Workpiece machining method and workpiece with convex pattern

Abstract

In a workpiece processing method that includes the step of providing a convex pattern on the outer periphery of the surface of a workpiece having bump electrodes, and further protecting the workpiece surface by attaching a surface protection sheet to the convex pattern, the costs required for forming and disposing of the convex pattern are reduced. [Solution] In the processing method for workpiece 1 of the present invention, when the maximum height of the convex pattern 3 is H1, the average height of the convex pattern is H2, and the average height of the bump electrode 2 is Hb, then H2

Description

[Technical Field]

[0001] The present invention relates to a method for machining a workpiece. In particular, it relates to a method for machining a workpiece that can reduce the TTV of the workpiece after back grinding, even if the workpiece has bump electrodes formed on it. Furthermore, the present invention relates to a workpiece with a convex pattern to which this machining method is preferably applied. [Background technology]

[0002] A known method for densely mounting semiconductor chips and other circuit-equipped chips onto a substrate involves joining the chip to electrodes on the substrate via bump electrodes formed on the circuit surface of the chip. Bump electrodes are convex electrodes that protrude in the thickness direction of the chip.

[0003] Such chips are obtained as individual workpiece fragments by fragmenting a workpiece, such as a wafer on which a circuit has been formed. With the rapid progress of miniaturization and multi-functionality of electronic devices on which such chips are mounted, there is a demand for smaller, lower-profile, and higher-density chips. To miniaturize and lower the profile of a chip, it is common practice to form a circuit on the surface of a wafer and then grind the back surface of the wafer to reduce the thickness of the chip.

[0004] During back-grinding of a wafer, a surface protection sheet called backgrind tape is applied to the wafer surface to protect the circuits on the wafer surface and to hold the wafer in place.

[0005] When grinding the back surface of a workpiece with bump electrodes formed on it, the difference in height between the bump electrodes and the circuit surface could cause damage to the workpiece or poor adhesion between the workpiece and the surface protection sheet.

[0006] Patent Document 1 discloses a surface protection sheet that conforms well to the surface of a semiconductor wafer having steps or protrusions, and can be peeled off without damaging the semiconductor wafer or leaving any adhesive residue.

[0007] However, when the surface protection sheet conforms to the workpiece surface, a problem arises in which the TTV (Total Thickness Variation) deteriorates at the outer periphery of the workpiece. Since no circuit is formed at the outer periphery of the workpiece, no bump electrodes are formed there either. As a result, a difference in height occurs between the outer periphery of the workpiece and the region where the circuit and bump electrodes are formed (the inner periphery of the workpiece).

[0008] As shown in Figure 1(A), when a surface protection sheet 20 is attached to the circuit surface (surface 1a) on which the bump electrode 2 is formed, the surface protection sheet 20 attached to the surface 1a of the workpiece conforms to the shape of the workpiece surface 1a and the bump electrode 2. Therefore, differences in height may exist in the surface protection sheet 20, reflecting the differences in height on the workpiece 1.

[0009] When grinding the back surface of a workpiece with a height difference in such a surface protection sheet, as shown in Figure 1(B), the workpiece is placed so that the surface side, i.e., the surface protection sheet 20, is in contact with the chuck table 100, and the workpiece 1 is fixed to the chuck table 100 via the surface protection sheet 20, for example by suction. At this time, the surface protection sheet 20 located on the bump electrode 2 is in contact with the chuck table 100 when placed on the chuck table 100, but the surface protection sheet 20 located on the area where the bump electrode 2 is not formed, i.e., on the outer circumference of the workpiece 1, may not be in sufficient contact with the chuck table 100 or a gap C may be created between the chuck table 100 and the surface protection sheet 20 due to the height difference.

[0010] When suction occurs in this state, as shown in Figure 1(C), the surface protection sheet 20 on the outer circumference of the workpiece 1 deforms due to the suction and falls towards the chuck table 100, becoming fixed in place. Since the outer circumference of the workpiece 1 is in close contact with the surface protection sheet 20, the outer circumference of the workpiece 1 deforms along with the deformation of the surface protection sheet 20. In other words, the outer circumference of the workpiece 1 is fixed in a curved position toward the chuck table 100 compared to the inner circumference of the workpiece 1 (the region where the bump electrode 2 is formed).

[0011] After fixing the workpiece 1 and the surface protection sheet 20 to the chuck table 100, the back surface 1b of the workpiece 1 is ground. The workpiece after grinding is shown in Figure 1(D). Since the back surface of the workpiece is ground to almost the same height, as shown in Figure 1(D), the outer circumference of the workpiece 1 is thicker than the inner circumference of the workpiece 1 because it was fixed closer to the chuck table 100. As a result, when the suction is released, as shown in Figure 1(E), the deformation of the surface protection sheet 10 is released, and the height difference between the inner circumference and the outer circumference of the workpiece 1 becomes apparent. Therefore, in this case, the thickness of the outer circumference of the workpiece 1 tends to be the maximum thickness of the workpiece 1, and the thickness of the inner circumference of the workpiece 1 tends to be the minimum thickness of the workpiece 1.

[0012] The difference between the maximum and minimum thickness of a workpiece is called TTV (Total Thickness Variation), and the above-mentioned height difference can cause the TTV to become large. When the TTV is large, cracks may easily form in the workpiece, and problems may occur when the workpiece is broken down into individual pieces.

[0013] Since Patent Document 1 does not take the above problem into consideration, even if the adhesive tape for semiconductor wafer processing described in Patent Document 1 is attached to a semiconductor wafer on which bump electrodes etc. are formed, a height difference occurs between the inner circumference and the outer circumference of the semiconductor wafer. As a result, there was a problem in that the TTV of the semiconductor wafer after backside grinding became large.

[0014] Patent Document 2 aims to provide a workpiece processing method that can reduce the TTV of the workpiece, A process of attaching a surface protection sheet to the surface of a workpiece that has a front and back surface, The process includes grinding the back surface of a workpiece to which a surface protection sheet has been attached, The surface of the workpiece has bump electrodes, A workpiece processing method has been proposed in which, during the process of applying a surface protection sheet, a convex pattern is formed on the outer periphery of the workpiece surface in a region different from the area where bump electrodes are formed. [Prior art documents] [Patent Documents]

[0015] [Patent Document 1] Japanese Patent Publication No. 2017-171896 [Patent Document 2] International release WO2023 / 189168 [Overview of the Initiative] [Problems that the invention aims to solve]

[0016] Patent Document 2 describes a method for creating a convex pattern on the outer periphery of a workpiece surface that is relatively wide and uniform in height. However, forming such a convex pattern requires a large amount of raw material, resulting in high raw material costs. Furthermore, the convex pattern is peeled off and discarded along with the surface protection sheet. If the volume of the convex pattern is large, the amount of waste also increases, leading to higher disposal costs and increased environmental burden. Therefore, when employing a workpiece processing method like that described in Patent Document 2, there is a need to reduce the costs required for forming and discarding the convex pattern. [Means for solving the problem]

[0017] The embodiments of the present invention are as follows. (1) A step of attaching a surface protection sheet to the surface of a workpiece that has a front and back surface, A step of grinding the back surface of the workpiece with the surface protection sheet attached thereto. The surface of the workpiece has bump electrodes. In the step of attaching the surface protection sheet, a convex pattern is formed in a region different from the region where the bump electrodes are formed on the outer peripheral portion of the surface of the workpiece. The relationship between the maximum height H1 of the convex pattern, the average height H2 of the convex pattern, and the average height Hb of the bump electrodes satisfies H2 < H1 < 2Hb / 3. The surface protection sheet is attached to the top of the convex pattern, a method for processing a workpiece.

[0018] (2) The method for processing a workpiece according to (1), wherein the relationship between the maximum height H1 of the convex pattern and the average height Hb of the bump electrodes satisfies 0 < H1 < Hb / 4.

[0019] (3) The method for processing a workpiece according to (1), wherein the convex pattern is made of resin.

[0020] (4) Further having a step of peeling the surface protection sheet from the workpiece after back grinding, The method for processing a workpiece according to (1), wherein at least a part of the convex pattern is peeled off together with the surface protection sheet.

[0021] (5) Having a step of forming a groove on the surface of the workpiece before the step of attaching the surface protection sheet to the surface of the workpiece, In the step of grinding the back surface of the workpiece, starting from the groove, the workpiece is separated into a plurality of workpiece pieces, the method for processing a workpiece according to any one of (1) to (4).

[0022] (6) Having a step of forming a modified region inside the workpiece before the step of grinding the back surface of the workpiece, In the step of grinding the back surface of the workpiece, starting from the modified region, the workpiece is separated into a plurality of workpiece pieces, the method for processing a workpiece according to any one of (1) to (4).

[0023] (7) A workpiece having bump electrodes on its surface, The workpiece surface has a convex pattern in a region different from the region where the bump electrode is formed, The relationship between the maximum height H1 of the convex pattern, the average height H2 of the convex pattern, and the average height Hb of the bump electrode is such that H2 Workpiece with a convex pattern.

[0024] (8) The workpiece with a convex pattern according to (7), wherein the convex pattern has multiple layers.

[0025] (9) The workpiece with a convex pattern according to (7), wherein the convex pattern is substantially a single layer. [Effects of the Invention]

[0026] According to the present invention, in a workpiece processing method that includes a step of forming a convex pattern, the TTV of the workpiece can be reduced, and the costs required for forming and discarding the convex pattern can be reduced. [Brief explanation of the drawing]

[0027] [Figure 1] Figures 1(A) to 1(E) are schematic cross-sectional diagrams illustrating the increase in TTV of a workpiece when it is machined using conventional methods. [Figure 2] Figure 2 is a schematic plan view showing a workpiece processing method according to this embodiment, in which a convex pattern is formed on the outer circumference of a workpiece having a bump electrode. [Figure 3] Figure 3 is a schematic cross-sectional view of the line IIB-IIB in Figure 2. [Figure 4A] Figure 4A is a schematic cross-sectional view of various convex patterns, shown in Figure 3 with the outer periphery magnified. [Figure 4B] Figure 4B is a schematic cross-sectional view of various convex patterns, shown in Figure 3 with the outer periphery magnified. [Figure 4C] ​Figure 4C is a schematic cross-sectional view of various convex patterns, shown in Figure 3 with the outer periphery magnified. [Figure 4D] Figure 4D is a schematic cross-sectional view of various convex patterns, shown in Figure 3 with the outer periphery magnified. [Figure 4E] Figure 4E is a schematic cross-sectional view of various convex patterns, shown in Figure 3 with the outer periphery magnified. [Figure 4F] Figure 4F is a schematic cross-sectional view of various convex patterns, shown in Figure 3 with the outer periphery magnified. [Figure 5A] Figure 5A is a schematic cross-sectional view of a surface protection sheet suitably used in the workpiece processing method according to this embodiment. [Figure 5B] Figure 5B is a schematic cross-sectional view showing another example of a surface protection sheet suitably used in the workpiece processing method according to this embodiment. [Figure 5C] Figure 5C is a schematic cross-sectional view showing another example of a surface protection sheet suitably used in the workpiece processing method according to this embodiment. [Figure 6] Figures 6(A) to 6(C) are schematic cross-sectional diagrams illustrating how the TTV of a workpiece decreases when it is processed by the method according to this embodiment. [Modes for carrying out the invention]

[0028] The present invention will be described in detail below with reference to the drawings, based on specific embodiments. First, the main terms used in this specification will be explained.

[0029] A workpiece refers to a plate-like body to which a surface protective sheet is attached and then separated into individual pieces. Examples of workpieces include circular wafers (including those having an orientation flat), rectangular panel-level packages, and strips (strip-shaped substrates) with molded resin encapsulation. Among these, wafers are preferred from the viewpoint of easily obtaining the effects of the present invention. Wafers may be semiconductor wafers such as silicon wafers, gallium arsenide wafers, silicon carbide wafers, gallium nitride wafers, and indium phosphate wafers, or insulating wafers such as glass wafers, lithium tantalate wafers, and lithium niobate wafers. They may also be reconfigured wafers made of resin and semiconductors used in the manufacture of fan-out packages, etc. From the viewpoint of easily obtaining the effects of the present invention, semiconductor wafers or insulating wafers are preferred as wafers, and semiconductor wafers are more preferred.

[0030] Workpiece fragmentation refers to the process of dividing a workpiece into individual circuit components to obtain individual workpiece fragments. For example, if the workpiece is a wafer, the individual workpiece fragments are chips; if the workpiece is a panel-level package or a strip (a strip-shaped substrate) with molded resin encapsulation, the individual workpiece fragments are semiconductor packages.

[0031] The "front surface" of a workpiece refers to the side on which circuits, electrodes, etc., are formed, while the "back surface" of a workpiece refers to the side on which circuits, etc., are not formed.

[0032] Bump electrodes are electrodes formed on the surface of a workpiece, protruding in the thickness direction of the workpiece. Typically, multiple bump electrodes are formed on a single circuit (workpiece fragment). The cross-sectional shape of the bump electrode in the thickness direction can be columnar, conical, circular, etc., and its tip is usually curved.

[0033] DBG (Dicing Before Grinding) is a method of forming grooves of a predetermined depth on the surface side of a wafer, and then grinding the wafer from the back side to separate it into individual pieces. The grooves formed on the surface side of the wafer are created by methods such as blade dicing, laser dicing, or plasma dicing.

[0034] Furthermore, LDBG (Laser Dicing Before Grinding) is a variation of DBG, which involves creating a modified region inside the wafer using a laser, and then fragmenting the wafer using stress during back-side grinding.

[0035] A "workpiece fragment group" refers to multiple workpiece fragments held on a surface protection sheet after the workpiece has been fragmented. These workpiece fragments, as a whole, form a shape similar to that of the workpiece. Similarly, a "chip group" refers to multiple chips held on a surface protection sheet after the wafer, which is the workpiece, has been fragmented. These chips, as a whole, form a shape similar to that of the wafer.

[0036] The term "(meth)acrylate" is used to refer to both "acrylate" and "methacrylate," and the same applies to other similar terms.

[0037] "Energy rays" refer to ultraviolet rays, electron beams, etc., and are preferably ultraviolet rays.

[0038] Unless otherwise specified, "weight-average molecular weight" is a polystyrene-converted value measured by gel permeation chromatography (GPC). Measurements using this method are performed, for example, with a high-speed GPC instrument "HLC-8120GPC" manufactured by Tosoh Corporation and a high-speed column "TSK guard column H XL -H", "TSK Gel GMH" XL "TSK Gel G2000 H XLThe following components (all manufactured by Tosoh Corporation) are connected in this order, and the test is performed with a column temperature of 40°C and a liquid delivery rate of 1.0 mL / min, using a differential refractometer as the detector.

[0039] A release sheet is a removable sheet that protects the adhesive layer. The term "sheet" is used in a general sense, including films, and does not limit the thickness of the sheet.

[0040] In descriptions of compositions such as adhesive compositions, the mass ratios are based on the active ingredient (solid content), and unless otherwise specified, the solvent is not included.

[0041] (1. Workpiece machining method) As a method for processing a workpiece, one example of processing a workpiece in which circuits, etc., are formed on one side (front) and not on the other side (back) is grinding the back side of the workpiece. By grinding the back side, it is possible to reduce the thickness of the individual workpiece pieces obtained by separating the workpiece into individual pieces.

[0042] In addition to the circuit, bump electrodes may be formed on the surface of the workpiece to electrically connect the circuit to the substrate or other components. As described above, the bump electrodes are formed to protrude in the thickness direction of the workpiece. Therefore, there is a difference in height between the tip of the bump electrode and the area of ​​the workpiece where the bump electrode is not formed. The area where the bump electrode is not formed roughly coincides with the area where the circuit is not formed, and this area is located on the outer periphery of the workpiece.

[0043] On the other hand, before grinding the back surface of the workpiece, a surface protection sheet is attached to the surface of the workpiece to protect circuits and other structures formed on the workpiece surface. At this time, it is necessary to fix the bump electrodes with the surface protection sheet to protect them from the force applied during back surface grinding.

[0044] The inventors have found that by employing the following method, the above-mentioned height difference can be reduced, resulting in a workpiece with a small TTV and reduced costs. The following describes in detail the workpiece processing method according to this embodiment.

[0045] The workpiece processing method according to this embodiment comprises at least the following steps 1 and 2. Step 1: A process of applying a surface protection sheet to the surface of a workpiece that has a front and back side. Step 2: Grinding the back surface of the workpiece to which the surface protection sheet has been attached. The following describes a workpiece processing method for the case where the workpiece is a wafer and the workpiece fragments are chips.

[0046] (1.1.Process 1) In step 1, a surface protection sheet is attached to the surface of the wafer. However, in this embodiment, as shown in Figures 2 and 3, before the surface protection sheet is attached, a convex pattern 3 is formed in the radial direction of the wafer 1 in the area outside the region where the bump electrodes 2 are formed (the outer periphery of the wafer surface) (step A). ​​In other words, step A is performed before step 1.

[0047] (1.2. Convex pattern) The convex pattern 3 is a pattern formed on the surface of the wafer 1, protruding in the thickness direction of the wafer 1. Because such a convex pattern 3 is formed on a region where the bump electrodes 2 are not formed, when the surface protection sheet 10 is attached to the surface of the wafer 1, the portion corresponding to the convex pattern 3 becomes higher in addition to the portion corresponding to the region where the bump electrodes 2 are formed. As a result, the height difference between the region where the bump electrodes 2 are formed (the inner circumference of the wafer) and the region where they are not formed (the outer circumference of the wafer) becomes smaller.

[0048] In this embodiment, the relationship among the maximum height H1 of the convex pattern, the average height H2 of the convex pattern, and the average height Hb of the bump electrode is within the range of H2 < H1 < 2Hb / 3. Therefore, the TTV of the wafer 1 after back grinding can be reduced, contributing to the reduction of raw material costs and disposal costs. The convex pattern is formed continuously with substantially the same shape on the outer peripheral portion of the workpiece. The maximum height H1 of the convex pattern is the shortest distance from the workpiece surface to the apex of the convex pattern. The maximum height H1 of the convex pattern can be obtained from the cross-sectional observation of the convex pattern.

[0049] The average height H2 of the convex pattern can be obtained from the cross-sectional observation of the convex pattern. Specifically, the average height H2 is obtained by S / L from the bottom length L of the convex pattern and the cross-sectional area S obtained from the result of the cross-sectional observation. The average height Hb of the bump electrode can be obtained from the cross-sectional observation of the bump electrode.

[0050] In this embodiment, the maximum height H1 of the convex pattern is the maximum value among the maximum heights obtained by cross-sectional observation of at least three points or more, and the average height H2 of the convex pattern is the average value of the heights obtained by cross-sectional observation of at least three points or more. Bumps are generally formed with uniform sizes. The average height Hb of the bump electrode in this embodiment is the average value of the heights obtained by cross-sectional observation of at least three points or more.

[0051] The relationship between the maximum height H1 of the convex pattern and the average height H2 of the convex pattern is preferably within the range of 6H2 / 5 < H1. Also, the relationship between the maximum height H1 of the convex pattern and the average height Hb of the bump electrode is preferably within the range of H1 < 3Hb / 5. Further, the relationship among the maximum height H1 of the convex pattern, the average height H2 of the convex pattern, and the average height Hb of the bump electrode is preferably within the range of 0 < H2 < Hb / 4. When the maximum height H1 of the convex pattern, the average height H2 of the convex pattern, and the average height Hb of the bump electrode are within the above ranges, the convex pattern can achieve a predetermined height with less raw material, contributing to the reduction of raw material costs and disposal costs.

[0052] If H1 is outside the range of the above relationship, the TTV of the wafer tends to increase after backside grinding.

[0053] The maximum height H1 of the convex pattern is preferably 2 to 2000 μm, more preferably 3 to 1000 μm, and particularly preferably 5 to 500 μm. The average height H2 of the convex pattern is preferably 1 to 1500 μm, more preferably 2 to 800 μm, and particularly preferably 3 to 400 μm.

[0054] (1.2.1. Shape of the convex pattern) The shape of the convex pattern can vary, as long as it is formed in a way that achieves the above-mentioned effects. For example, as shown in Figures 2 and 3, the convex pattern may be formed in a ring shape that includes the outer edge of the wafer. Such a convex pattern is easy to form while reliably achieving the above-mentioned effects.

[0055] The cross-sectional shape of the convex pattern may be, for example, a mountain-like shape with a raised center, as shown in Figure 4A; a structure consisting of multiple convex parts with the highest convex part located in the center, as shown in Figure 4B; a structure consisting of multiple convex parts with the highest convex part located on the outer circumference, as shown in Figure 4C; a structure consisting of multiple convex parts with the highest convex part located on the inner circumference, as shown in Figure 4D; a tapered structure with the outer circumference being higher and the inner circumference being lower, as shown in Figure 4E; or a tapered structure with the inner circumference being higher and the outer circumference being lower, as shown in Figure 4F. The tops of each convex part may be rounded.

[0056] The convex pattern may be formed over the entire area where the bump electrode 2 is not formed (the outer periphery of the workpiece), but from the viewpoint of saving raw materials, it is preferable to form it on a part of the outer periphery of the workpiece. As shown in Figure 4A, if the width of the area where the bump electrode 2 is not formed is W, the ratio of the width W to the bottom length L of the convex pattern (L / W) is preferably in the range of 0.1 to 1, and more preferably in the range of 0.2 to 1. If the convex pattern consists of multiple protrusions, the bottom length L of the convex pattern is the sum of the bottom lengths of all the protrusions. Also, although the convex pattern shown in Figure 2 is a continuous pattern, it may also be a pattern formed intermittently.

[0057] The means for forming the convex pattern should be determined according to the material used to form the convex pattern. For example, when forming a convex pattern using a curable resin, a means for applying the liquid resin before curing can be used. Specifically, coating devices such as die coaters, curtain coaters, spray coaters, slit coaters, and knife coaters; printing devices such as screen printers and inkjet printers; and dispensing devices such as dispensers can be used.

[0058] The convex patterns in Figures 4B to 4D are obtained by applying multiple linear coatings with narrow line widths and allowing them to cure. The convex patterns in Figures 4E and 4F can be formed by bonding the convex patterns in Figures 4C and 4D together under pressure while maintaining the height differences of each convex portion. For example, this bonding can be achieved by the pressure applied when a surface protective sheet is attached. In this case, cross-sectional observation of the resulting convex pattern reveals that it has a multilayer structure derived from multiple linear coatings. Note that having a multilayer structure means that a clear boundary line can be observed within the convex pattern when observing its cross-section.

[0059] Furthermore, the convex patterns in Figures 4E and 4F can also be obtained by applying wind, centrifugal force, static electricity, etc., to a single-layer coating of a curable resin with the shape shown in Figure 4A, while it is still uncured, to induce deformation. For example, the convex pattern in Figure 4E is obtained by applying wind to the uncured coating from the inside to the outside of the wafer to deform the convex pattern and then curing it. Similarly, the convex pattern in Figure 4F is obtained by applying wind to the uncured coating from the outside to the inside of the wafer to deform the convex pattern and then curing it. In this case, cross-sectional observation of the obtained convex pattern confirms that it is essentially a single-layer structure. A single-layer structure means that no clear boundary line is observed within the convex pattern when observing its cross-section.

[0060] As described above, the present invention provides a workpiece with a convex pattern having the above-mentioned convex pattern and bump electrodes on its surface. In such a workpiece with a convex pattern, the convex pattern may consist of multiple layers or may be substantially a single layer.

[0061] (1.2.2. Shear storage modulus of convex patterns) In this embodiment, it is preferable that the shear storage modulus of the convex pattern at 23°C is 50 MPa or higher. Having the shear storage modulus at 23°C within this range results in a relatively hard convex pattern, suppressing the height difference between the outer and inner edges of the wafer even during backside grinding.

[0062] The material constituting the convex pattern can be any material that provides the above-mentioned effects. In this embodiment, from the viewpoint of ease of forming the convex pattern, the material constituting the convex pattern is preferably a resin, and preferably its shear storage modulus at 23°C is within the above range. The resin is preferably an adhesive resin that adheres closely to the wafer surface. Examples of adhesive resins include acrylic resin, silicone resin, urethane resin, epoxy resin, phenolic resin, urea resin, alkyd resin, vinyl acetate resin, vinyl chloride resin, amide resin, imide resin, chloroprene rubber, nitrile rubber, styrene-butadiene rubber, nylon, polycarbonate, and polypropylene.

[0063] Furthermore, it is preferable that the adhesive resin is curable. Before curing, it is easy to form a predetermined shape, and after curing, it is easy to make the material hard enough to prevent abrasives, cooling water, etc. from penetrating the circuit surface of the wafer during back grinding, and to suppress the height difference between the inner and outer edges of the wafer. Note that if the material constituting the convex pattern is a curable resin, the shear storage modulus of the convex pattern at 23°C is the shear storage modulus after curing.

[0064] Examples of curable resins include thermosetting resins and energy ray curable resins, with energy ray curable resins being preferred from the viewpoint of removing convex patterns from the wafer surface after backside grinding. Specific examples of curable resins include combinations of urethane acrylate, polymerizable monomer, and photopolymerization initiator, as exemplified in the intermediate layer composition described later.

[0065] (1.3. Surface protection sheet) After forming the convex pattern (after step A), a surface protection sheet 10 is attached to the wafer surface as shown in Figure 6A. In this embodiment, the surface protection sheet is attached so as to cover at least the bump electrodes present on the wafer surface. "Attaching so as to cover the bump electrodes" means attaching the surface protection sheet so that it is in contact with the tops of the bump electrodes. Similarly, the surface protection sheet is attached so that the tops of the convex pattern are in contact with the surface protection sheet.

[0066] By covering the bump electrodes with a surface protection sheet, the bump electrodes are fixed in place, and damage to the bump electrodes is suppressed even if external forces are applied to them during backside grinding. Furthermore, by applying the surface protection sheet so that it contacts the bump electrodes and the tops of the convex patterns, the height difference in the inner circumference of the workpiece can be suppressed. Therefore, the overall TTV of the wafer can be reduced.

[0067] The surface protection sheet only needs to be configured to fix the bump electrodes when it is attached to the wafer surface. In this embodiment, it is preferable that the surface protection sheet has a base material and an adhesive layer. The adhesive layer is attached to the wafer surface, and the bump electrodes are held and fixed by the adhesive layer.

[0068] From the viewpoint of securely fixing the bump electrode to the surface protection sheet, it is more preferable that the surface protection sheet consists of a base material and an adhesive layer, or has a structure in which the base material, an intermediate layer, and an adhesive layer are laminated in this order. The intermediate layer, together with the adhesive layer, sufficiently follows the shape of the bump electrode formed on the surface of the workpiece, so that even if the bump electrode is large, the bump electrode can be securely held by the adhesive layer and the intermediate layer. As a result, even if the workpiece is ground very thin and force is applied to the bump electrode, the adhesive layer and the intermediate layer can sufficiently protect the bump electrode, etc. Furthermore, if the bump electrode, etc. penetrates the adhesive layer, the intermediate layer will hold the bump electrode, etc.

[0069] Below, we will describe in detail a surface protection sheet consisting of a base material and an adhesive layer (Figure 5A), and a surface protection sheet having a structure in which the base material, intermediate layer, and adhesive layer are laminated in that order (Figure 5B).

[0070] (1.3.1. Base material) The base material is the component responsible for the rigidity of the surface protection sheet. The base material is not limited as long as it is made of a material that can support the workpiece. For example, various resin films used as base materials for backgrind tapes are examples. By using such resin films, the workpiece can be held without damage even if its thickness is reduced by grinding. The base material may consist of a single-layer film made of one resin film, or it may consist of a multi-layer film made by laminating multiple resin films.

[0071] In this embodiment, suitable materials for the base material include, for example, polyesters such as polyethylene terephthalate, polyethylene naphthalate, polybutylene terephthalate, and fully aromatic polyesters, as well as polyamides, polycarbonates, polyacetals, modified polyphenylene oxide, polyphenylene sulfide, polysulfone, polyether ketone, and biaxially oriented polypropylene. Among these, polyester is preferred, and polyethylene terephthalate is more preferred.

[0072] The thickness of the substrate affects the rigidity of the surface protection sheet, so it should be set according to the material of the substrate. In this embodiment, the thickness of the substrate is preferably 25 μm or more and 200 μm or less, more preferably 35 μm or more and 150 μm or less, and even more preferably 40 μm or more and 150 μm or less.

[0073] At least one main surface of the substrate may be subjected to an adhesive treatment such as corona treatment to improve adhesion with the layer formed on the main surface. Alternatively, an easy-adhesion layer may be formed on at least one main surface of the substrate to improve adhesion with the layer formed on the main surface.

[0074] (1.3.2. Adhesive layer) The adhesive layer is attached to the surface of the workpiece (the surface on which the circuit and bump electrodes are formed) and protects the surface and supports the workpiece until it is peeled off. In this embodiment, the adhesive layer can contact the bump electrodes formed on the surface of the workpiece and fix the bump electrodes to the surface protection sheet. As a result, even if the workpiece is ground very thin and force is applied to the bump electrodes, etc., the surface protection sheet can adequately protect the bump electrodes, etc. Furthermore, even if the workpiece is fragmented, contact between the fragmented workpiece pieces can be suppressed. The adhesive layer may consist of one layer (single layer) or of two or more layers.

[0075] The thickness of the adhesive layer is not particularly limited as long as it is thick enough to adequately support the workpiece. In this embodiment, the thickness of the adhesive layer is preferably 1 μm or more and 100 μm or less, and more preferably 5 μm or more and 50 μm or less. Note that the thickness of the adhesive layer refers to the total thickness of the adhesive layer. For example, the thickness of an adhesive layer composed of multiple layers refers to the total thickness of all the layers constituting the adhesive layer.

[0076] The composition of the adhesive layer is not limited as long as it has sufficient tackiness to protect the surface of the workpiece. In this embodiment, the adhesive layer is preferably composed of, for example, an acrylic adhesive, a urethane adhesive, a rubber adhesive, a silicone adhesive, or the like.

[0077] Furthermore, it is preferable that the adhesive layer be formed from an energy-ray curable adhesive. By forming the adhesive layer of the surface protection sheet from an energy-ray curable adhesive, it adheres to the workpiece with high adhesive strength when attached, and when peeled off from the workpiece, the adhesive strength can be reduced by irradiation with energy rays. Therefore, while appropriately protecting the circuits of the workpiece, etc., it is prevented from damaging the circuits on the workpiece surface or transferring the adhesive onto the workpiece when peeling off the surface protection sheet.

[0078] In this embodiment, the energy-ray curable adhesive is preferably composed of an adhesive composition containing an acrylic adhesive. It is preferable to use an acrylic polymer as the acrylic adhesive.

[0079] Any known acrylic polymer may be used as the acrylic polymer, but in this embodiment, a functional group-containing acrylic polymer is preferred. The functional group-containing acrylic polymer may be a homopolymer formed from one type of acrylic monomer, a copolymer formed from multiple types of acrylic monomers, or a copolymer formed from one or more types of acrylic monomers and monomers other than acrylic monomers.

[0080] In this embodiment, the functional group-containing acrylic polymer is preferably an acrylic copolymer obtained by copolymerizing an alkyl (meth)acrylate with a functional group-containing monomer.

[0081] Examples of alkyl (meth)acrylates include methyl (meth)acrylate, ethyl (meth)acrylate, n-propyl (meth)acrylate, isopropyl (meth)acrylate, n-butyl (meth)acrylate, isobutyl (meth)acrylate, t-butyl (meth)acrylate, n-pentyl (meth)acrylate, n-hexyl (meth)acrylate, 2-ethylhexyl (meth)acrylate, isooctyl (meth)acrylate, and n-octyl (meth)acrylate.

[0082] Functional group-containing monomers are monomers that contain reactive functional groups. Reactive functional groups are functional groups that can react with other compounds, such as crosslinking agents, which will be described later. Examples of functional groups in functional group-containing monomers include hydroxyl groups, carboxyl groups, and epoxy groups, with hydroxyl groups being preferred.

[0083] Examples of hydroxyl group-containing monomers include hydroxyalkyl (meth)acrylates such as hydroxymethyl (meth)acrylate, 2-hydroxyethyl (meth)acrylate, 2-hydroxypropyl (meth)acrylate, 3-hydroxypropyl (meth)acrylate, 2-hydroxybutyl (meth)acrylate, 3-hydroxybutyl (meth)acrylate, and 4-hydroxybutyl (meth)acrylate; and non-(meth)acrylic unsaturated alcohols such as vinyl alcohol and allyl alcohol (unsaturated alcohols that do not have a (meth)acryloyl skeleton).

[0084] The adhesive composition preferably further contains an energy-ray curable compound having an energy-ray curable group. The energy-ray curable compound having an energy-ray curable group is preferably a compound having one or more selected from isocyanate groups, epoxy groups, and carboxyl groups, and more preferably a compound having an isocyanate group.

[0085] Examples of compounds having an isocyanate group include 2-methacryloyloxyethyl isocyanate, meta-isopropenyl-α,α-dimethylbenzyl isocyanate, methacryloyl isocyanate, allyl isocyanate, 1,1-(bisacryloyloxymethyl)ethyl isocyanate; acryloyl monoisocyanate compounds obtained by the reaction of a diisocyanate compound or polyisocyanate compound with hydroxyethyl (meth)acrylate; and acryloyl monoisocyanate compounds obtained by the reaction of a diisocyanate compound or polyisocyanate compound with a polyol compound with hydroxyethyl (meth)acrylate. The isocyanate group undergoes an addition reaction to the hydroxyl group of a functional group-containing acrylic polymer.

[0086] The adhesive composition preferably further contains a crosslinking agent. The crosslinking agent, for example, reacts with the functional group to crosslink the resins contained in the functional group-containing acrylic polymer.

[0087] Examples of crosslinking agents include isocyanate-based crosslinking agents (crosslinking agents having an isocyanate group) such as tolylene diisocyanate, hexamethylene diisocyanate, xylylene diisocyanate, and adducts of these diisocyanates; epoxy-based crosslinking agents (crosslinking agents having a glycidyl group) such as ethylene glycol glycidyl ether; aziridine-based crosslinking agents (crosslinking agents having an aziridinyl group) such as hexa[1-(2-methyl)-aziridinyl]triphosphatriaidine; metal chelate-based crosslinking agents (crosslinking agents having a metal chelate structure) such as aluminum chelate; and isocyanurate-based crosslinking agents (crosslinking agents having an isocyanuric acid skeleton).

[0088] For reasons such as improving the cohesive force of the adhesive and thus the adhesive strength of the adhesive layer, and because they are readily available, the crosslinking agent is preferably an isocyanate-based crosslinking agent.

[0089] The adhesive composition may further contain a photopolymerization initiator. The inclusion of a photopolymerization initiator in the adhesive composition allows the curing reaction to proceed sufficiently even when irradiated with relatively low-energy rays such as ultraviolet light.

[0090] Examples of photopolymerization initiators include benzoin compounds, acetophenone compounds, acylphosphine oxide compounds, titanocene compounds, thioxanthone compounds, and peroxide compounds, as well as photosensitizers such as amines and quinones. Specifically, examples include α-hydroxycyclohexylphenyl ketone, benzoin, benzoin methyl ether, benzoin ethyl ether, benzoin isopropyl ether, benzyl diphenyl sulfide, benzyl dimethyl ketal, tetramethylthiuram monosulfide, azobisisobutyronitrile, dibenzyl, diacetyl, β-chloranthraquinone, and 2,4,6-trimethylbenzoyldiphenylphosphine oxide.

[0091] Figure 5B shows a surface protection sheet 10 according to another embodiment having a configuration in which an intermediate layer 12 and an adhesive layer 13 are laminated in that order on a substrate 11.

[0092] The surface protection sheet 10 is not limited to the configuration shown in Figure 5B and may have other layers. That is, other layers may be formed between the substrate 11 and the intermediate layer 12, or other layers may be formed between the intermediate layer 12 and the adhesive layer 13.

[0093] The components of the surface protection sheet 1 shown in Figure 5B will be described in detail below. The base material and adhesive layer are the same as described above, so their explanation will be omitted.

[0094] (1.3.3. Middle Class) As shown in Figure 5B, the intermediate layer 12 is a layer placed between the substrate 11 and the adhesive layer 13. In this embodiment, the intermediate layer may consist of one layer (single layer) or of two or more layers.

[0095] The thickness of the intermediate layer should be set considering the height of the bump electrodes on the semiconductor wafer. In this embodiment, the thickness of the intermediate layer is preferably 50 μm or more and 600 μm or less, and more preferably 150 μm or more and 500 μm or less. Note that the thickness of the intermediate layer refers to the total thickness of the intermediate layer. For example, the thickness of an intermediate layer composed of multiple layers refers to the total thickness of all the layers that make up the intermediate layer.

[0096] The composition of the intermediate layer is not particularly limited, but in this embodiment, the intermediate layer is preferably composed of a resin-containing composition (intermediate layer composition). The intermediate layer composition preferably contains the following components.

[0097] (1.3.3.1. Urethane (meth)acrylate) Urethane (meth)acrylate is a compound having at least a (meth)acryloyl group and a urethane bond, and has the property of polymerizing upon energy ray irradiation. In this embodiment, urethane (meth)acrylate is a component that provides flexibility to the intermediate layer and securely fixes the bump electrode.

[0098] The urethane (meth)acrylate may be monofunctional or polyfunctional. In this embodiment, polyfunctional urethane (meth)acrylate is preferred, and from the viewpoint of securely fixing the bump electrode, bifunctional urethane (meth)acrylate is preferred.

[0099] The urethane (meth)acrylate may be an oligomer, a polymer, or a mixture thereof. In this embodiment, a urethane (meth)acrylate oligomer is preferred.

[0100] Urethane (meth)acrylate can be obtained, for example, by reacting a terminal isocyanate urethane prepolymer, which is obtained by reacting a polyol compound with a polyvalent isocyanate compound, with a (meth)acrylate having a hydroxyl group. Note that one or more types of urethane (meth)acrylate may be used.

[0101] The content of urethane (meth)acrylate in the intermediate layer composition is preferably 20% by mass or more, more preferably 25% by mass or more, and even more preferably 30% by mass or more. Furthermore, the content of urethane (meth)acrylate in the intermediate layer composition is preferably 70% by mass or less, more preferably 65% ​​by mass or less, and even more preferably 50% by mass or less.

[0102] (1.3.3.2. Polymerizable monomers) The polymerizable monomer is preferably a polymerizable compound other than the urethane (meth)acrylate mentioned above, and is a compound that can be polymerized with other components by irradiation with energy rays. In this embodiment, the polymerizable monomer is a compound having one reactive unsaturated double bond group.

[0103] Examples of polymerizable monomers include (meth)acrylates having alkyl groups with 1 to 30 carbon atoms; (meth)acrylates having functional groups such as hydroxyl groups, amide groups, amino groups, and epoxy groups; (meth)acrylates having an alicyclic structure; (meth)acrylates having an aromatic structure; (meth)acrylates having a heterocyclic structure; and vinyl compounds such as styrene, hydroxyethyl vinyl ether, hydroxybutyl vinyl ether, N-vinylformamide, N-vinylpyrrolidone, and N-vinylcaprolactam.

[0104] Examples of (meth)acrylates having alkyl groups with 1 to 30 carbon atoms include methyl (meth)acrylate, ethyl (meth)acrylate, n-propyl (meth)acrylate, isopropyl (meth)acrylate, n-butyl (meth)acrylate, isobutyl (meth)acrylate, t-butyl (meth)acrylate, n-pentyl (meth)acrylate, n-hexyl (meth)acrylate, 2-ethylhexyl (meth)acrylate, n-octyl (meth)acrylate, nonyl (meth)acrylate, decyl (meth)acrylate, undecyl (meth)acrylate, dodecyl (meth)acrylate, tridecyl (meth)acrylate, tetradecyl (meth)acrylate, hexadecyl (meth)acrylate, octadecyl (meth)acrylate, and eicosyl (meth)acrylate.

[0105] Examples of functional group-containing (meth)acrylates include hydroxyl group-containing (meth)acrylates such as 2-hydroxyethyl (meth)acrylate, 2-hydroxypropyl (meth)acrylate, 3-hydroxypropyl (meth)acrylate, 2-hydroxybutyl (meth)acrylate, 3-hydroxybutyl (meth)acrylate, and 4-hydroxybutyl (meth)acrylate; (meth)acrylamide, N,N-dimethyl (meth)acrylamide, N-butyl (meth)acrylamide, and N-methylol (meth)acrylamide. Examples include amide group-containing compounds such as crillamide, N-methylolpropane(meth)acrylamide, N-methoxymethyl(meth)acrylamide, and N-butoxymethyl(meth)acrylamide; amino group-containing (meth)acrylates such as primary amino group-containing (meth)acrylates, secondary amino group-containing (meth)acrylates, and tertiary amino group-containing (meth)acrylates; and epoxy group-containing (meth)acrylates such as glycidyl(meth)acrylate, methylglycidyl(meth)acrylate, and allylglycidyl ether.

[0106] Examples of (meth)acrylates having an alicyclic structure include isobornyl (meth)acrylate, dicyclopentenyl (meth)acrylate, dicyclopentanyl (meth)acrylate, dicyclopentenyloxy (meth)acrylate, cyclohexyl (meth)acrylate, trimethylcyclohexyl (meth)acrylate, and adamantane (meth)acrylate.

[0107] Examples of (meth)acrylates having an aromatic structure include phenylhydroxypropyl (meth)acrylate, benzyl (meth)acrylate, and 2-hydroxy-3-phenoxypropyl (meth)acrylate.

[0108] Examples of (meth)acrylates having a heterocyclic structure include tetrahydrofurfuryl(meth)acrylate and morpholine(meth)acrylate.

[0109] In this embodiment, the polymerizable monomer preferably includes (meth)acrylates having an alkyl group with 1 to 30 carbon atoms and (meth)acrylates having an alicyclic structure. From the viewpoint of reliably fixing the bump electrode, (meth)acrylates having an alkyl group with 4 to 14 carbon atoms are preferred, and as (meth)acrylates having an alicyclic structure, isobornyl (meth)acrylate and trimethylcyclohexyl (meth)acrylate are preferred.

[0110] Furthermore, if the intermediate layer composition contains a crosslinking agent, (meth)acrylates having functional groups that can react with the crosslinking agent are undesirable. This is because the crosslinked structure formed by the crosslinking reaction may excessively harden the intermediate layer. For example, an intermediate layer composition containing a polyisocyanate-based crosslinking agent and a (meth)acrylate having hydroxyl groups is undesirable.

[0111] The content of polymerizable monomers in the intermediate layer composition is preferably 20% by mass or more, and more preferably 30% by mass or more. Furthermore, the content of polymerizable monomers in the intermediate layer composition is preferably 80% by mass or less, and more preferably 70% by mass or less.

[0112] Furthermore, the mass ratio of urethane (meth)acrylate to polymerizable monomer (urethane (meth)acrylate / polymerizable monomer) in a total of 100 parts by mass of urethane (meth)acrylate and polymerizable monomer is preferably 20 / 80 to 80 / 20, and more preferably 30 / 70 to 70 / 30.

[0113] (1.3.3.3. Photopolymerization Initiators) When the intermediate layer composition contains the above-mentioned urethane (meth)acrylate and polymerizable monomer, it is preferable that the intermediate layer composition also contains a photopolymerization initiator. By including a photopolymerization initiator, polymerization proceeds reliably, and the intermediate layer can be easily obtained.

[0114] Examples of photopolymerization initiators include benzoin compounds, acetophenone compounds, acyl phosphinoxide compounds, titanocene compounds, thioxanthone compounds, and peroxide compounds, as well as photosensitizers such as amines and quinones. Specifically, examples include 1-hydroxycyclohexylphenyl ketone, 2-hydroxy-2-methyl-1-phenylpropan-1-one, benzoin, benzoin methyl ether, benzoin ethyl ether, benzoin isopropyl ether, and 2,2-dimethoxy-1,2-diphenylethane-1-one. These photopolymerization initiators may be used individually or in combination of two or more.

[0115] The amount of photopolymerization initiator added is preferably 0.05 parts by mass or more and 15 parts by mass or less, and more preferably 0.5 parts by mass or more and 10 parts by mass or less, based on 100 parts by mass of the total of urethane (meth)acrylate and polymerizable monomer.

[0116] (1.3.3.4. Chain Movement Agent) The intermediate layer composition preferably contains a chain transfer agent. The chain transfer agent can induce a chain transfer reaction and regulate the progress of the curing reaction of the intermediate layer composition. By including a chain transfer agent, components with relatively short molecular chains can remain even after curing, so that the cured polymer has a relatively flexible crosslinked structure.

[0117] Examples of chain transfer agents include thiol group-containing compounds. Examples of thiol group-containing compounds include nonyl mercaptan, 1-dodecanethiol, 1,2-ethanedithiol, 1,3-propanedithiol, triazinethiol, triazinedithiol, triazinetrithiol, 1,2,3-propanetrithiol, tetraethylene glycol-bis(3-mercaptopropionate), trimethylolpropanetris(3-mercaptopropionate), pentaerythritoltetrakis(3-mercaptopropionate), and pentaerythritol. Examples include lysritol tetrakisthioglucolate, dipentaerythritol hexakis(3-mercaptopropionate), tris[(3-mercaptopropionyloxy)-ethyl]-isocyanurate, 1,4-bis(3-mercaptobutyryloxy)butane, pentaerythritol tetrakis(3-mercaptobutyrate), and 1,3,5-tris(3-mercaptobutyloxyethyl)-1,3,5-triazine-2,4,6-(1H,3H,5H)-trione. The chain transfer agent may be used individually or in combination of two or more types.

[0118] The amount of chain transfer agent is preferably 0.1 parts by mass or more and 10 parts by mass or less, and more preferably 0.3 parts by mass or more and 5 parts by mass or less, based on 100 parts by mass of the total of urethane (meth)acrylate and polymerizable monomer.

[0119] (1.4. Manufacturing of surface protection sheets) Surface protection sheets can be manufactured by known methods. For example, first, the adhesive composition described above is prepared. Next, the prepared composition is applied to the surface of a release sheet, substrate, etc., using a known coating method, and then drying, curing, etc., to obtain a surface protection sheet consisting of an adhesive layer and a substrate having a predetermined structure. Alternatively, the intermediate layer composition and adhesive composition described above are prepared. Next, the prepared compositions are applied to the surface of a release sheet, substrate, etc., using a known coating method, and then drying, curing, etc., to obtain a surface protection sheet consisting of an adhesive layer, an intermediate layer and a substrate.

[0120] (1.5.Process 2) As shown in Figure 6(A), a convex pattern 3 is formed on the surface of the workpiece 1, and a surface protection sheet 10 is attached to the surface of the workpiece. Then, the back surface 1b of the workpiece is ground.

[0121] As shown in Figure 6(B), the workpiece 1 with the surface protection sheet 10 attached is placed on the chuck table 100 with its surface side facing up. The chuck table 100 has, for example, a porous holding surface, and the surface protection sheet 10 is attracted to and fixed to the chuck table 100 by suction from the side opposite to the side on which the workpiece 1 is placed.

[0122] At this time, since a convex pattern 3 is formed on the workpiece 1, the height difference of the surface protection sheet 10 between the inner circumference and the outer circumference of the workpiece 1 becomes small. Therefore, even when suction is applied, the outer circumference of the workpiece 1 is hardly deformed and is adsorbed onto the chuck table 100.

[0123] After the workpiece 1 is fixed to the chuck table 100, the back surface 1b of the workpiece 1 is ground using, for example, a grinding wheel. The thickness of the workpiece after back surface grinding is, for example, about 25 μm to 600 μm.

[0124] After grinding, when the suction is released, the workpiece 1 and the surface protection sheet 10 are released from the chuck table 100. In this embodiment, since process A is adopted in process 1, the thickness of the workpiece 1 becomes almost uniform from the inner circumference to the outer circumference, as shown in Figure 6(C). As a result, the TTV of the workpiece 1 after grinding is suppressed to a small extent.

[0125] (1.6.Step 3) In this embodiment, it is preferable to peel off the surface protection sheet from the workpiece after back grinding. That is, the workpiece processing method according to this embodiment preferably includes a step (step 3) of peeling off the surface protection sheet from the workpiece after back grinding. Step 3 is carried out, for example, as follows.

[0126] If the adhesive layer of the surface protection sheet is formed from an energy-ray curable adhesive, the adhesive layer is cured and shrunk by irradiation with energy rays, reducing its adhesive strength to the workpiece. Next, a pickup tape is attached to the back side of the workpiece after backside grinding, and its position and orientation are adjusted so that it can be picked up. At this time, the ring frame positioned on the outer circumference of the workpiece is also attached to the pickup tape, and the outer edge of the pickup tape is fixed to the ring frame. The workpiece and ring frame may be attached to the pickup tape at the same time, or at different times. Next, the surface protection sheet is peeled off from the workpiece held on the pickup tape.

[0127] In this embodiment, when the surface protection sheet is peeled off, it is preferable that at least a portion of the convex pattern formed on the surface of the workpiece is peeled off together with the surface protection sheet, and it is particularly preferable that the entire convex pattern is peeled off.

[0128] At least a portion of the convex pattern is peeled off together with the surface protective sheet, making it easier to remove the convex pattern from the workpiece surface.

[0129] Furthermore, the workpiece processing method according to this embodiment is also applicable to DBG or LDBG. In this case, it is preferable to use the surface protection sheet 10 shown in Figure 5C as the surface protection sheet. In the surface protection sheet 10 shown in Figure 5C, a buffer layer 14 is formed on the main surface 11b of the base material 11, which is opposite to the main surface on which the adhesive layer 13 is formed.

[0130] The buffer layer is a softer layer compared to the base material, and it relieves stress during back-side grinding of the workpiece, preventing cracking and chipping. Furthermore, the presence of the buffer layer makes it easier for the workpiece to be properly held on the chuck table.

[0131] The thickness of the buffer layer is preferably 1 to 200 μm, more preferably 10 to 100 μm, and even more preferably 20 to 80 μm. By setting the thickness of the buffer layer within the above range, the buffer layer can appropriately relieve stress during backside grinding.

[0132] The buffer layer may be a layer formed from a buffer layer composition containing an energy ray polymerizable compound, or it may be a film such as a polypropylene film, an ethylene-vinyl acetate copolymer film, an ionomer resin film, an ethylene-(meth)acrylic acid copolymer film, an ethylene-(meth)acrylic acid ester copolymer film, an LDPE film, or an LLDPE film.

[0133] When the workpiece processing method according to this embodiment is applied to DBG or LDBG, in addition to steps 1 to 3 above, the method includes a step (step 4) of forming a groove from the surface side of the workpiece, or forming a modified region inside the workpiece from the surface or back surface.

[0134] When forming a modified region on the workpiece, it is preferable to perform step 1 (and step A) before step 4. On the other hand, when forming grooves on the surface of the workpiece by dicing or the like, step 1 (and step A) is performed after step 4. That is, in step 1, a surface protection sheet is applied to the surface of the workpiece having grooves formed in step 4, which will be described later.

[0135] In step 4, grooves are formed from the surface side of the workpiece. Alternatively, a modified region is formed inside the workpiece from either the surface or back surface.

[0136] The grooves formed in this process are shallower than the thickness of the workpiece. The grooves can be formed by dicing using conventionally known dicing equipment. In addition, in step 3 described above, the workpiece is divided into multiple chips along the grooves by back grinding.

[0137] Furthermore, the modified region is a brittle part of the workpiece, and it is the starting point for the modified region of the workpiece to break down and be fragmented into chips due to the thinning of the workpiece by grinding during the grinding process or the force applied by grinding. In other words, in step 4, the groove and the modified region are formed along the dividing line that occurs when the workpiece is divided and fragmented into chips in step 3 described above.

[0138] The modified region is formed by irradiating the inside of the workpiece with a laser focused on it, and the modified region is formed inside the workpiece. The laser irradiation may be performed from either the front or back side of the workpiece. In the embodiment in which the modified region is formed, if step 4 is performed after step 1 and the laser irradiation is performed from the surface of the workpiece, the laser will be irradiated onto the workpiece through a surface protection sheet.

[0139] In DBG, the back grinding in step 2 is performed to thin the workpiece at least to the bottom of the groove. This back grinding creates a groove that penetrates the workpiece, dividing the workpiece into individual chips.

[0140] In LDBG, the grinding surface (back surface of the workpiece) may reach the modified region through grinding, but it does not need to reach the modified region precisely. In other words, grinding should be done up to a position close to the modified region so that the workpiece is broken down into individual chips starting from the modified region. For example, the actual fragmentation of the chips may be performed by applying a pickup tape, as described later, and then stretching the pickup tape.

[0141] Additionally, after the backside grinding is complete and prior to picking up the chip, dry polishing may be performed.

[0142] Step 3 can be carried out as described above, but the surface protection sheet will be peeled off from the individualized workpiece. Even in this case, damage to the individualized workpiece is suppressed.

[0143] The shape of the individualized chips (workpiece fragments) obtained through step 2 may be rectangular or elongated, such as a rectangle. The thickness of the individualized chips is not particularly limited, but is preferably around 5 to 100 μm, and more preferably 10 to 45 μm. LDBG makes it easy to set the thickness of the individualized semiconductor chips to 50 μm or less, more preferably 10 to 45 μm. The size of the individualized chips is not particularly limited, but the chip size is preferably 600 mm. 2 Less than, more preferably 400 mm 2 Less than 120 mm, more preferably 120 mm 2 It is less than.

[0144] According to the workpiece processing method of this embodiment, by performing step A in step 1, the TTV of the workpiece after back grinding can be reduced. Furthermore, by devising the shape of the convex pattern, it is possible to reduce the costs associated with forming and discarding the convex pattern.

[0145] Although embodiments of the present invention have been described above, the present invention is not limited in any way to the embodiments described above, and may be modified in various ways within the scope of the present invention. [Examples]

[0146] The invention will be described in more detail below using examples, but the present invention is not limited to these examples.

[0147] (1) Convex pattern The following energy-ray curable resin was used as the material for forming the convex pattern.

[0148] (Energy ray curable resin A) Energy-ray curable resin A was obtained by combining 40 parts by mass of a bifunctional polyol-modified urethane acrylate (molecular weight 7300), 60 parts by mass of isobornyl acrylate, and 0.5 parts by mass of a photopolymerization initiator (Omnirad 1173, manufactured by IGM Resin).

[0149] (2) Surface protection sheet As the surface protection sheet, the surface protection sheet shown below was used.

[0150] (Surface protection sheet A) 48 parts by mass of urethane acrylate oligomer (manufactured by Arkema Co., Ltd., CN9021 NS), and 3.4 parts by mass of a photopolymerization initiator (manufactured by IGM Resin, Omnirad1173) and 1.0 part by mass of a chain transfer agent (manufactured by Showa Denko K.K., Karenz MT PE1) were blended with respect to a total of 100 parts by mass of 26 parts by mass of isobornyl acrylate, 16 parts by mass of trimethylcyclohexyl acrylate, and 10 parts by mass of lauryl acrylate to obtain an intermediate layer composition.

[0151] The obtained intermediate layer composition was applied onto a PET film (manufactured by Toray Industries, Inc., thickness 75 μm) which is a base material so as to have a thickness of 400 μm by a knife coater method to form an intermediate layer composition layer. Further, a PET-based release film (manufactured by Lintec Corporation, SP-PET752150, thickness 75 μm) was laminated onto the intermediate layer composition layer immediately after coating. Then, using a high-pressure mercury lamp, ultraviolet irradiation was performed on the intermediate layer composition layer under the conditions of an illuminance of 120 mW / cm 2 and a light quantity of 240 mJ / cm 2 . Further, using a metal halide lamp, ultraviolet irradiation was performed under the conditions of an illuminance of 330 mW / cm 2 and a light quantity of 1260 mJ / cm 2 to cure the intermediate layer composition layer, thereby forming an intermediate layer with a thickness of 400 μm on the PET film which is a base material.

[0152] Next, 80 parts by mass of 2-ethylhexyl acrylate (2EHA) and 20 parts by mass of 2-hydroxyethyl acrylate (HEA) were copolymerized to an acrylic copolymer. To this copolymer, 2-isocyanate ethyl methacrylate (MOI) was added at an addition rate of 80 equivalents relative to the hydroxyl groups (100 equivalents) derived from HEA. To 100 parts by mass of this energy-curable acrylic copolymer (weight-average molecular weight: 800,000), 1.5 parts by mass of trimethylolpropane adduct tolylene diisocyanate (Tosoh Corporation, Coronate L) was added as a crosslinking agent, and 2.2 parts by mass of 2,2-dimethoxy-2-phenylacetophenone (IGM Resin Corporation, Omnirad651) was added as a photopolymerization initiator. Toluene was then added to adjust the solid content concentration to 30%, and the mixture was stirred for 30 minutes to prepare the adhesive composition.

[0153] Next, a solution of the prepared adhesive composition was applied to a PET release film (Lintec Corporation, SP-PET382150, 38 μm thick), dried to form a 10 μm thick adhesive layer, and an adhesive sheet was produced.

[0154] The release film on the intermediate layer obtained above was removed, and the intermediate layer and the adhesive layer of the adhesive sheet were bonded together to produce a surface protection sheet A having the structure of a substrate / intermediate layer / adhesive layer.

[0155] (Example 1) On a 720 μm thick wafer with bump electrodes having a bump electrode height of 250 μm and a pitch of 500 μm, energy-ray curable resin A was applied using a dispenser (Shotmaster 350DS, manufactured by Musashi Engineering Co., Ltd.) as shown in Figure 4B. A raised portion P1 was applied to the outermost circumference, with a cured height of 100 μm and width of 0.5 mm. A raised portion P2 was applied to the inner circumference, with a cured height of 50 μm and width of 0.5 mm. A raised portion P3 was applied to the innermost circumference, with a cured height of 30 μm and width of 0.5 mm. After application, each raised portion P1, P2, and P3 was irradiated with a high-pressure mercury lamp at an irradiance of 150 mW / cm². 2 , irradiation amount 400mJ / cm 2Under these conditions, ultraviolet light was irradiated to harden each protrusion, forming a convex pattern.

[0156] After forming the convex pattern on the wafer, a surface protection sheet A was applied using a tape laminator (Lintec Corporation, RAD-3510F / 12) at a lamination table temperature of 65°C.

[0157] After applying surface protection sheet A, the wafer was placed on the chuck table so that surface protection sheet A was in contact with the chuck table, and the wafer was fixed to the chuck table by suction. Next, the back surface of the wafer was ground with a grinding wheel, and the grinding was stopped when the wafer thickness reached 300 μm. After the grinding was completed, the TTV evaluation shown below was performed.

[0158] (TTV evaluation of the outer edge of the wafer after grinding) The wafer thickness was measured using a digital microscope (Keyence VHX-7000) in a 15mm section from the edge of the ground wafer inward. The maximum and minimum values ​​were selected, and the difference between the maximum and minimum values ​​was calculated. This measurement was performed at four locations on the ground wafer (top, bottom, right, and left – positions where the wafer was rotated 90°), and the average difference between the maximum and minimum values ​​was calculated. The results are shown in Table 1.

[0159] (Examples 2-5 and Comparative Example 1) Backside grinding and evaluation of a wafer with bump electrodes were performed using the same method as in Example 1, except that the heights of each protrusion P1, P2, and P3 constituting the convex pattern were set to the values ​​shown in Table 1. The results are shown in Table 1.

[0160] (Example 6) On a 720 μm thick wafer with bump electrodes of 250 μm height and 500 μm pitch, energy-ray curable resin A was applied to the outermost periphery using a dispenser (Shotmaster 350DS, Musashi Engineering Co., Ltd.) to a thickness of 100 μm and 2 mm. After application, the coating was deformed by blowing air from the inside to the outside of the wafer using a blower, and then irradiated with a high-pressure mercury lamp at an illuminance of 150 mW / cm². 2 , irradiation amount 400mJ / cm 2 The coating film was cured by irradiating it with ultraviolet light under the specified conditions, forming a convex pattern with the shape shown in Figure 4E. Observation of the cross-section of the convex pattern revealed that the maximum height H1 was 130 μm and the average height H2 was 90 μm, confirming that it was a single-layer structure. Backside grinding and evaluation of the wafer with bump electrodes were performed using the same method as in Example 1. The results are shown in Table 1.

[0161] (Example 7) Except for the fact that, during deformation of the coating film in Example 6, an electrostatic generator was placed outside the wafer to generate a potential difference of 100 kV, and the coating film was deformed by generating static electricity, the same procedure as in Example 6 was used to form a convex pattern with the shape shown in Figure 4E. Observation of the cross-section of the convex pattern revealed that the maximum height H1 was 130 μm and the average height H2 was 90 μm, confirming that it was a single-layer structure. Backside grinding and evaluation of the wafer with bump electrodes were performed using the same method as in Example 1. The results are shown in Table 1.

[0162] [Table 1]

[0163] Table 1 shows that when a surface protection sheet is applied to a wafer with bump electrodes and the wafer is back-ground, the TTV of the wafer after back-ground grinding can be reduced by forming a convex pattern whose height is appropriately set according to the height of the bump electrodes. Furthermore, by adjusting the maximum and average height of the convex pattern, sufficient effect can be obtained even if the amount of resin used to form the convex pattern is reduced, and the costs required for forming and disposing of the convex pattern can be reduced. [Explanation of Symbols]

[0164] 1…Work 2…Bump electrodes 3…Convex pattern 10…Surface protection sheet 11...Base material 12…Middle class 13…Adhesive layer 14...Buffer layer 20…Surface protective sheet

Claims

1. A process of attaching a surface protection sheet to the surface of a workpiece that has a front and back surface, The process includes grinding the back surface of the workpiece to which the surface protection sheet is attached, The surface of the workpiece has bump electrodes, In the process of applying the surface protection sheet, a convex pattern is formed on the outer periphery of the workpiece surface in a region different from the region where the bump electrode is formed. The relationship between the maximum height H1 of the convex pattern, the average height H2 of the convex pattern, and the average height Hb of the bump electrode satisfies H2 < H1 < 2Hb / 3. A method for processing a workpiece, wherein the surface protection sheet is attached to the top of a convex pattern.

2. The method for machining a workpiece according to claim 1, wherein the relationship between the maximum height H1 of the convex pattern and the average height Hb of the bump electrode satisfies 0 < H1 < Hb / 4.

3. The method for processing a workpiece according to claim 1, wherein the convex pattern is made of resin.

4. The process further includes a step of peeling off the surface protection sheet from the workpiece after backside grinding, The method for processing a workpiece according to claim 1, wherein at least a portion of the convex pattern is peeled off together with the surface protective sheet.

5. The process includes a step of forming grooves on the surface of the workpiece before the step of attaching a surface protection sheet to the surface of the workpiece. A method for processing a workpiece according to any one of claims 1 to 4, wherein, in the step of grinding the back surface of the workpiece, the workpiece is fragmented into a plurality of workpiece fragments, starting from a groove.

6. The process includes a step of forming a modified region inside the workpiece before the step of grinding the back surface of the workpiece. A method for processing a workpiece according to any one of claims 1 to 4, wherein, in the step of grinding the back surface of the workpiece, the workpiece is fragmented into a plurality of workpiece fragments starting from the modified region.

7. A workpiece having bump electrodes on its surface, The workpiece surface has a convex pattern in a region different from the region where the bump electrode is formed, The relationship between the maximum height H1 of the convex pattern, the average height H2 of the convex pattern, and the average height Hb of the bump electrode satisfies H2 < H1 < 2Hb / 3. Workpiece with a convex pattern.

8. The workpiece with a convex pattern according to claim 7, wherein the convex pattern has multiple layers.

9. The workpiece with a convex pattern according to claim 7, wherein the convex pattern is substantially a single layer.

Images