less. Having the Young's modulus within this range can prevent a decrease in the rigidity of the glass substrate and suppress deformation of the glass substrate, thereby suppressing deterioration in the manufacturability of semiconductor devices. The Young's modulus of the glass substrate 10 may be a value measured based on ultrasonic propagation using an Olympus 38DL PLUS.

[0049] (Average thermal expansion coefficient of glass substrate) The glass substrate 10 preferably has an average coefficient of thermal expansion (CTE) in the range of 50°C to 200°C of 3 ppm / °C to 12 ppm / °C, more preferably 3.2 ppm / °C to 10 ppm / °C. Having an average coefficient of thermal expansion (CTE) in this range can prevent breakage of the glass substrate 10. The average coefficient of thermal expansion (CTE) can be measured in accordance with DIN-51045-1, a standard for thermal expansion measurement. Specifically, a sample is measured in the range of 30 to 300°C using a dilatometer (DIL 402 Expedis) manufactured by NETZSCH, and the average coefficient of thermal expansion in the range of 50 to 200°C can be used as the average coefficient of thermal expansion (CTE).

[0050] (Light transmittance of glass substrate) The glass substrate 10 preferably has an average transmittance of 50% or more, more preferably 60% or more, for light with a wavelength of 300 nm or more and 1100 nm or less. The glass substrate 10 also preferably has an average transmittance of 80% or more, more preferably 90% or more, for light with a wavelength of 500 nm or more and 1100 nm or less. Semiconductor devices can be manufactured appropriately when the transmittance for light in this wavelength range is within this range. The transmittance can be measured by measuring a spectral transmittance curve using, for example, a UV-visible spectrophotometer (U-4150 model, manufactured by Hitachi High-Technologies Corporation). The average transmittance is the average value of the transmittance for light of each wavelength in the wavelength range (here, 300 nm to 1100 nm or 500 nm to 1000 nm).

[0051] (Glass substrate composition) The glass substrate 10 preferably contains the following compounds in terms of mass % (wt %) based on oxides: By making the glass substrate 10 have the following composition, it is possible to properly support members. SiO2: 40 wt% or more and 75 wt% or less is preferable, and 50 wt% or more and 75 wt% or less is more preferable. Al2O3: preferably 0 wt% or more and 20 wt% or less, more preferably 0 wt% or more and 15 wt% or less B2O3: preferably 0 wt% or more and 20 wt% or less, more preferably 0 wt% or more and 10 wt% or less MgO: 0 wt% to 25 wt% is preferable CaO: preferably 0 wt% or more and 25 wt% or less, and more preferably 0 wt% or more and 15 wt% or less SrO: preferably 0 wt% or more and 10 wt% or less BaO: preferably 0 wt% or more and 20 wt% or less, more preferably 0 wt% or more and 15 wt% or less Li2O: 0 wt% or more and 40 wt% or less is preferable Na2O: preferably 0 wt% or more and 15 wt% or less K2O: 0 wt% or more and 10 wt% or less is preferable ZrO2: 0 wt% or more and 10 wt% or less is preferable, 0 wt% or more and 8 wt% or less is more preferable, and 0 wt% or more and 5 wt% or less is even more preferable. TiO2: 0 wt% to 5 wt% is preferred Y2O3: 0 wt% to 10 wt% is preferable

[0052] The glass substrate 10 has the above-described structure, and therefore the deterioration of dimensional accuracy is suppressed, making it suitable as a glass substrate for supporting semiconductor devices.

[0053] The glass substrate 10 may have a mark described in WO2018 / 150759 provided at any position on the glass substrate 10.

[0054] (Glass substrate manufacturing method) Next, a method for manufacturing the glass substrate 10 according to this embodiment will be described. Fig. 5 is a schematic diagram illustrating the method for manufacturing the glass substrate according to this embodiment. In this manufacturing method, as shown in Fig. 5, a glass mother plate 10a is prepared, the glass mother plate 10a is cut to produce a raw glass plate 10b, the surface of the glass plate 10b is polished to produce a glass plate 10c, and the surface of the glass plate 10c is further polished while undergoing edge processing, thereby producing the glass substrate 10.

[0055] The method for manufacturing the glass substrate 10 according to this embodiment will be described in more detail below. Fig. 6 is a flowchart illustrating the method for manufacturing the glass substrate according to this embodiment.

[0056] (Glass mother plate production step) As shown in FIG. 6 , in this manufacturing method, a glass mother plate 10a is produced (step S10; glass mother plate production step). The glass mother plate 10a is a plate-shaped glass that serves as a base material for the glass substrate 10. In this embodiment, for example, in the glass mother plate production step, a glass raw material may be vitrified through any glass melting and forming method such as float, fusion, or ingot forming, and, if necessary, mechanical processing such as slicing, to produce the glass mother plate 10a having a desired composition. Furthermore, in the glass mother plate production step, the glass mother plate 10a may be produced in any size, for example, a glass mother plate 10a having a side length of 300 mm to 1000 mm and a thickness of 0.5 mm to 4 mm.

[0057] (Glass substrate measurement step) Next, in this manufacturing method, the thickness, thickness deviation, and warpage of the produced glass mother sheet 10a are measured (step S12; glass mother sheet measuring step). In the glass mother sheet measuring step, for example, the thickness may be measured at each position (coordinate) on a plane along the surface (main surface) of the glass mother sheet 10a, and the average value of the thicknesses at each position may be defined as the thickness of the glass mother sheet 10a. However, the thickness of the glass mother sheet 10a is not limited to the average value of the thicknesses at each position. For example, the thickness at a predetermined position, such as the center position of the glass mother sheet 10a, may be defined as the thickness of the glass mother sheet 10a, or the maximum or minimum value of the thickness at each position may be defined as the thickness of the glass mother sheet 10a. A laser displacement meter may be used to measure the thickness.

[0058] In the glass substrate measuring step, the thickness deviation of the glass substrate 10a is measured. The thickness deviation refers to the deviation of the thickness at each position (each coordinate) on a plane along the surface of the glass substrate 10a. For example, the thickness may be measured at each position (coordinate) on a plane along the surface of the glass substrate 10a, and the difference between the maximum and minimum values ​​of the thickness at each position may be taken as the thickness deviation.

[0059] In the glass substrate measurement step, the amount of warpage of the glass substrate 10a is measured. The amount of warpage of the glass substrate 10a refers to the amount of warpage excluding the amount of warpage due to its own weight. The amount of warpage of the glass substrate 10a is calculated by dividing the maximum value of the difference in the amount of warpage at each position (each coordinate) on a plane along the surface of the glass substrate 10a between a case where one surface of the glass substrate 10a is supported by the support member B in the same manner as described in FIG. 3 and a case where the other surface of the glass substrate 10a is supported by the support member B in the same manner as described in FIG. 3 by two. That is, the amount of warpage at position (coordinate) i on a plane along the surface of the glass substrate 10a when one surface of the glass substrate 10a is supported by the support member B facing vertically downward is defined as TAa. (i) When the other surface of the glass substrate 10a is supported by the support member B facing vertically downward, the deflection amount at a position (coordinate) i on a plane along the surface of the glass substrate 10a is defined as TBa. (i) Then, the deflection amount TAa for each position i is (i) and deflection TBa (i) The maximum difference between these values ​​is called MAX(TAa (i) -TBa (i) In this case, the amount of warpage ΔTa of the glass substrate 10a is calculated by the following formula (2). (i) -TBa (i) ) divided by 2.

[0060] ΔTa = |MAX(TAa (i) -TBa (i) )| / 2 ···(2)

[0061] (Glass mother board sorting step) Next, in this manufacturing method, the glass mother sheets 10a are sorted based on the measured thickness of the glass mother sheets 10a (step S14: glass mother sheet sorting step). That is, in the glass mother sheet sorting step, the multiple glass mother sheets 10a are classified by thickness. For example, in the glass mother sheet sorting step, the range between the upper and lower limit values ​​of the possible thickness of the glass mother sheets 10a is divided into multiple numerical ranges, and the glass mother sheets 10a to be sorted are assigned to a group whose numerical range includes the thickness of the glass mother sheet 10a. In the glass mother sheet sorting step, similar sorting is performed on each of the multiple glass mother sheets 10a, and each of the multiple glass mother sheets 10a is assigned to one of the groups. Note that if the thickness of the glass mother sheet 10a falls outside the range between the upper and lower limit values, it may be rejected as a defective product. By sorting the glass mother sheets 10a based on their thickness in this way, for example, the polishing allowance can be determined for each group, thereby reducing thickness variation among the manufactured glass substrates 10.

[0062] As described above, in this embodiment, the glass mother sheets 10a are selected using the thickness among the measured thickness, thickness deviation, and warpage. However, the present invention is not limited to this, and the glass mother sheets 10a may also be selected using the thickness deviation and warpage. That is, in the glass mother sheet selection step, the glass mother sheets 10a may be selected based on at least one of the thickness, thickness deviation, and warpage. Note that the selection based on the thickness deviation and warpage may also be performed in a manner similar to the selection based on the thickness. That is, for example, the ranges between the upper and lower limits of the thickness deviation and warpage are divided into multiple numerical ranges, and the glass mother sheets 10a to be selected are assigned to a group of numerical ranges that include the thickness deviation and warpage of the glass mother sheets 10a.

[0063] (Glass plate production step) Next, in this manufacturing method, the glass substrate 10a is cut to produce glass substrates 10b (step S16; glass substrate production step). In this step, the glass substrate 10a is machined to produce multiple glass substrates 10b. The glass substrates 10b can be considered as glass substrates before polishing. In this step, the glass substrate 10a can be cut by any method, such as laser processing using a laser beam or machining using a diamond wheel. In addition, in the glass substrate production step, glass substrates 10b of any size can be produced, such as glass substrates 10b with a side length of 300 mm to 1000 mm and a thickness of 0.5 mm to 4 mm.

[0064] (First polishing condition setting step) Next, in this manufacturing method, first polishing conditions for the raw glass plate 10b are set based on the thickness, thickness deviation, and warpage of the glass plate 10a (step S18; first polishing condition setting step). The first polishing conditions are polishing conditions for polishing the surface of the glass plate 10b, and are polishing conditions for setting the thickness, thickness deviation, and warpage of the glass plate 10a to predetermined levels. In the first polishing condition setting step, the first polishing conditions are set for each raw glass plate 10b. Setting the first polishing conditions based on the thickness, thickness deviation, and warpage can prevent a decrease in the dimensional accuracy of the glass substrate 10 to be manufactured.

[0065] More specifically, in the first polishing condition setting step, the polishing allowance of the raw glass sheet 10b is determined as the first polishing condition based on the thickness of the raw glass sheet 10a. The polishing allowance refers to the thickness of the raw glass sheet 10b to be removed by polishing, and can be considered as the difference between the thickness of the raw glass sheet 10b before and after polishing. Furthermore, in the first polishing condition setting step, the polishing allowance is determined based on the sorting results (selected groups) in the glass sheet sorting step. That is, the polishing allowance is preset for each group of thicknesses of the glass sheet 10a, and in the first polishing condition setting step, the polishing allowance set for the group of glass sheet 10a corresponding to the raw glass sheet 10b to be polished is set as the polishing allowance as the first polishing condition.

[0066] In the first polishing condition setting step, a polishing method for the glass substrate 10b is determined based on the thickness deviation and warpage of the glass substrate 10a. The polishing method is a polishing condition that indicates how the glass substrate 10b is polished. Examples of the polishing method include the load value applied by the polishing pad to the glass substrate 10b and the rotation speed of the polishing pad. For example, the polishing load value for thicker portions is increased to selectively increase the amount of polishing for thicker portions.

[0067] The order of execution of steps S14 to S18 is not limited to this order and may be arbitrary. For example, steps S14 and S18 may be executed after step S16, or steps S14 and S18 may be executed before step S16.

[0068] (First polishing step) After the first polishing conditions are set, the manufacturing method involves polishing the raw glass plate 10b under the first polishing conditions to produce a glass plate 10c (step S20; first polishing step). In the first polishing step, the surface of the raw glass plate 10b is polished under the first polishing conditions. That is, in the first polishing step, the surface of the raw glass plate 10b is polished using the polishing method set as the first polishing conditions so that the polishing stock removal set as the first polishing conditions is removed, thereby producing the glass plate 10c. Note that in the first polishing step, both one surface (one main surface) and the other surface (the other main surface) of the raw glass plate 10b are polished, but this is not limited thereto; it is sufficient that at least one of the one surface and the other surface is polished. In the first polishing step, the glass plate 10c may be produced to any thickness. For example, polishing may be performed so that the thickness of the glass plate 10c is 0.5 mm or more and 4 mm or less. Note that the polishing method for the raw glass plate 10b may be any method, such as lapping or surface grinding.

[0069] After the glass plate 10c is produced, the manufacturing method measures the thickness, thickness deviation, and warpage of the glass plate 10c (step S22; glass plate measurement step). In the glass plate measurement step, for example, the thickness may be measured at each position (coordinate) on a plane along the surface (main surface) of the glass plate 10c, and the average value of the thicknesses at each position may be taken as the thickness of the glass plate 10c. However, the thickness of the glass plate 10c is not limited to the average value of the thicknesses at each position. For example, the thickness of the glass plate 10c may be the thickness at a predetermined position, such as the center position, of the glass plate 10c, or the maximum or minimum value of the thickness at each position may be taken as the thickness of the glass plate 10c.

[0070] In the glass plate measurement step, the thickness deviation of the glass plate 10c is measured. The thickness deviation refers to the deviation in thickness at each position (each coordinate) on a plane along the surface of the glass plate 10c. For example, the thickness may be measured at each position (coordinate) on a plane along the surface of the glass plate 10c, and the difference between the maximum and minimum values ​​of the thickness at each position may be taken as the thickness deviation.

[0071] In the glass plate measurement step, the amount of warpage of the glass plate 10c is measured. The amount of warpage of the glass plate 10c refers to the amount of warpage excluding the amount of warpage due to its own weight. The amount of warpage of the glass plate 10c is calculated by dividing the maximum value of the difference in the amount of warpage at each position (each coordinate) on a plane along the surface of the glass plate 10c between a case where one surface of the glass plate 10c is supported by the support member B in the same manner as described in FIG. 3 and a case where the other surface of the glass plate 10c is supported by the support member B in the same manner as described in FIG. 3 by two. That is, the amount of warpage at position (coordinate) i on a plane along the surface of the glass plate 10c when one surface of the glass plate 10c is supported by the support member B facing vertically downward is calculated as TAb (i) When the other surface of the glass plate 10c is supported by the support member B facing vertically downward, the deflection amount at a position (coordinate) i on a plane along the surface of the glass plate 10c is defined as TBb. (i) Then, the deflection amount TAb for each position i is (i) and deflection TBb (i) The maximum difference between these values ​​is called MAX(TAb (i) -TBb (i) In this case, the amount of warpage ΔTb of the glass plate 10c is calculated by the following formula (3). That is, the amount of warpage ΔTb is calculated by the following formula (3): (i) -TBa (i) ) divided by 2.

[0072] ΔTb=|MAX(TAb (i) -TBb (i) )| / 2 ···(3)

[0073] (Glass plate sorting step) Next, in this manufacturing method, the glass plates 10c are sorted based on the measured thickness of the glass plates 10c (step S24: glass plate sorting step). That is, in the glass plate sorting step, the multiple glass plates 10c are classified by thickness. For example, in the glass plate sorting step, the range between the upper and lower limit values ​​of the thickness that the glass plates 10c can have is divided into multiple numerical ranges, and the glass plates 10c to be sorted are assigned to a group whose numerical range includes the thickness of the glass plate 10c. In the glass plate sorting step, similar sorting is performed on each of the multiple glass plates 10c, and each of the multiple glass plates 10c is assigned to one of the groups. Note that if the thickness of the glass plate 10c falls outside the range between the upper and lower limit values, it may be rejected as a defective product. By sorting the glass plates 10c based on their thickness in this manner, for example, the polishing allowance can be determined for each group, thereby stabilizing the thickness of the manufactured glass substrate 10.

[0074] As described above, in this embodiment, the glass sheets 10c are selected using the thickness among the measured thickness, thickness deviation, and warpage. However, the present invention is not limited to this, and the glass sheets 10c may also be selected using the thickness deviation and warpage. That is, in the glass sheet selection step, the glass sheets 10c may be selected based on at least one of the thickness, thickness deviation, and warpage. Note that the selection based on the thickness deviation and warpage may also be performed in a manner similar to the selection based on thickness. That is, for example, the ranges between the upper and lower limits of the thickness deviation and warpage are divided into multiple numerical ranges, and the glass sheets 10c to be selected are assigned to a group of numerical ranges that include the thickness deviation and warpage of the glass sheets 10c.

[0075] (Second polishing condition setting step) Next, in this manufacturing method, second polishing conditions for the glass plate 10c are set based on the thickness, thickness deviation, and amount of warpage of the glass plate 10c (step S26; second polishing condition setting step). The second polishing conditions are polishing conditions for polishing the surface of the glass plate 10c, and are polishing conditions for setting the thickness, thickness deviation, and amount of warpage of the glass plate 10c to predetermined levels. In the second polishing condition setting step, the second polishing conditions are set for each glass plate 10c. Setting the second polishing conditions based on the thickness, thickness deviation, and amount of warpage can suppress a decrease in the dimensional accuracy of the glass substrate 10 to be manufactured.

[0076] More specifically, in the second polishing condition setting step, a polishing allowance for the glass plate 10c is determined as the second polishing condition based on the thickness of the glass plate 10c. Furthermore, in the second polishing condition setting step, a polishing allowance is determined based on the selection result (selected group) in the glass plate selection step. That is, the polishing allowance is set in advance for each group of thicknesses of the glass plate 10c, and in the second polishing condition setting step, the polishing allowance set for the group of glass plates 10c to be polished is set as the polishing allowance as the second polishing condition.

[0077] In the second polishing condition setting step, a polishing method for the glass plate 10c is determined based on the thickness deviation and warpage of the glass plate 10c. The polishing method is a polishing condition that indicates how the glass plate 10c is polished. Examples of the polishing method include the load value applied by the polishing pad to the glass plate 10c and the rotation speed of the polishing pad.

[0078] (End face processing step) Next, in this manufacturing method, the edge surface (side surface) of the glass plate 10c is processed (step S28; edge surface processing step). Specifically, in the edge surface processing step, the edge surface of the glass plate 10c is chamfered to form an edge surface portion 10C2 on the edge surface of the glass plate 10c. As a result, an edge surface portion 10C1 and an edge surface portion 10C2 are formed on the edge surface of the glass plate 10c. The chamfering may be performed by any method, but for example, the edge surface may be chamfered by bringing a rotating grindstone into contact with the edge surface, or may be chamfered by chemical treatment. Furthermore, in the edge surface processing step, the edge surface portion 10C1 may be polished.

[0079] The order of steps S26 and S28 is arbitrary; for example, step S28 may be executed first, and then step S26 may be executed.

[0080] (Second polishing step) Next, in this manufacturing method, the glass plate 10c is polished under the second polishing conditions to produce the glass substrate 10 (step S30; second polishing step). This produces the glass substrate 10 described in this embodiment. In the second polishing step, the surface of the glass plate 10c is polished under the second polishing conditions. That is, in the second polishing step, the surface of the glass plate 10c is polished using a polishing method set as the second polishing conditions so that the polishing stock removal set as the second polishing conditions is removed by polishing, thereby producing the glass substrate 10. In the second polishing step, both one surface (one main surface) and the other surface (the other main surface) of the glass plate 10c are polished. Note that the glass plate 10c may be polished by any method, and for example, polishing using cerium oxide may be used.

[0081] (Washing step) After the glass substrate 10 is manufactured, the glass substrate 10 is cleaned (step S32; cleaning step). Any cleaning method may be used, and for example, at least one of cleaning with an alkaline detergent, ultrasonic cleaning, and scrubbing cleaning may be used.

[0082] (Inspection step) After the glass substrate 10 is cleaned, the glass substrate 10 is inspected (step S34; inspection step). In the inspection step, the dimensions of the glass substrate 10 are measured and the glass substrate 10 is inspected.

[0083] (effect) As described above, in the manufacturing method of the glass substrate 10 according to this embodiment, the glass mother sheet 10a is produced (glass mother sheet producing step), the thickness, thickness deviation, and warpage of the glass mother sheet 10a are measured (glass mother sheet measuring step), the glass mother sheet 10a is selected based on the thickness of the glass mother sheet 10a (glass mother sheet selecting step), the selected glass mother sheet 10a is cut to produce a plurality of raw glass sheets 10b (glass mother sheet producing step), first polishing conditions for the glass mother sheet 10b are set based on the thickness, thickness deviation, and warpage of the glass mother sheet 10a (first polishing condition setting step), and the first Based on the polishing conditions, the surface of the raw glass plate 10b is polished to produce a glass plate 10c (first polishing step), the thickness, thickness deviation, and amount of warping of the glass plate 10c are measured (glass plate measurement step), the glass plate 10c is selected based on the thickness of the glass plate 10c (glass plate selection step), second polishing conditions for the glass plate 10c are set based on the thickness, thickness deviation, and amount of warping of the glass plate 10c (second condition setting step), and the surface of the selected glass plate 10c is polished based on the second polishing conditions to produce a rectangular glass substrate 10 with a side length of 300 mm or more and a thickness of 0.5 mm or more.

[0084] Here, glass substrates supporting semiconductor devices require dimensional accuracy, such as thickness and warpage. However, because glass substrates supporting semiconductor devices have relatively large thicknesses and areas, it can be difficult to ensure high dimensional accuracy. In contrast, according to the present manufacturing method, by selecting the glass mother plates 10a and glass plates 10c based on thickness, thickness variation among the glass substrates 10 is reduced, and by setting the first polishing conditions and the second polishing conditions based on the thickness, thickness deviation, and warpage, the thickness and thickness deviation can be set to appropriate values, thereby suppressing a decrease in dimensional accuracy.

[0085] In this manufacturing method, it is preferable to set the polishing allowance of the glass raw material plate 10b based on the results of sorting the glass mother plate 10a as the first polishing condition for the glass raw material plate 10b, and to set the polishing method based on the thickness deviation and warpage of the glass mother plate 10a. In this manufacturing method, it is preferable to set the polishing allowance of the glass plate 10c based on the results of sorting the glass plate 10c, and to set the polishing method based on the thickness deviation and warpage of the glass plate 10c as the second polishing condition for the glass plate 10c. In this manufacturing method, the polishing allowance is determined from the results of sorting by thickness, and the polishing method is determined from the thickness deviation and warpage, so that deterioration of dimensional accuracy can be more effectively suppressed.

[0086] In this manufacturing method, the surface of the glass plate 10c is polished (first polishing), and then the edge of the glass plate 10c is processed (edge ​​processing step). According to this manufacturing method, since the edge is processed after the first polishing, the influence of variations in plate thickness can be minimized and the deterioration of the dimensional accuracy of the edge shape can also be suppressed.

[0087] Furthermore, the glass substrate 10 according to this embodiment is designed to support semiconductor devices and has a rectangular shape with a side length L of 300 mm or more, a thickness D of 0.7 mm or more, a warpage ΔT of 1 mm or less, a thickness D of 0.5 mm to 4.0 mm, a deviation in thickness D of 5 μm or less, a local thickness variation (LTV) of 2 μm or less in a 50 mm x 50 mm area, a width W of the edge surface 10C2 of 1 mm or less, and a radius of curvature formed by any three points in the region with the smallest curvature (curved portion 10C1b) between the edge surface 10C2 and the end surface 10C1 (straight portion 10C1a) of 0.05 mm or more. This shape of the glass substrate 10 minimizes degradation of dimensional accuracy, making it suitable as a support substrate for semiconductor devices. While such a glass substrate 10 is preferably manufactured using the manufacturing method described in this embodiment, any method may be used.

[0088] Example 1 Next, Example 1 will be described. Table 1 shows each example. In each example, the manufacturing process of the glass substrate is different.

[0089] [Table 1]

[0090] (Example 1) In Example 1, a glass mother plate production step was performed to produce glass mother plates so that their composition fell within the composition range of the glass substrate 10 of this embodiment. A glass mother plate selection step was then performed to select the glass mother plates based on their thickness. A first polishing condition setting step was then performed to set first polishing conditions for the glass mother plates based on the thickness, thickness deviation, and warpage of the glass mother plates. A glass mother plate production step was then performed to cut the glass mother plates to produce glass mother plates. A first polishing step was then performed to polish the glass mother plates to produce glass plates. In the first polishing step, the glass mother plates were polished based on the first polishing conditions. A glass plate selection step was then performed to select the glass plates based on their thickness. A second polishing condition setting step was then performed to set second polishing conditions for the glass plates based on the thickness, thickness deviation, and warpage of the glass plates. An edge processing step was then performed to process the edge of the glass plate. A second polishing step was then performed to polish the glass plates to produce glass substrates. In the second polishing step, the glass plate was polished under the second polishing conditions.

[0091] (Example 2) In Example 2, the glass mother plate selection step, the first polishing condition setting step, the glass plate selection step, and the second polishing condition setting step were not performed. That is, in Example 2, the polishing conditions were not determined based on the thickness, thickness deviation, and warpage of the glass mother plate and the glass plate. In addition, in Example 2, the edge processing step was not performed. Other than these, the glass substrate was produced under the same conditions as in Example 1.

[0092] (Example 3) In Example 3, the glass plate selection step and the second polishing condition setting step were not performed. That is, in Example 3, the second polishing conditions were not determined based on the thickness, thickness deviation, and warpage of the glass plate. In addition, in Example 3, the edge processing step was not performed. Other than these, the glass substrate was produced under the same conditions as in Example 1.

[0093] (Example 4) In Example 4, the glass mother plate selection step, the first polishing condition setting step, the glass plate selection step, and the second polishing condition setting step were not performed. That is, in Example 4, the polishing conditions were not determined based on the thickness, thickness deviation, and warpage of the glass mother plate and the glass plate. Other than these, the glass substrate was produced under the same conditions as in Example 1.

[0094] (Example 5) In Example 5, the glass plate selection step and the second polishing condition setting step were not performed. That is, in Example 5, the second polishing conditions were not determined based on the thickness, thickness deviation, or warpage of the glass plate. Except for these, the glass substrate was produced under the same conditions as in Example 1.

[0095] (Example 6) In Example 6, the glass substrate selection step and the first polishing condition setting step were not performed. That is, the first polishing conditions were not determined based on the thickness and thickness deviation of the glass substrate in Example 6. Other than these, the glass substrate was produced under the same conditions as in Example 1.

[0096] (shape measurement) The amount of warpage, thickness, thickness deviation, LTV, edge width, and curvature radius of the curved portion of the glass substrate produced in each example refer to the amount of warpage ΔT, thickness D, deviation of thickness D, LTV at 50 mm × 50 mm, width W of edge 10C2, and curvature radius of curved portion 10C1b, respectively, as described in this embodiment, and the measurement methods are also the same as those described in this embodiment. Note that the glass substrates of Examples 2 to 6 could not be measured in part due to a high risk of affecting the equipment.

[0097] (evaluation) In the thickness evaluation, a thickness of 1.49 mm or more and 1.51 mm or less was marked with a circle, and a thickness outside the range was marked with an X. In addition, in the evaluation of thickness deviation, a thickness deviation of 5 μm or less was marked with a circle, a thickness deviation of more than 5 μm and less than 10 μm was marked with a triangle, and a thickness deviation of more than 10 μm was marked with an X. In addition, semiconductor devices were installed on the glass substrates of each example to evaluate whether the semiconductor device manufacturing process could be carried out. If the manufacturing process could be carried out, a circle was marked, and if it could not, an X was marked. A sample that scored a circle for thickness, thickness deviation, and process flow was considered to have passed. In Example 1, which is an embodiment, all thicknesses and thickness deviations were marked as "good," indicating that a decrease in dimensional accuracy was suppressed. Furthermore, the glass substrate of Example 1 had thicknesses and thickness deviations within appropriate ranges, so no particular problems occurred in the process flow and it was suitable as a support substrate. On the other hand, in Examples 2 to 6, which are comparative examples, at least one of the thicknesses and thickness deviations was marked as "good," indicating that the dimensional accuracy decreased. Furthermore, the glass substrates of Examples 2 and 3 had thicknesses and thickness deviations that were not within appropriate ranges, and there were areas without edge surfaces, so the edge surfaces could not be recognized, some shape measurements could not be performed, and the process flow was not possible. Furthermore, the glass substrates of Examples 4 to 6 had large thickness deviations, so some shape measurements could not be performed, and the devices could not be properly installed, so the process flow was not possible.

[0098] The end face shape was also evaluated as an option. A circle was marked when the length D1 of the straight portion 10C1a in the Z direction in FIG. 2B was 0.1 mm or more and 0.5 mm or less, and an X was marked when the length D1 of the straight portion 10C1a in the Z direction was less than 0.1 mm or more than 0.5 mm. In Examples 1 and 4 to 6, the end face processing step was performed, which is preferable because it suppresses a decrease in the dimensional accuracy of the end face shape.

[0099] Although the embodiments of the present invention have been described above, the embodiments are not limited to the contents of these embodiments. Furthermore, the above-described components include those that can be easily imagined by a person skilled in the art, those that are substantially the same, and those that are within the scope of what is called equivalents. Furthermore, the above-described components can be combined as appropriate. Furthermore, various omissions, substitutions, or modifications of the components can be made without departing from the spirit of the above-described embodiments. [Explanation of symbols]

[0100] 10 Glass substrate 10a Glass mother board 10b Glass plate 10c Glass Plate 10A, 10B surface 10C end face 10C1 End section 10C2 Edge surface

Claims

1. A glass substrate supporting a semiconductor device, It has a rectangular shape with a side length of 300 mm or more, The thickness is 0.7 mm or more, The warpage of the glass substrate is 1 mm or less, The thickness of the glass substrate is 0.5 mm or more and 4.0 mm or less, The deviation in thickness of the glass substrate is 5 μm or less; The LTV (Local Thickness Variation) of the glass substrate in a 50 mm x 50 mm area is 2 μm or less; The Young's modulus of the glass substrate is 70 GPa or more and 120 GPa or less. Glass substrate.

2. A glass substrate supporting a semiconductor device, It has a rectangular shape with a side length of 300 mm or more, The thickness is 0.7 mm or more, The warpage of the glass substrate is 1 mm or less, The thickness of the glass substrate is 0.5 mm or more and 4.0 mm or less, The deviation in thickness of the glass substrate is 5 μm or less; The LTV (Local Thickness Variation) of the glass substrate in a 50 mm x 50 mm area is 2 μm or less; The glass substrate has an average coefficient of thermal expansion (CTE) of 3 ppm / °C or more and 12 ppm / °C or less at 50°C to 200°C. Glass substrate.

Images