JP7764779B2 - adhesive tape - Google Patents

Description

The present invention relates to an adhesive tape used to temporarily fix substrates and components.

In response to the recent trend toward more sophisticated electronic devices and the expansion into mobile applications, there has been a growing demand for higher density and integration of semiconductor devices, leading to the development of larger capacity, higher density IC packages.

The manufacturing method for these semiconductor devices involves, for example, first applying adhesive tape to a semiconductor substrate (semiconductor wafer) as a substrate, and then using a dicing saw to cut the semiconductor substrate in the thickness direction while securing the periphery of the semiconductor substrate with a wafer ring, thereby cutting and separating (singulating) the semiconductor substrate into individual semiconductor elements (semiconductor chips). Next, an expanding process is performed in which the adhesive tape is radially stretched using the wafer ring to form gaps between adjacent semiconductor elements. This is followed by a pickup process in which the diced semiconductor elements are picked up while being pushed up with a needle. The picked-up semiconductor elements are then transferred to a mounting process in which they are mounted on a metal lead frame or substrate (e.g., a tape substrate, an organic hard substrate, etc.). In the mounting process, the picked-up semiconductor elements are adhered to the lead frame or substrate via, for example, an underfill material, and then encapsulated on the lead frame or substrate to produce a semiconductor device.

In recent years, various studies have been conducted on the adhesive tape (dicing tape) used in the manufacture of such semiconductor devices (see, for example, Patent Document 1).

This adhesive tape generally has a substrate (film substrate) and an adhesive layer formed on the substrate, with the semiconductor substrate secured by the adhesive layer. With adhesive tapes of this configuration, a pickup process for picking up semiconductor elements is carried out after the dicing process for dicing the semiconductor substrate, as in the semiconductor device manufacturing method described above. That is, in the dicing process, a disk-shaped dicing saw is used to separate the semiconductor substrate into individual pieces to obtain semiconductor elements. In the pickup process, energy is applied to the adhesive layer to reduce its adhesive strength, and then a needle is used to lift the semiconductor element. While maintaining this state, the semiconductor element is picked up by suction using a vacuum collet or air tweezers, for example.

However, when manufacturing semiconductor devices using this adhesive tape, the dicing process, in which a dicing saw is used to separate a semiconductor substrate into individual semiconductor elements, poses the following problems. Specifically, during the dicing process, when the semiconductor substrate is diced (cut) in the thickness direction using a dicing saw to separate the semiconductor substrate into individual semiconductor elements, the cutting of the semiconductor substrate is generally continued until it reaches the middle of the thickness of the base material of the adhesive tape, in order to ensure reliable separation of the semiconductor substrate. Therefore, the adhesive layer of the adhesive tape located closer to the semiconductor substrate than the base material is naturally cut by the dicing saw when the semiconductor substrate is cut. This poses a problem in that portions of the cut adhesive layer become glued to the side surfaces of the semiconductor elements obtained by separating the semiconductor substrate.

JP 2009-245989 A

The present invention aims to provide an adhesive tape that can accurately suppress or prevent glue adhesion, which occurs when a semiconductor wafer serving as a substrate attached to the adhesive tape is cut in the thickness direction to obtain semiconductor chips as components. This occurs when part of the adhesive layer of the adhesive tape adheres to the side of the resulting semiconductor chip.

These objects can be achieved by the present invention described in (1) to (12) below.
(1) An adhesive tape used for temporarily fixing a semiconductor wafer as a substrate and a component, the adhesive tape comprising a base material and an adhesive layer laminated on one surface of the base material, the semiconductor wafer being fixed thereto, the semiconductor wafer being diced into individual pieces to form semiconductor chips as a plurality of components, and the semiconductor chips being then detached from the adhesive layer,
the adhesive layer contains a base resin having adhesiveness and a curable resin that is cured by the application of energy, and the adhesive strength of the adhesive layer is reduced by applying energy to the adhesive layer to cure the adhesive layer;
a pressure-sensitive adhesive tape characterized in that a test piece made of the pressure-sensitive adhesive layer having dimensions of 0.5 mm thick × 6 mm wide × 20 mm long is prepared, and the test piece is elongated in the longitudinal direction at 25°C, at a chuck distance of 10 mm, and at a tensile speed of 50 mm/min, before the application of energy, and in a displacement-stress curve measured when the test piece is elongated, the breaking elongation W is 200% or more and 1000% or less, and the work required for the elongation of the test piece to reach 100% is 300 J/ ^m3 or more and 2000 J/ ^m3 or less.

(2) The adhesive tape according to (1) above, wherein the curable resin is at least one of urethane acrylate and bisphenol A-based epoxy acrylate.

(3) The adhesive tape according to (1) or (2) above, wherein the base resin is an acrylic resin.

(4) The pressure-sensitive adhesive tape according to (3), wherein the acrylic resin is a copolymer having an amide monomer as a structural unit.

(5) The adhesive tape according to any one of (1) to (4) above, wherein the adhesive layer further contains a crosslinking agent, and the crosslinking agent is an isocyanate-based crosslinking agent.

(6) The adhesive tape according to any one of (1) to (5) above, wherein the adhesive layer further contains a silicone material as an oil repellent.

(7) The adhesive tape according to any one of (1) to (6) above, wherein the substrate contains an antistatic agent.

(8) The adhesive tape according to any one of (1) to (7) above is used when a semiconductor wafer is fixed onto the adhesive layer as the substrate, the semiconductor wafer is cut partway through the thickness of the base material to separate the semiconductor wafer into individual pieces, and then the adhesive tape is stretched in the length direction, and the semiconductor chips are pushed up from the base material side and pulled out from the opposite side of the base material, thereby detaching the semiconductor chips from the adhesive layer.

(9) The adhesive tape according to any one of (1) to (8) above, wherein the adhesive layer has a thickness of 5 μm or more and 15 μm or less.

(10) The pressure-sensitive adhesive tape according to any one of (1) to (9) above, wherein the stress X at 100% elongation in the displacement-stress curve is 0.04 MPa or more and 0.4 MPa or less.

(11) The adhesive tape according to any one of (1) to (10) above, which satisfies the following requirement A:
Requirement A: The adhesive tape has a peel strength A of 30 cN/20 mm or more and 300 cN/20 mm or less, as measured when a 20 mm wide adhesive tape is attached to a #2000 polished silicon wafer with the adhesive layer facing the silicon wafer, the energy is applied to the adhesive layer, and then one end of the adhesive tape is held and peeled off at a speed of 1000 mm/min in a direction of 30° at 25°C.

(12) The pressure-sensitive adhesive tape according to any one of (1) to (11) above, which satisfies the following requirement B:
Requirement B: When a silicon substrate is fixed to the adhesive tape and then cut into individual pieces using a 30 μm thick blade in the thickness direction until it reaches half the thickness of the base material at a rotation speed of 30,000 rpm and a processing speed of 60 mm/s to obtain silicon chips measuring 4 mm long x 4 mm wide, the adhesion rate of the adhesive layer adhering to the side surface of the silicon chip due to the cutting must be 5.0% or less.

According to the present invention, the adhesive tape satisfies the following: a test piece made of the adhesive layer measuring 0.5 mm thick x 6 mm wide x 20 mm long is prepared, and the test piece is stretched longitudinally at 25°C with a chuck distance of 10 mm and a tensile speed of 50 mm/min before the application of energy to the adhesive layer; a fracture elongation W in a displacement-stress curve measured before the application of energy is 200% to 1000%; and the work required to reach 100% elongation of the test piece is 300 J/ ^m³ to 2000 J/ ^m³ . Therefore, when a semiconductor wafer as a substrate attached to the adhesive tape is cut in the thickness direction to obtain semiconductor chips as components before the application of energy to the adhesive layer of the adhesive tape, the adhesive tape can accurately suppress or prevent the occurrence of glue adhesion, in which a part of the adhesive layer of the adhesive tape adheres to the side surfaces of the obtained semiconductor chips.

1 is a longitudinal sectional view showing an example of a semiconductor device manufactured using the pressure-sensitive adhesive tape of the present invention. 2 is a longitudinal sectional view illustrating a method for manufacturing the semiconductor device shown in FIG. 1 using the adhesive tape of the present invention. FIG. 2 is a longitudinal sectional view illustrating a method for manufacturing the semiconductor device shown in FIG. 1 using the adhesive tape of the present invention. FIG. FIG. 3 is an enlarged cross-sectional view of the needle and its surroundings located in the area [A] surrounded by the dotted line in FIG. 2 . 1 is a longitudinal cross-sectional view showing an embodiment of a pressure-sensitive adhesive tape. 6A to 6C are longitudinal cross-sectional views for explaining a method for manufacturing the adhesive tape shown in FIGS. 5A to 5C.

The pressure-sensitive adhesive tape of the present invention will be described in detail below.
First, before describing the pressure-sensitive adhesive tape of the present invention, a semiconductor device manufactured using the pressure-sensitive adhesive tape of the present invention will be described.

<Semiconductor Device>
Fig. 1 is a longitudinal cross-sectional view showing an example of a semiconductor device manufactured using the pressure-sensitive adhesive tape of the present invention. In the following description, the upper side in Fig. 1 will be referred to as "top" and the lower side as "bottom." In addition, in each drawing referred to in this specification, dimensions in the left-right direction and/or thickness direction are exaggerated and differ significantly from the actual dimensions.

The semiconductor device 10 shown in Figure 1 includes a semiconductor chip (semiconductor element) 20, an interposer (substrate) 30 that supports the semiconductor chip 20, a plurality of conductive bumps (terminals) 70, and a molded portion (sealing portion) 17 that seals the semiconductor chip 20.

The interposer 30 is an insulating substrate made of various resin materials, such as polyimide, epoxy, cyanate, and bismaleimide triazine (BT resin). The planar shape of the interposer 30 is typically quadrangular, such as a square or rectangle.

Terminals 41 made of a conductive metal material such as copper are provided in a predetermined shape on the top surface (one side) of the interposer 30.

In addition, the interposer 30 has multiple vias (through holes) (not shown) formed through it in the thickness direction.

One end (upper end) of each bump 70 is electrically connected to a portion of the terminal 41 via a respective via, and the other end (lower end) protrudes from the lower surface (other surface) of the interposer 30.

The portion of the bump 70 that protrudes from the interposer 30 is approximately spherical (ball-shaped).

The bumps 70 are primarily made of a brazing material such as solder, silver solder, copper solder, or phosphorus copper solder.

In addition, terminals 41 are formed on the interposer 30. Terminals 21 of the semiconductor chip 20 are electrically connected to these terminals 41 via connection portions 81.

In this embodiment, as shown in FIG. 1, the terminals 21 are configured to protrude from the surface formed on the semiconductor chip 20, and the terminals 41 are also configured to protrude from the interposer 30.

The gap between the semiconductor chip 20 and the interposer 30 is filled with an underfill material made from various resin materials, and the hardened underfill material forms a sealing layer 80. This sealing layer 80 improves the bonding strength between the semiconductor chip 20 and the interposer 30 and prevents the intrusion of foreign matter, moisture, etc. into the gap.

Furthermore, on the upper side of the interposer 30, a molded portion 17 is formed to cover the semiconductor chip 20 and the interposer 30 and is made of a hardened semiconductor sealing material (sealant). This seals the semiconductor chip 20 within the semiconductor device 10 and prevents the intrusion of foreign matter, moisture, etc. into the semiconductor chip 20.

1, the semiconductor chip 20 (semiconductor element) has a semiconductor chip body 23 (semiconductor element body) and terminals 21 protruding from the underside of the semiconductor chip body 23. A circuit (not shown) is built into the upper surface of the semiconductor chip body 23, and the semiconductor chip body ₂₃ is mainly made of a semiconductor material such as Si, SiC, GaN, or _Ga2O3 .

The semiconductor device 10 and semiconductor chip 20 having such a configuration are manufactured, for example, by a semiconductor device manufacturing method using adhesive tape, as follows.

<Method of manufacturing semiconductor device>
Figures 2 and 3 are longitudinal cross-sectional views illustrating a method for manufacturing the semiconductor device shown in Figure 1 using the adhesive tape of the present invention, and Figure 4 is an enlarged cross-sectional view of the periphery of a needle located in the area [A] surrounded by a dotted line in Figure 2. In the following description, the upper side in Figures 2 to 4 will be referred to as "upper" and the lower side will be referred to as "lower." In addition, in each of the drawings referred to in this specification, the dimensions in the left-right direction and/or thickness direction are exaggerated and differ significantly from the actual dimensions.

[1A] First, prepare an adhesive tape 100 composed of a laminate having a base material 4 and an adhesive layer 2 laminated on the upper surface of the base material 4. As shown in Figure 2(a), place a semiconductor substrate 7 (semiconductor wafer) at the center 122 on the adhesive layer 2 and apply light pressure to laminate (attach) the semiconductor substrate 7 (attachment process).

This semiconductor substrate 7 has pre-formed on its top surface the circuits that will be included in the semiconductor chips 20 (semiconductor chip main body 23) that will be formed by singulation, and pre-formed on its bottom surface with terminals 21. The semiconductor substrate 7 is attached to the adhesive tape 100 with the top surface on which the circuits are formed facing the adhesive layer 2. Therefore, the top surface of the semiconductor substrate 7 on which the circuits are formed, i.e., the uneven surface on which the unevenness is formed, is bonded to the adhesive layer 2. Furthermore, the semiconductor substrate 7 typically has a diameter of approximately 6 to 12 inches and a thickness of approximately 100 to 600 μm.

In this main step [1A], the pressure-sensitive adhesive tape 100 of the present invention is used. That is, when the semiconductor substrate 7 as the substrate in this step [1A] is attached to the pressure-sensitive adhesive tape 100 with the uneven surface on which the circuit is formed facing the pressure-sensitive adhesive tape 100, the pressure-sensitive adhesive tape 100 used satisfies the following conditions: a test piece of the pressure-sensitive adhesive layer 2 having a thickness of 0.5 mm, a width of 6 mm, and a length of 20 mm is prepared before energy is applied to the pressure-sensitive adhesive layer 2, and the test piece is elongated in the longitudinal direction at 25°C under conditions of a chuck distance of 10 mm and a tensile speed of 50 mm/min, the fracture elongation W in the displacement-stress curve is measured to be 200% or more and 1000% or less, and the work required to reach 100% elongation of the test piece is 300 J/m3 or ^more and 2000 J/m3 ^or less.

As a result, in this step [1A], when the semiconductor substrate 7 serving as a substrate is attached to the adhesive tape 100 with the uneven surface on which the circuit is formed facing the adhesive tape 100, the semiconductor substrate 7 can be reliably attached to the adhesive tape 100 while accurately suppressing or preventing the adhesive layer 2 from unintentionally embedding itself in the unevenness of this uneven surface; a detailed explanation of this will be provided later.

[2A] Next, as shown in Figure 2(b), the adhesive tape 100 with the semiconductor substrate 7 laminated thereon is placed on the dicer table 200.

[3A] Next, the outer periphery 121 of the adhesive layer 2 is fixed with a wafer ring 9, and then a disc-shaped dicing saw (blade) (not shown) is used to cut (diced) the semiconductor substrate 7 as a substrate in the thickness direction to separate the semiconductor substrate 7, thereby obtaining semiconductor chips 20 as components on the adhesive tape 100 (singulation process; see Figure 2(c)).

At this time, the adhesive tape 100 has a cushioning effect, preventing cracks, chips, etc. when cutting the semiconductor substrate 7.

In addition, the cutting of the semiconductor substrate 7 using the blade is carried out so that it reaches partway through the thickness of the base material 4, as shown in Figure 2(c). This ensures that the semiconductor substrate 7 is reliably divided into individual pieces.

At this time, cutting water is supplied to the semiconductor substrate 7 while the semiconductor substrate 7 is being cut in order to prevent the scattering of dust generated when the semiconductor substrate 7, adhesive layer 2, and base material 4 are cut, and to prevent the semiconductor substrate 7 from being unnecessarily heated.

In this main step [3A], the pressure-sensitive adhesive tape 100 of the present invention is used. That is, when semiconductor chips 20 as components are obtained by cutting the semiconductor substrate 7 as a substrate in the thickness direction to separate it in this step [3A], the pressure-sensitive adhesive tape 100 used satisfies the following conditions: a test piece made of the pressure-sensitive adhesive layer having dimensions of 0.5 mm thick x 6 mm wide x 20 mm long is prepared before energy is applied to the pressure-sensitive adhesive layer 2; the test piece is elongated in the longitudinal direction at 25°C with a chuck distance of 10 mm and a tensile speed of 50 mm/min, and the fracture elongation W in a displacement-stress curve measured before the energy is applied is 200% or more and 1000% or less; and the work required to reach 100% elongation of the test piece is 300 J/ ^m3 or more and 2000 J/m3 ^or less.

As a result, in this step [3A], when the semiconductor substrate 7 is cut in the thickness direction to separate it into individual semiconductor chips 20, it is possible to effectively suppress or prevent the adhesion of glue from occurring on the side surfaces of the semiconductor chips 20 that are the resulting components, as part of the adhesive layer 2 provided on the adhesive tape 100; this will be explained in more detail later.

[4A] Next, the adhesive tape 100 with the semiconductor substrate 7 attached and fixed by the wafer ring 9 is transferred from the dicing device (not shown) to a pickup device (not shown). With the adhesive layer 2 fixed at the outer periphery 121 by the wafer ring 9, the center 310 is pushed upward against the outer periphery 320 of the table 300, thereby radially stretching the adhesive tape 100. This forms gaps with a fixed distance between the individual semiconductor substrates 7, i.e., the semiconductor chips 20 as components (expanding process; see Figure 2(d)).

Prior to the next step [5A], energy is applied to the adhesive layer 2 to reduce its adhesive strength to the semiconductor chip 20. This application of energy to the adhesive layer 2 may be performed after the expanding step in this step [4A], or prior to the expanding step.

[5A] Next, with a gap formed as a result of step [4A], the semiconductor chip 20 is picked up on the stage 400 by suction using a vacuum collet or air tweezers (pick-up step; see Figure 2(e)).

More specifically, the pickup of the semiconductor chip 20 is performed as follows. That is, first, in step [4A], the expanding step is performed, in which the adhesive tape 100 is radially expanded. Furthermore, after the expanding step in step [4A], in which the adhesive tape 100 is radially expanded, or prior to the expanding step, energy is applied to the adhesive layer 2 to harden the adhesive layer 2 and reduce the adhesive strength of the adhesive layer 2. Then, the needle 430 (not shown in FIG. 2) is changed from a state in which it is housed in the ejector head 410 as shown in FIG. 4(a) to a state in which it is protruded from the ejector head 410 as shown in FIG. 4(b). That is, the needle 430 is protruded in the thickness direction. As a result, the semiconductor chip 20 attached to the adhesive tape 100 is pushed up using the needle 430, causing it to be peeled off from the adhesive tape 100. Thereafter, the semiconductor chip 20 is picked up by suction using a vacuum collet or air tweezers, as shown in Figure 4(c).

In this main step [5A], the pressure-sensitive adhesive tape 100 of the present invention is used. That is, when the semiconductor chip 20 serving as a component is picked up from the pressure-sensitive adhesive tape 100 in this step [5A] by suction or the like using a vacuum collet or air tweezers, the pressure-sensitive adhesive tape 100 used satisfies the following conditions: a test piece of the pressure-sensitive adhesive layer 2 having a thickness of 0.5 mm, a width of 6 mm, and a length of 20 mm is prepared before energy is applied to the pressure-sensitive adhesive layer 2; the test piece is elongated in the longitudinal direction at 25°C, with a chuck distance of 10 mm, and a tensile speed of 50 mm/min, and the measured displacement-stress curve shows a breaking elongation W of 200% or more and 1000% or less; and the work required to reach 100% elongation of the test piece is 300 J/ ^m3 or more and 2000 J/m3 ^or less.

As a result, in this step [5A], when the semiconductor chip 20 serving as a component is picked up from the adhesive tape 100 by suction using a vacuum collet or air tweezers, this improved pick-up performance effectively reduces or prevents a portion of the adhesive layer 2 of the adhesive tape 100 from leaving adhesive residue on the back surface of the semiconductor chip 20 serving as the component; a detailed explanation of this will be provided later.

By going through the above steps [1A] to [5A], the semiconductor chips 20 are separated (diced) from the semiconductor substrate 7 using the adhesive tape 100. That is, with the semiconductor substrate 7 fixed on the adhesive layer 2 of the adhesive tape 100, the tape is cut from the semiconductor substrate 7 to reach partway through the thickness of the base material 4, thereby dicing the semiconductor substrate 7 into multiple semiconductor chips 20. Energy is then applied to the adhesive layer 2 to harden it, and with gaps formed between the semiconductor chips 20 at regular intervals, the semiconductor chips 20 are pushed up from the base material 4 side and pulled out from the opposite side of the base material 4, thereby separating the semiconductor chips 20 from the adhesive layer 2.

[6A] Next, the picked-up semiconductor chip 20 is transferred from the vacuum collet or air tweezers to a mounting probe or the like and turned upside down. Then, as shown in FIG. 3(a), the semiconductor chip 20 is placed on the interposer 30 with the terminals 21 of the semiconductor chip 20 facing the terminals 41 of the interposer 30 via the solder bumps 85 provided on the terminals 41. In other words, the semiconductor chip 20 (semiconductor element) is placed on the interposer 30 (substrate) with the surface of the semiconductor chip 20 on which the terminals 21 are formed facing downwards.

[7A] Next, as shown in FIG. 3(b), the interposer 30 and the semiconductor chip 20 are brought close together while the solder bumps 85 interposed between the terminals 21 and 41 are heated.

As a result, the molten solder bumps 85 come into contact with both terminals 21 and 41, and by cooling in this state, connection portions 81 are formed, resulting in an electrical connection between terminals 21 and 41 via connection portions 81 (mounting process; see Figure 3(c)).

[8A] Next, the gap formed between the semiconductor chip 20 and the interposer 30 is filled with an underfill material (sealing material) made of various resin materials, and then the underfill material is cured to form a sealing layer 80 made of the cured underfill material (sealing layer formation process; see Figure 3(d)).

[9A] Next, a molded portion 17 (sealing portion) is formed on the upper side of the interposer 30 so as to cover the semiconductor chip 20 and the interposer 30, thereby sealing the semiconductor chip 20 between the interposer 30 and the molded portion 17, and bumps 70 are formed to protrude from the underside of the interposer 30 and are electrically connected to part of the terminals 41 through vias provided in the interposer 30 (see Figure 3(e)).

Here, sealing with the molded portion 17 is performed by, for example, preparing a mold with an internal space corresponding to the shape of the molded portion 17 to be formed, and filling this internal space with powdered semiconductor encapsulation material so that it covers the semiconductor chip 20 and interposer 30 placed within the internal space. Then, in this state, the semiconductor encapsulation material is heated to harden it, resulting in a hardened semiconductor encapsulation material.

A semiconductor device 10 is obtained by the semiconductor device manufacturing method having the steps described above. More specifically, after performing steps [1A] to [9A], steps [4A] to [9A] are repeatedly performed, thereby manufacturing multiple semiconductor devices 10 in one batch from a single semiconductor substrate 7.

The following describes the adhesive tape 100 of the present invention, which is used in the manufacturing method of such a semiconductor device 10.

<Adhesive tape 100>
5 is a longitudinal cross-sectional view showing an embodiment of the adhesive tape. In the following description, the upper side in FIG. 5 will be referred to as "top" and the lower side as "bottom."

In the present invention, the adhesive tape 100 is configured as a laminate including a substrate 4 and an adhesive layer 2 laminated on the upper surface (one surface) of the substrate 4, and is used to temporarily fix a semiconductor substrate 7 (substrate) and a semiconductor chip 20 (component). The adhesive layer 2 of the adhesive tape 100 contains a base resin having adhesiveness and a curable resin that hardens when energy is applied, and when energy is applied to the adhesive layer 2, the adhesive layer 2 hardens, thereby reducing its adhesive strength. This pressure-sensitive adhesive tape 100 satisfies the following conditions: a test piece made of the pressure-sensitive adhesive layer 2 measuring 0.5 mm thick × 6 mm wide × 20 mm long is prepared, and the test piece is elongated in the longitudinal direction at 25°C under conditions of a chuck distance of 10 mm and a tensile speed of 50 mm/min before the application of energy to the pressure-sensitive adhesive layer 2; in the displacement-stress curve measured, the breaking elongation W is 200% or more and 1000% or less; and the work required to reach an elongation of 100% of the test piece is 300 J/ ^m3 or more and 2000 J/ ^m3 or less.

Here, in step [3A], when the semiconductor substrate 7 is diced (cut) in the thickness direction using a dicing saw to separate the semiconductor substrate 7 into multiple semiconductor chips 20, the cutting of the semiconductor substrate 7 is typically continued until it reaches halfway through the thickness direction of the base material 4 of the adhesive tape 100, as shown in Figure 2(c), in order to ensure that the semiconductor substrate 7 is reliably separated into individual chips.

At this time, in the adhesive tape 100, the adhesive layer 2 is located closer to the semiconductor substrate 7 than the base material 4 and is bonded to the semiconductor substrate 7, so naturally, when the semiconductor substrate 7 is cut, it is cut together with the semiconductor substrate 7 by the dicing saw.

The dicing saw (blade) used to cut the semiconductor substrate 7 has an overall disk shape with sharp diamond abrasive grains embedded in its tip (the end of the disk). While rotating this dicing saw in contact with the semiconductor substrate 7, it is moved in a grid pattern along the horizontal and vertical directions of the semiconductor substrate 7 when viewed from above. This causes the adhesive layer 2 to be torn apart along with the semiconductor substrate 7 by the dicing saw, cutting it in its thickness direction.

As described above, when the adhesive layer 2 is torn apart by the dicing saw, a portion of the cut adhesive layer 2 adheres to the side of the semiconductor chip 20, creating a problem of adhesive adhesion, as explained in the background art section above.

As a result of extensive research into this issue, the inventors have discovered that the occurrence of adhesive adhesion on the side surfaces of the semiconductor chip 20 is related to the breaking elongation W in the displacement-stress curve measured on the adhesive layer 2 before energy is applied, and the amount of work done when the test specimen is stretched.

Further investigation by the inventors has revealed that the breaking elongation W and the magnitude of the work load are related not only to the occurrence of adhesive adhesion on the side surfaces of the semiconductor chip 20 in step [3A], but also to the adherence of the semiconductor substrate 7 to the adhesive tape 100 in step [1A] and the pick-up of the semiconductor chip 20 from the adhesive tape 100 in step [5A].

Then, as a result of further investigations by the present inventors, it was found that the following three problems can be solved by preparing a test piece of the adhesive layer 2 having dimensions of 0.5 mm in thickness × 6 mm in width × 20 mm in length before applying energy to the adhesive layer 2, and stretching this test piece in the longitudinal direction under conditions of 25°C, a chuck distance of 10 mm, and a tensile speed of 50 mm/min, and measuring the fracture elongation W in the displacement-stress curve, and setting the work required to reach 100% elongation to 300 J/m3 ^{or more} and 2000 J/m3 ^{or less} , as described above, and thereby completing the present invention.

1. In step [3A], when semiconductor chips 20 are obtained by cutting semiconductor substrate 7 in the thickness direction to separate it, the occurrence of glue adhesion, in which a portion of adhesive layer 2 adheres to the side surface of semiconductor chip 20, can be accurately suppressed or prevented. This improves the reliability of semiconductor device 10 in which semiconductor chip 20 is sealed with molded portion 17. 2. In step [1A], when semiconductor substrate 7 is attached to adhesive tape 100 with the uneven surface on which the circuit is formed facing adhesive tape 100, the adhesive layer 2 can be accurately suppressed or prevented from unintentionally embedding into the unevenness of this uneven surface, while allowing semiconductor substrate 7 to be reliably attached to adhesive tape 100. 3. The inventors discovered that in step [5A], when the semiconductor chip 20 is picked up from the adhesive tape 100 by suction using a vacuum collet or air tweezers, it is possible to effectively suppress or prevent the adhesive residue of part of the adhesive layer 2 of the adhesive tape 100 from remaining on the back surface of the semiconductor chip 20, leading to the completion of the present invention.

The substrate 4 and adhesive layer 2 of this adhesive tape 100 (dicing tape) are described in detail below.

<Base material 4>
The substrate 4 is mainly made of a resin material, has a sheet shape, and has the function of supporting the adhesive layer 2 provided on this substrate 4. It also serves to realize the expansion in the expanding step of step [4A], in which the adhesive tape 100 is expanded in the planar direction. Furthermore, it serves to realize the pushing-up by the needles 430 in the pick-up step of step [5A], in which the individualized semiconductor chips 20 are picked up in a state in which they are pushed up by the needles 430.

Such resin materials are not particularly limited, but examples include thermoplastic resins such as polyolefin resins, polyvinyl chloride resins, polystyrene resins such as polystyrene thermoplastic elastomers, polyester resins (ester polymers) such as polyethylene terephthalate, polyethylene naphthalate, polybutylene terephthalate, polybutylene naphthalate, and polyester thermoplastic elastomers, polyurethane, polyimide, polyamide, polyether ketones such as polyether ether ketone, polyethersulfone, fluororesin, silicone resin, cellulose resin, acrylic resin, polyvinyl isoprene, and polycarbonate (carbonate polymer), as well as mixtures of these thermoplastic resins.

In particular, it is preferable to use polyolefin resin, polyvinyl chloride resin, polystyrene resin, or a mixture of these as the resin material. By using these resin materials, it is possible to reliably impart extensibility (expandability) to the substrate 4 in the expanding step [4A], and it is also possible to reliably achieve the needles 430 pushing up when the individual semiconductor chips 20 are picked up in a pushed-up state by the needles 430 in the step [5A]. Furthermore, it is possible to reliably suppress or prevent contamination of the adhesive tape 100 by cutting debris from the substrate 4 during dicing in the step [3A].

Such polyolefin resins are not particularly limited, but examples include polyethylene resins such as polypropylene, linear low-density polyethylene, low-density polyethylene, and very low-density polyethylene; polyethylene copolymers such as ethylene-vinyl acetate copolymer (EVA), ethylene-methyl methacrylate copolymer (EMMA), and ethylene-methacrylate copolymer (EMAA); and ionomers such as ethylene-based ionomers crosslinked with zinc ions, sodium ions, or potassium ions. One or more of these can be used in combination.

Polyvinyl chloride resins are polymers having a plurality of repeating units of the group represented by -CH ₂ -CHCl-, and specifically include homopolymers of vinyl chloride, copolymers of vinyl chloride with vinyl monomers (polymerizable monomers) copolymerizable with vinyl chloride, and post-chlorinated vinyl chloride polymers. Although one or a combination of two or more of these can be used, homopolymers are generally used.

Examples of copolymers of vinyl chloride with copolymerizable vinyl monomers include vinyl chloride-vinyl acetate copolymer, vinyl chloride-ethylene copolymer, and vinyl chloride-acrylic copolymer.

Furthermore, examples of polystyrene-based resins include, but are not limited to, polystyrene, poly(α-methylstyrene), polychlorostyrene, poly(m-propylstyrene), high impact polystyrene (HIPS), acrylonitrile-butadiene-styrene copolymer (ABS), acrylonitrile-styrene copolymer (AS), styrene-methacrylic acid copolymer, styrene-methacrylic acid alkyl ester copolymer, styrene-methacrylic acid glycidyl ester copolymer, styrene-acrylic acid copolymer, styrene-acrylic acid alkyl ester copolymer, styrene-maleic acid copolymer, and styrene-fumaric acid copolymer, as well as styrene-based thermoplastic elastomers such as styrene-butadiene copolymer and styrene-isoprene copolymer, and these may be used alone or in combination of two or more.

It is also preferable that the substrate 4 contains a conductive material that has electrical conductivity. By including such a conductive material, the conductive material can function as an antistatic agent, accurately suppressing or preventing the generation of static electricity in the semiconductor chip 20 during the singulation process [3A] and the pick-up process [5A].

In this way, when the base material 4 contains a conductive material, the surface resistivity of the surface of the base material 4 opposite to the adhesive layer 2 is preferably set to 1.0×10 ¹³ (Ω/□) or less, and more preferably set to 1.0×10 ¹¹ (Ω/□) or less, thereby more accurately suppressing or preventing the generation of static electricity in the semiconductor chip 20 in the singulation step [3A] and the pick-up step [5A].

This conductive material is not particularly limited as long as it is conductive, but examples include surfactants, permanently antistatic polymers (IDPs), metal materials, metal oxide materials, and carbon-based materials, and one or more of these can be used in combination.

Among these, examples of surfactants include anionic surfactants, cationic surfactants, nonionic surfactants, and amphoteric surfactants.

As permanently antistatic polymers (IDPs), all IDPs can be used, such as polyether and polyolefin block polymers, polyester amides, polyester amides, polyether ester amides, and polyurethanes.

Metal materials include gold, silver, copper or silver-coated copper, nickel, etc., and powders of these metals are preferably used.

Examples of metal oxide materials include indium tin oxide (ITO), indium oxide (IO), antimony tin oxide (ATO), indium zinc oxide (IZO), tin oxide (SnO ₂ ), and zinc oxide (ZnO), and powders of these metal oxides are preferably used.

Further examples of carbon-based materials include carbon black, carbon nanotubes such as single-walled carbon nanotubes and multi-walled carbon nanotubes, carbon nanofibers, CN nanotubes, CN nanofibers, BCN nanotubes, BCN nanofibers, graphene, etc.

Among these, the conductive material is preferably at least one of surfactants, permanently antistatic polymers (IDPs), metal oxide materials, and carbon black. These materials have low temperature dependence of resistivity, so even if the base material 4 is heated when dicing the semiconductor substrate 7 in step [3A], the amount of change in its surface resistivity can be minimized.

In addition, if static electricity generation in the semiconductor chip 20 is to be prevented without incorporating a conductive material into the base material 4, an antistatic layer containing a conductive material may be formed on the surface opposite the adhesive layer 2. This will achieve the same effect as if the base material 4 contained a conductive material.

Furthermore, the substrate 4 may contain softeners such as mineral oil, fillers such as calcium carbonate, silica, talc, mica, and clay, antioxidants, light stabilizers, lubricants, dispersants, neutralizers, colorants, etc.

Furthermore, when the base material 4 contains a constituent material other than the resin material as the main material, the content of the resin material in the base material 4 is preferably 50% by weight or more and 95% by weight or less, and more preferably 65% by weight or more and 90% by weight or less.

Furthermore, the thickness of the base material 4 is preferably, for example, 70 μm or more and 150 μm or less, and more preferably 80 μm or more and 120 μm or less. When the thickness of the base material 4 is within this range, the base material 4 can more reliably perform its function, and the dicing of the semiconductor substrate 7 in step [3A] can be carried out with excellent workability. Furthermore, in step [5A], the pick-up of the semiconductor chips 20, in which the individual semiconductor chips 20 are pushed up by the needles 430, can be carried out with greater precision.

Functional groups such as carboxyl groups, hydroxyl groups, and amino groups that are reactive with the constituent materials contained in the adhesive layer 2 may be exposed on the surface of the substrate 4.

The substrate 4 may also be composed of a laminate (multilayer body) made up of multiple layers composed of different resin materials.

<Adhesive layer 2>
The adhesive layer 2 adheres to and supports the semiconductor substrate 7 when the semiconductor substrate 7 is diced in the step [3A], and in the step [4A], energy is applied to the adhesive layer 2 to harden the adhesive layer 2, so that the semiconductor chips 20 obtained by dicing the semiconductor substrate 7 have enough adhesiveness to be able to be picked up in the step [5A].

Such adhesive layer 2 is composed of a resin composition containing, as its main materials, (1) a base resin having adhesive properties and (2) a curable resin that hardens the adhesive layer 2.

In the present invention, 1. in the step [3A], dicing of the semiconductor substrate 7 is performed with excellent accuracy while preventing glue adhesion to the side surfaces of the semiconductor chip 20, 2. in the step [1A], when the semiconductor substrate 7 is attached to the adhesive tape 100, preventing the adhesive layer 2 from unintentionally embedding into the unevenness of the uneven surface of the semiconductor substrate 7, and 3. in the step [5A], when the semiconductor chip 20 is picked up from the adhesive tape 100, for the purposes of preventing glue residue from occurring on the back surface of the semiconductor chip 20, the type and content, etc. of each component (constituent material) contained in the resin composition constituting the adhesive layer 2 are set so as to satisfy the following conditions before energy is applied to the adhesive layer 2: the fracture elongation W is 200% or more and 1000% or less, and the workload until the elongation of the test piece reaches 100% is 300 J/m ³ or more and 2000 J ^/ m 3 or less.

Each component contained in this resin composition will be described in detail below.
(1) Base Resin The base resin has adhesiveness and is contained in the resin composition in order to impart adhesiveness to the semiconductor substrate 7 to the adhesive layer 2 .

Such base resins include known adhesive layer components such as acrylic resins (adhesives), silicone resins (adhesives), polyester resins (adhesives), polyvinyl acetate resins (adhesives), polyvinyl ether resins (adhesives), styrene elastomer resins (adhesives), polyisoprene resins (adhesives), polyisobutylene resins (adhesives), and urethane resins (adhesives). Among these, acrylic resins are preferred. By using an acrylic resin as the base resin, it is relatively easy to set the breaking elongation W in the displacement-stress curve before energy is applied to the adhesive layer 2 and the work required to elongate the test specimen to 100% within the above ranges. Furthermore, acrylic resins have excellent heat resistance and are relatively easy and inexpensive to obtain, making them a preferred base resin.

Acrylic resins are polymers (homopolymers or copolymers) whose main monomer component is (meth)acrylic acid ester as the base polymer.

The (meth)acrylic acid ester is not particularly limited, but examples thereof include methyl (meth)acrylate, ethyl (meth)acrylate, propyl (meth)acrylate, isopropyl (meth)acrylate, butyl (meth)acrylate, isobutyl (meth)acrylate, s-butyl (meth)acrylate, t-butyl (meth)acrylate, pentyl (meth)acrylate, hexyl (meth)acrylate, heptyl (meth)acrylate, octyl (meth)acrylate, isooctyl (meth)acrylate, 2-ethylhexyl (meth)acrylate, nonyl (meth)acrylate, isononyl (meth)acrylate, and decyl (meth)acrylate. Examples of suitable acrylates include alkyl (meth)acrylates such as methyl (meth)acrylate, isodecyl (meth)acrylate, undecyl (meth)acrylate, dodecyl (meth)acrylate, tridecyl (meth)acrylate, tetradecyl (meth)acrylate, pentadecyl (meth)acrylate, hexadecyl (meth)acrylate, heptadecyl (meth)acrylate, and octadecyl (meth)acrylate; cycloalkyl (meth)acrylates such as cyclohexyl (meth)acrylate; and aryl (meth)acrylates such as phenyl (meth)acrylate. These acrylates may be used alone or in combination of two or more. Among these, alkyl (meth)acrylates such as methyl (meth)acrylate, ethyl (meth)acrylate, butyl (meth)acrylate, 2-ethylhexyl (meth)acrylate, and octyl (meth)acrylate are preferred. Alkyl (meth)acrylates are particularly excellent in heat resistance and are relatively easy and inexpensive to obtain. Furthermore, by including a (meth)acrylic acid alkyl ester as the base resin, it is relatively easy to set within the above range the magnitude of the breaking elongation W in the displacement-stress curve of the test specimen made of the adhesive layer 2 before energy is applied to the adhesive layer 2, and the magnitude of the work required to reach 100% elongation of the test specimen.

In this specification, the term "(meth)acrylic acid ester" is used to include both acrylic acid ester and methacrylic acid ester.

The acrylic resin used contains copolymerizable monomers as structural units as monomer components constituting the polymer, as needed, for the purposes of adjusting the magnitude of the breaking elongation W in the displacement-stress curve before energy is applied to the adhesive layer 2, and the amount of work required to reach 100% elongation of the test specimen, as well as modifying properties such as cohesive strength and heat resistance.

Such copolymerizable monomers are not particularly limited, but examples include hydroxyl group-containing monomers such as 2-hydroxyethyl (meth)acrylate, 2-hydroxypropyl (meth)acrylate, 4-hydroxybutyl (meth)acrylate, and 6-hydroxyhexyl (meth)acrylate; epoxy group-containing monomers such as glycidyl (meth)acrylate; carboxyl group-containing monomers such as (meth)acrylic acid, itaconic acid, maleic acid, fumaric acid, crotonic acid, and isocrotonic acid; and acid-free monomers such as maleic anhydride and itaconic anhydride. hydroxyl group-containing monomers, amide monomers such as (meth)acrylamide, N,N-dimethyl(meth)acrylamide, N-butyl(meth)acrylamide, N-methylol(meth)acrylamide, N-methylolpropane(meth)acrylamide, N-methoxymethyl(meth)acrylamide, and N-butoxymethyl(meth)acrylamide, amino group-containing monomers such as aminoethyl (meth)acrylate, N,N-dimethylaminoethyl (meth)acrylate, and t-butylaminoethyl (meth)acrylate, and (meth)acrylonitrile olefin-based monomers such as ethylene, propylene, isoprene, butadiene, and isobutylene; styrene-based monomers such as styrene, α-methylstyrene, and vinyltoluene; vinyl ester-based monomers such as vinyl acetate and vinyl propionate; vinyl ether-based monomers such as methyl vinyl ether and ethyl vinyl ether; halogen-atom-containing monomers such as vinyl chloride and vinylidene chloride; alkoxy-group-containing monomers such as methoxyethyl (meth)acrylate and ethoxyethyl (meth)acrylate; and monomers having a nitrogen atom-containing ring such as N-vinyl-2-pyrrolidone, N-methylvinylpyrrolidone, N-vinylpyridine, N-vinylpiperidone, N-vinylpyrimidine, N-vinylpiperazine, N-vinylpyrazine, N-vinylpyrrole, N-vinylimidazole, N-vinyloxazole, N-vinylmorpholine, N-vinylcaprolactam, and N-(meth)acryloylmorpholine. These monomers may be used alone or in combination of two or more. Among these, amide-based monomers are preferred. This makes it relatively easy to set the magnitude of the breaking elongation W in the displacement-stress curve of the test specimen made of adhesive layer 2 before energy is applied to adhesive layer 2, and the magnitude of the work required to reach 100% elongation of the test specimen, within the above range.

The content of these copolymerizable monomers is preferably 40% by weight or less, and more preferably 10% by weight or less, of the total monomer components constituting the acrylic resin.

The copolymerizable monomer may be contained at the end of the main chain of the polymer that constitutes the acrylic resin, or may be contained within the main chain, or may be contained both within and at the end of the main chain.

Furthermore, the copolymerizable monomer may contain a polyfunctional monomer for the purpose of crosslinking between polymers, etc.

Examples of polyfunctional monomers include 1,6-hexanediol (meth)acrylate, (poly)ethylene glycol di(meth)acrylate, (poly)propylene glycol di(meth)acrylate, neopentyl glycol di(meth)acrylate, pentaerythritol di(meth)acrylate, trimethylolpropane tri(meth)acrylate, pentaerythritol tri(meth)acrylate, dipentaerythritol hexa(meth)acrylate, glycerin di(meth)acrylate, epoxy (meth)acrylate, polyester (meth)acrylate, urethane (meth)acrylate, divinylbenzene, butyl di(meth)acrylate, and hexyl di(meth)acrylate, and these can be used alone or in combination of two or more.

Ethylene-vinyl acetate copolymers and vinyl acetate polymers can also be used as copolymerizable monomer components.

Such acrylic resins (polymers) can be produced by polymerizing a single monomer component or a mixture of two or more monomer components. Furthermore, polymerization of these monomer components can be carried out using polymerization methods such as solution polymerization, emulsion polymerization, bulk polymerization, and suspension polymerization.

The acrylic resin preferably contains a small amount of low-molecular-weight compounds to prevent contamination of the semiconductor substrate 7 and other components by the acrylic resin when dicing the semiconductor substrate 7 in step [3A]. In this case, the weight-average molecular weight of the acrylic resin is preferably set to between 300,000 and 5,000,000, more preferably between 400,000 and 4,000,000, and even more preferably between 500,000 and 1,500,000. Note that, depending on the type of monomer component, if the weight-average molecular weight of the acrylic resin is less than 500,000, its ability to prevent contamination of the semiconductor substrate 7 and other components may be reduced, resulting in the risk of adhesive residue remaining when the semiconductor chip 20 is peeled off.

The acrylic resin (base resin) used preferably has a glass transition point of -70°C or higher and -50°C or lower, more preferably -65°C or higher and -55°C or lower. By using an acrylic resin with a glass transition point within this range as the base resin, it is relatively easy to set the breaking elongation W in the displacement-stress curve of the test specimen made of adhesive layer 2 before energy is applied to adhesive layer 2, and the amount of work required to reach 100% elongation of the test specimen, within the above range.

The acrylic resin preferably has a functional group (reactive functional group) that is reactive with a crosslinking agent or photopolymerization initiator, such as a hydroxyl group or a carboxyl group (particularly a hydroxyl group). This allows the crosslinking agent or photopolymerization initiator to be linked to the acrylic resin, which is a polymer component, thereby effectively suppressing or preventing leakage of the crosslinking agent or photopolymerization initiator from the adhesive layer 2. As a result, the adhesion of the adhesive layer 2 to the semiconductor substrate 7 is reliably reduced during the energy ray irradiation in step [4A].

(2) Curable Resin The curable resin has a curing property such that it is cured by, for example, irradiation with energy rays. As a result of this curing, the base resin is incorporated into the cross-linked structure of the curable resin, resulting in a decrease in the adhesive strength of the adhesive layer 2.

Such curable resins are, for example, low-molecular-weight compounds that have at least two polymerizable carbon-carbon double bonds as functional groups in the molecule that can be three-dimensionally crosslinked by irradiation with energy rays such as ultraviolet rays or electron beams. Specific examples include trimethylolpropane tri(meth)acrylate, pentaerythritol tri(meth)acrylate, pentaerythritol tetra(meth)acrylate, tetramethylolmethane tetra(meth)acrylate, tetraethylene glycol di(meth)acrylate, 1,6-hexanediol di(meth)acrylate, neopentyl glycol di(meth)acrylate, dipentaerythritol hexa(meth)acrylate, and dipentaerythritol monohydroxypenta(meth)acrylate. acrylate, 1,4-butylene glycol di(meth)acrylate, polyethylene glycol di(meth)acrylate, esterification products of (meth)acrylic acid and polyhydric alcohol such as glycerin di(meth)acrylate, ester acrylate oligomers, cyanurate compounds having a carbon-carbon double bond-containing group such as 2-propenyl-di-3-butenyl cyanurate, tris(2-acryloxyethyl)isocyanurate, tris(2-methacryloxyethyl)isocyanurate, 2-hydroxyethyl Examples of suitable curable resins include isocyanurate compounds having a carbon-carbon double bond-containing group, such as bis(2-acryloxyethyl)isocyanurate, bis(2-acryloxyethyl)2-[(5-acryloxyhexyl)oxy]ethyl isocyanurate, tris(1,3-diacryloxy-2-propyl-oxycarbonylamino-n-hexyl)isocyanurate, tris(1-acryloxyethyl-3-methacryloxy-2-propyl-oxycarbonylamino-n-hexyl)isocyanurate, and tris(4-acryloxy-n-butyl)isocyanurate; commercially available oligoester acrylates; aromatic and aliphatic urethane acrylates; and bisphenol A-based epoxy acrylates. These may be used alone or in combination. Among these, it is preferable to use at least one of urethane acrylates and bisphenol A-based epoxy acrylates. This allows the curable resin to be more reliably cured by the application of energy, i.e., by irradiation with energy rays. Furthermore, based on the test specimen made of adhesive layer 2, the magnitude of the breaking elongation W in the displacement-stress curve before energy is applied to adhesive layer 2 and the magnitude of the work required to reach 100% elongation of the test specimen can be set within the above ranges relatively easily.

In addition, although there are no particular limitations on the curable resin, it is preferable to use a mixture of two or more curable resins with different weight-average molecular weights. Using such a curable resin makes it easy to control the degree of crosslinking of the resin due to energy ray irradiation. Furthermore, such a curable resin may be, for example, a mixture of a first curable resin and a second curable resin having a higher weight-average molecular weight than the first curable resin.

When the curable resin is a mixture of a first curable resin and a second curable resin, the weight-average molecular weight of the first curable resin is preferably approximately 100 to 1,000, and more preferably approximately 200 to 500. The weight-average molecular weight of the second curable resin is preferably approximately 1,000 to 30,000, more preferably approximately 1,000 to 10,000, and even more preferably approximately 2,000 to 5,000. Furthermore, the number of functional groups of the first curable resin is preferably 1 to 5, and the number of functional groups of the second curable resin is preferably 6 or more. By satisfying this relationship, the above-mentioned effects can be more significantly exhibited.

The curable resin is preferably blended in an amount of 30 to 200 parts by weight, and more preferably 50 to 140 parts by weight, per 100 parts by weight of the base resin. This ensures that the functions exhibited by the addition of the curable resin and base resin to the resin composition can be exerted reliably by both the curable resin and the base resin.

Note that when a double-bond-introduced acrylic resin is used as the aforementioned acrylic resin, i.e., when one having a carbon-carbon double bond in a side chain, in the main chain, or at the end of the main chain is used, the addition of this curable resin to the resin composition may be omitted. This is because, when the acrylic resin is a double-bond-introduced acrylic resin, the adhesive layer 2 hardens upon irradiation with energy rays due to the function of the carbon-carbon double bond contained in the double-bond-introduced acrylic resin, thereby reducing the adhesive strength of the adhesive layer 2.

(3) Photopolymerization initiator The adhesive layer 2 has its adhesiveness to the semiconductor substrate 7 reduced by irradiation with energy rays. When ultraviolet rays or the like are used as the energy rays, it is preferable that the resin composition constituting the adhesive layer 2 contains a photopolymerization initiator to facilitate the initiation of polymerization of the curable resin.

Examples of photopolymerization initiators include 2,2-dimethoxy-1,2-diphenylethan-1-one, 1-[4-(2-hydroxyethoxy)-phenyl]-2-hydroxy-2-methyl-1-propan-1-one, 2-hydroxy-1-{4-[4-(2-hydroxy-2-methyl-propionyl)-benzyl]phenyl}-2-methyl-propan-1-one, benzyl diphenyl sulfide, tetramethylthiuram monosulfide, 4-(2-hydroxyethoxy)phenyl (2-hydroxy-2-propyl) ketone, α-hydroxy-α,α'-dimethylacetophenone, 2-methyl-2-hydroxypropiophenone, 1-hydroxycyclohexyl phenyl ketone, and Michler's ketone. acetophenone, methoxyacetophenone, 2,2-dimethoxy-2-phenylacetophenone, 2,2-diethoxyacetophenone, 2-methyl-1-[4-(methylthio)-phenyl]-2-morpholinopropane-1, benzoin methyl ether, benzoin ethyl ether, benzoin propyl ether, benzoin isopropyl ether, benzoin isobutyl ether, benzyl, benzoin, dibenzyl, α-hydroxycyclohexyl phenyl ketone, benzil dimethyl ketal, 2-hydroxymethylphenylpropane, 2-naphthalenesulfonyl chloride, 1-phenone-1,1-propanedione-2-(o-ethoxycarbonyl)oxime, benzophenone, benzyl benzoylbenzoic acid, 4,4'-dimethylaminobenzophenone, 4,4'-diethylaminobenzophenone, 4,4'-dichlorobenzophenone, 3,3'-dimethyl-4-methoxybenzophenone, o-acryloxybenzophenone, p-acryloxybenzophenone, o-methacryloxybenzophenone, p-methacryloxybenzophenone, p-(meth)acryloxyethoxybenzophenone, benzophenone-4-carboxylic acid esters of acrylates such as 1,4-butanediol mono(meth)acrylate, 1,2-ethanediol mono(meth)acrylate, and 1,8-octanediol mono(meth)acrylate, thioxanthone, and 2-chlorothioxanthone, Examples include 2-methylthioxanthone, 2,4-dimethylthioxanthone, isopropylthioxanthone, 2,4-dichlorothioxanthone, 2,4-diethylthioxanthone, 2,4-diisopropylthioxanthone, azobisisobutyronitrile, β-chloroanthraquinone, camphorquinone, halogenated ketones, acylphosphinoxides, acylphosphonates, polyvinylbenzophenone, chlorothioxanthone, dodecylthioxanthone, dimethylthioxanthone, diethylthioxanthone, 2-ethylanthraquinone, t-butylanthraquinone, and 2,4,5-triarylimidazole dimer, and these can be used alone or in combination of two or more.

The photopolymerization initiator is preferably blended in an amount of 0.1 to 50 parts by weight, and more preferably 0.5 to 10 parts by weight, per 100 parts by weight of the base resin. By adjusting the amount of photopolymerization initiator blended as described above, the photopolymerization initiator can be made to reliably exhibit the functions that are exhibited by adding the photopolymerization initiator to the resin composition.

(4) Crosslinking Agent Furthermore, the resin composition constituting the adhesive layer 2 may contain a crosslinking agent. By containing a crosslinking agent, the adhesive layer 2 can be adjusted to have an appropriate hardness. Therefore, based on the test specimen constituted by the adhesive layer 2, the magnitude of the breaking elongation W in the displacement-stress curve before energy is applied to the adhesive layer 2 and the magnitude of the work required to reach 100% elongation of the test specimen can be set within the above ranges relatively easily.

The crosslinking agent is not particularly limited, but examples include isocyanate-based crosslinking agents, epoxy-based crosslinking agents, urea resin-based crosslinking agents, methylol-based crosslinking agents, chelate-based crosslinking agents, aziridine-based crosslinking agents, melamine-based crosslinking agents, polyvalent metal chelate-based crosslinking agents, acid anhydride-based crosslinking agents, polyamine-based crosslinking agents, and carboxyl group-containing polymer-based crosslinking agents. Of these, isocyanate-based crosslinking agents are preferred. Isocyanate-based crosslinking agents can more significantly enhance the effects achieved by using a crosslinking agent.

Isocyanate-based crosslinking agents are not particularly limited, but examples include polyisocyanate compounds of polyvalent isocyanates, trimers of polyisocyanate compounds, trimers of isocyanate-terminated compounds obtained by reacting a polyisocyanate compound with a polyol compound, and blocked polyisocyanate compounds in which isocyanate-terminated urethane prepolymers are blocked with phenols, oximes, etc.

Examples of polyisocyanates include 2,4-tolylene diisocyanate, 2,6-tolylene diisocyanate, 1,3-xylylene diisocyanate, 1,4-xylylene diisocyanate, diphenylmethane-4,4'-diisocyanate, diphenylmethane-2,4'-diisocyanate, 3-methyldiphenylmethane diisocyanate, hexamethylene diisocyanate, isophorone diisocyanate, dicyclohexylmethane-4,4'-diisocyanate, dicyclohexylmethane-2,4'-diisocyanate, 4,4'-diphenylether diisocyanate, 4,4'-[2,2-bis(4-phenoxyphenyl)propane]diisocyanate, and 2,2,4-trimethyl-hexamethylene diisocyanate, and these may be used alone or in combination of two or more. Of these, at least one polyisocyanate selected from the group consisting of 2,4-tolylene diisocyanate, diphenylmethane-4,4'-diisocyanate, and hexamethylene diisocyanate is preferred.

The crosslinking agent is preferably blended in an amount of 0.01 to 30 parts by weight, and more preferably 0.1 to 20 parts by weight, per 100 parts by weight of the base resin. By adjusting the amount of crosslinking agent blended as described above, the crosslinking agent can be made to reliably exert its functions, which are exerted by adding the crosslinking agent to the resin composition.

(5) Plasticizer A plasticizer improves the flexibility of the adhesive layer 2, the adhesive strength of which decreases when energy is applied. As a result, the magnitude of the breaking elongation W in the displacement-stress curve before energy is applied to the adhesive layer 2 based on the test specimen made of the adhesive layer 2 and the magnitude of the work required to reach 100% elongation can be set within the above ranges relatively easily. Therefore, it is preferable that a plasticizer is contained in the adhesive layer 2, i.e., the resin composition.

The plasticizer is not particularly limited, but examples include phthalate ester plasticizers such as DOP (dioctyl phthalate), DBP (dibutyl phthalate), DIBP (diisobutyl phthalate), and DHP (diheptyl phthalate); aliphatic dibasic acid ester plasticizers such as DOA (di-2-ethylhexyl adipate), DIDA (diisodecyl adipate), and DOS (di-2-ethylhexyl sebacate); aromatic carboxylic acid ester plasticizers such as ethylene glycol benzoates; trimellitate ester plasticizers such as TOTM (trioctyl trimellitate); and adipate ester plasticizers, and these may be used alone or in combination of two or more.

The content of plasticizer in adhesive layer 2, i.e., the resin composition, is not particularly limited, but is preferably 0.1 to 5.0 parts by weight, and more preferably 0.5 to 3.0 parts by weight, per 100 parts by weight of base resin. This reliably improves the flexibility of adhesive layer 2. Therefore, based on the test specimen made of adhesive layer 2, the magnitude of the breaking elongation W in the displacement-stress curve before energy is applied to adhesive layer 2 and the magnitude of the work required to reach 100% elongation of the test specimen can be more easily set within the above ranges.

(6) Silicone Material Contrary to the plasticizer described above, a silicone material makes the adhesive layer 2 hard in the adhesive layer 2, the adhesive strength of which decreases when energy is applied. As a result, the magnitude of the breaking elongation W in the displacement-stress curve before energy is applied to the adhesive layer 2 based on the test specimen made of the adhesive layer 2 and the magnitude of the work required to make the elongation of the test specimen reach 100% can be set within the above ranges relatively easily. Therefore, it is preferable that a silicone material is contained in the adhesive layer 2, i.e., in the resin composition.

A silicone material is a compound composed of repeating structural units having a siloxane bond (-Si-O-Si-). Specific examples of these repeating structural units include those composed of repeating structural units having a siloxane bond of at least one of the following formulas (3) and (4):

(In formula (3), _X1 represents a hydrocarbon group or a hydroxyl group.)

(In formula (4), X ₂ and X ₃ each independently represent a hydrocarbon group or a hydroxyl group.)

Specific examples of repeating units having siloxane bonds include compounds having a polyorganosiloxane skeleton and compounds having a silsesquioxane skeleton. The silsesquioxane structure may be any structure, such as a random structure, a cage structure, or a ladder structure.

Examples of hydrocarbon groups include alkyl groups such as methyl, ethyl, propyl, and isopropyl; cycloalkyl groups such as cyclopropyl, cyclobutyl, cyclopentyl, and cyclohexyl; aryl groups such as phenyl, naphthyl, and 2-methylphenyl; aralkyl groups such as benzyl, diphenylmethyl, and naphthylmethyl; phenyl and biphenyl groups.

Specific examples of silicone materials having such a structure include, but are not limited to, compounds represented by the following formulas (5) and (6):

(In formula (5), Me represents a methyl group, and n represents an integer of 1 or more.)

(In formula (6), Me represents a methyl group, p represents an integer of 1 or more, and R1, R2, R3, and R4 each independently represent a hydrocarbon group, an organic group, or a hydrogen atom.)

The content of silicone material (polydimethylsiloxane) in adhesive layer 2, i.e., the resin composition, is not particularly limited, but is preferably 0.1 to 2.0 parts by weight, and more preferably 0.2 to 1.0 part by weight, per 100 parts by weight of base resin. This ensures that the hardness of adhesive layer 2 can be adjusted to be high, and therefore the amount of work required to reach 100% elongation in the displacement-stress curve of the test specimen made of adhesive layer 2 before energy is applied to adhesive layer 2 can be more easily set within the above range.

(7) Other Components Furthermore, the resin composition constituting the adhesive layer 2 may contain, in addition to the above-mentioned components (1) to (6), at least one of other components selected from the group consisting of a conductive material, a tackifier, an antioxidant, a tackifier, a filler, a colorant, a flame retardant, a softener, an antioxidant, a surfactant, and the like.

The conductive material among these is not particularly limited as long as it is conductive, but materials similar to those described as the conductive material contained in the substrate 4 can be used.

By including such a conductive material, the conductive material functions as an antistatic agent, effectively suppressing or preventing the generation of static electricity in the semiconductor chip 20 during the singulation process [3A] and the pick-up process [5A].

When a conductive material is contained in either the base material 4 or the adhesive layer 2, it is preferable to contain the conductive material in the base material 4. This makes it possible to more accurately suppress or prevent the generation of static electricity on the semiconductor chip 20 without having to reliably adhere the conductive material to the semiconductor chip 20.

The tackifiers among these include, but are not limited to, rosin resins, terpene resins, coumarone resins, phenolic resins, aliphatic petroleum resins, aromatic petroleum resins, and aliphatic-aromatic copolymer petroleum resins, and one or more of these can be used in combination.

As described above, by appropriately selecting the type and content of each of the components (1) to (7) contained in the adhesive layer 2, of which components (1) and (2) are essential components, the magnitude of the breaking elongation W in the displacement-stress curve before energy is applied to the adhesive layer 2 based on the test specimen composed of the adhesive layer 2, and the magnitude of the work required to reach 100% elongation of the test specimen, can be set within the above ranges. Therefore, the adhesive layer 2 can: 1. be capable of adhering to and supporting the semiconductor substrate 7 when dicing the semiconductor substrate 7 in step [3A], and can accurately suppress or prevent the occurrence of adhesive adhesion, in which a portion of the adhesive layer 2 provided on the adhesive tape 100 adheres to the side surface of the semiconductor chip 20 as the obtained component. Furthermore, the adhesive layer 2 can: 2. In step [1A], when the semiconductor substrate 7 is attached to the adhesive tape 100 with the uneven surface on which the circuit is formed facing the adhesive tape 100, the adhesive layer 2 can be reliably prevented from unintentionally embedding into the unevenness of the uneven surface, while the semiconductor substrate 7 can be reliably attached to the adhesive tape 100. Furthermore, 3. After applying energy to the adhesive layer 2 to harden it in step [5A], when the semiconductor chip 20 is picked up from the adhesive tape 100 by suction using a vacuum collet or air tweezers, the adhesive layer 2 can be reliably prevented from leaving a residue of adhesive on the back surface of the semiconductor chip 20, thereby improving the reliability of the semiconductor device 10 in which the semiconductor chip 20 is sealed in the molded portion 17.

As described above, the adhesive layer 2 having the above configuration is prepared by preparing a test piece of the adhesive layer 2 having dimensions of 0.5 mm thick × 6 mm wide × 20 mm long before applying energy to the adhesive layer 2, and stretching this test piece in the longitudinal direction under conditions of 25°C, a chuck distance of 10 mm, and a tensile speed of 50 mm/ ^min . In the displacement-stress curve measured when this test piece is stretched, it is sufficient that the breaking elongation W is 200% or more and 1000% or less, and the work required to reach 100% elongation of the test piece is 300 J/m3 or more and 2000 J/ ^m3 or less. However, the breaking elongation W is preferably 200% or more and 800% or less, more preferably 220% or more and 700% or less, and even more preferably 250% or more and 600% or less. Furthermore, the work load required to reach 100% elongation of the test piece is preferably 350 J/m3 ^{or more} and 1900 J/m3 ^or less, more preferably 450 J/m3 or ^more and 1800 J/m3 or ^less , and even more preferably 550 J/ ^m3 or more and 750 J/ ^m3 or less. By setting the magnitude of the breaking elongation W in the displacement-stress curve before energy is applied to the adhesive layer 2 and the magnitude of the work load required to reach 100% elongation within the above ranges, the effects 1 to 3 described above can be more significantly exhibited.

In addition, when the displacement-stress curve is plotted with displacement (elongation) on the X-axis and stress on the Y-axis, the amount of work required to reach 100% elongation of the test specimen can be calculated as the area of the region enclosed by the plotted displacement-stress curve and the X-axis.

Furthermore, the stress X at 100% elongation in the displacement-stress curve before energy is applied to the adhesive layer 2 is preferably 0.08 MPa or more and 0.35 MPa or less, and more preferably 0.10 MPa or more and 0.30 MPa or less. This allows the adhesive layer 2 to be said to have an appropriate hardness. Therefore, when the semiconductor substrate 7 is diced in step [3A], the adhesive layer 2, which is cut along with the semiconductor substrate 7, can be scraped off as fine powder. This can more accurately suppress or prevent the occurrence of glue adhesion, in which part of the adhesive layer 2 provided on the adhesive tape 100 adheres to the side surface of the semiconductor chip 20 as the obtained component.

As described above, by setting the magnitude of the breaking elongation W in the displacement-stress curve before energy is applied to the adhesive layer 2 and the magnitude of the work required to reach 100% elongation of the test specimen within the above ranges, it is possible to suppress the occurrence of adhesive adhesion, whereby the adhesive layer 2 adheres to the side surfaces of the semiconductor chip 20. The degree of adhesive adhesion is specifically set as follows: A silicon substrate is fixed to the adhesive tape 100, and then the silicon substrate is cut in the thickness direction to half the thickness of the substrate 4 using a 30 μm thick blade at a rotation speed of 30,000 rpm and a processing speed of 60 mm/s to obtain silicon chips measuring 4 mm long x 4 mm wide. The adhesion rate of the adhesive layer 2 adhering to the side surfaces of the silicon chips due to the cutting is preferably set to 5.0% or less, and more preferably 2.0% or less. This effectively suppresses the occurrence of adhesive adhesion on the side surfaces of the semiconductor chips 20 obtained by singulating the semiconductor substrate 7.

Furthermore, by setting the magnitude of the breaking elongation W in the displacement-stress curve before energy is applied to the adhesive layer 2 and the magnitude of the work required to reach 100% elongation of the test piece within the above ranges, in step [1A], when the semiconductor substrate 7 is attached to the adhesive tape 100 with the uneven surface on which the circuit is formed facing the adhesive tape 100, it is possible to prevent the adhesive layer 2 from unintentionally embedding itself into the unevenness of the uneven surface, and the degree of embedding into the unevenness is specifically set as shown below. Specifically, a transparent glass substrate is prepared having a plurality of square pillar-shaped protrusions (convex stripes) on its surface, each 200 μm high, 400 μm wide, and spaced 1.0 mm apart. Adhesive tape 100 is applied to the surface of the transparent glass substrate by pressing a 35 mm diameter, 400 mm wide roller against the surface at a pressure of 0.5 MPa while the stage is heated to 45°C. The conformance of adhesive tape 100, calculated by dividing the depth of the recesses formed by adhesive layer 2 conforming to the protrusions by the height of the protrusions, is preferably 30% or less, and more preferably 10% or less. This prevents adhesive layer 2 from unintentionally embedding itself into the irregularities of the textured surface.

Furthermore, by setting the magnitude of the breaking elongation W in the displacement-stress curve before energy is applied to the adhesive layer 2 and the magnitude of the work load until the elongation of the test piece reaches 100% within the above ranges, when the semiconductor chip 20 is picked up from the adhesive tape 100 by suction using a vacuum collet or air tweezers in step [5A], it is possible to accurately suppress or prevent the occurrence of adhesive residue, in which a portion of the adhesive layer 2 of the adhesive tape 100 remains on the back surface of the semiconductor chip 20. Specifically, the peel strength (peel strength A) of the adhesive tape 100 when this pick-up is performed is set as shown below. That is, after applying a 20 mm wide adhesive tape to a #2000 polished silicon wafer with the adhesive layer facing the silicon wafer, the energy is applied to the adhesive layer, and then the adhesive tape is held at one end and peeled at a 30° angle at 25°C at a rate of 1000 mm/min. The peel strength A measured is preferably 30 cN/20 mm or more and 300 cN/20 mm or less, more preferably 40 cN/20 mm or more and 200 cN/20 mm or less, and even more preferably 50 cN/20 mm or more and 110 cN/20 mm or less. Setting the peel strength A within this range can be said to be an appropriate strength for picking up the semiconductor chip 20, and can accurately suppress or prevent adhesive residue, where part of the adhesive layer 2 of the adhesive tape 100 remains on the back surface of the semiconductor chip 20.

The thickness of the adhesive layer 2 is not particularly limited, but is preferably 5 μm or more and 15 μm or less, and more preferably 5 μm or more and 10 μm or less. By setting the thickness of the adhesive layer 2 within this range, the adhesive layer 2 can exhibit good adhesion to the semiconductor substrate 7 in the singulation step [3A] and function as an adhesive layer with sufficient adhesion to exhibit good peelability between the adhesive layer 2 and the semiconductor substrate 7 in the pickup step [5A]. This can also more significantly enhance the effects 1 to 3 described above, which are obtained by setting the magnitude of the breaking elongation W in the displacement-stress curve before energy is applied to the adhesive layer 2 and the magnitude of the work required to reach 100% elongation of the test specimen within the above range.

The adhesive layer 2 may also be composed of a laminate (multilayer body) made up of multiple layers composed of different resin compositions.

Furthermore, in the adhesive tape 100 having a configuration in which the adhesive layer 2 is laminated on the substrate 4 as described above, when the adhesive tape 100 is viewed in a plan view, the number of bubbles having an area of 100 ^μm2 or more formed at the interface between the substrate 4 and the adhesive layer 2 is preferably 15.0 bubbles/mm2 or ^less , more preferably 0.01 bubbles/mm2 or ^more and 7.0 bubbles/ ^mm2 or ^less , and even more preferably 0.1 bubbles/ ^mm2 or more and 2.0 bubbles/ ^mm2 or less. By controlling the number of bubbles formed at the interface between the substrate 4 and the adhesive layer 2 and setting the number of bubbles having an area of 100 μm2 or more as described above, it is possible to more accurately suppress or prevent adhesive residue on the semiconductor chip 20 when the semiconductor chip 20 is picked up and the adhesive tape 100 is peeled off from the semiconductor chip 20 in the step [5A].

Next, the adhesive tape 100 having such a configuration can be manufactured, for example, as follows.

<Method of manufacturing adhesive tape>
Fig. 6 is a vertical cross-sectional view for explaining a method for producing the adhesive tape shown in Fig. 5. In the following explanation, the upper side in Fig. 6 will be referred to as "top" and the lower side as "bottom".

[1B] First, a substrate 4 is prepared (see FIG. 6(a)).
The method for producing the substrate 4 is not particularly limited, and examples thereof include common molding methods such as extrusion molding methods such as a calendar method, an inflation extrusion method, and a T-die extrusion method, and a wet casting method. When the substrate 4 is formed as a laminate, molding methods such as a co-extrusion method and a dry lamination method are used as the method for producing the substrate 4 having such a configuration.

The substrate 4 can be used without stretching, or, if necessary, can be subjected to uniaxial or biaxial stretching.

[2B] Next, an adhesive layer 2 is formed on the upper surface of the substrate 4 (see FIG. 6(b)).
The surface (upper surface) of the substrate 4 may be subjected to a surface treatment such as corona treatment, chromic acid treatment, matte treatment, ozone exposure treatment, flame exposure treatment, high-voltage shock exposure treatment, ionizing radiation treatment, primer treatment, or anchor coat treatment in order to improve the adhesion between the substrate 4 and the adhesive layer 2.

The adhesive layer 2 can also be obtained by coating or spraying a liquid material, which is a varnish made by dissolving the resin composition that constitutes the adhesive layer 2 in a solvent, onto the substrate 4, and then volatilizing the solvent.

The solvent is not particularly limited, but examples include methyl ethyl ketone, acetone, toluene, ethyl acetate, dimethyl formaldehyde, etc., and one or more of these can be used in combination.

In addition, the liquid material can be applied or sprayed onto the substrate 4 using methods such as die coating, curtain die coating, gravure coating, comma coating, bar coating, and lip coating.

[3B] Next, a portion of the adhesive layer 2 formed on the substrate 4 is removed in a circular pattern, leaving the substrate 4 in the thickness direction of the adhesive layer 2 so that the central and peripheral sides are separated, thereby forming the adhesive layer 2 with a central portion 122 and a peripheral portion 121 (see Figure 6(c)).

One method for removing a circular portion of the adhesive layer 2 is to punch out a portion of the adhesive layer 2 surrounding the area to be removed, and then remove the adhesive layer 2 located in the punched-out area.

The area to be removed can be punched out using, for example, a method using a roll-shaped mold or a method using a press mold. Of these, the method using a roll-shaped mold, which allows for continuous production of the adhesive tape 100, is preferred.

In this process, a portion of the adhesive layer 2 is punched out into a ring shape (circular shape) to form the central portion 122 and the outer periphery 121. However, the shape of the punched portion of the adhesive layer 2 may be any shape as long as it allows the outer periphery 121 of the adhesive layer 2 to be fixed with a wafer ring in the semiconductor device manufacturing method described above. Specific examples of punched shapes include the circular shape described above, as well as oval shapes such as ellipses and bale-shaped ovals, and polygonal shapes such as squares and pentagons.

[4B] Next, a separator 1 is laminated onto the adhesive layer 2 formed on the substrate 4 to obtain an adhesive tape 100 in which the adhesive layer 2 is coated with the separator 1 (see Figure 6(d)).

The method for laminating the separator 1 onto the adhesive layer 2 is not particularly limited, but for example, a lamination method using a roll or a lamination method using a press can be used. Of these, a lamination method using a roll is preferred from the standpoint of productivity, which allows for continuous production.

The separator 1 is not particularly limited, but examples include polypropylene film, polyethylene film, polyethylene terephthalate film, etc.

The separator 1 may have its surface treated with a release agent so that it can be peeled off when the adhesive tape 100 is used. Examples of release treatments include coating the surface of the separator 1 with a release agent, or creating fine irregularities on the surface of the separator 1. Examples of release agents include silicone-based, alkyd-based, and fluorine-based agents.

Through the above steps, an adhesive tape 100 coated with a separator 1 can be formed.

The adhesive tape 100 coated with the separator 1 manufactured in this embodiment is used after peeling the adhesive tape 100 from the separator 1 in the semiconductor device manufacturing method using the adhesive tape 100 described above.

When peeling the separator 1 from the adhesive layer 2 that it covers, it is preferable to peel the separator 1 at an angle of 90° or more and 180° or less relative to the surface of the adhesive layer 2. By setting the angle at which the separator 1 is peeled within this range, peeling can be reliably prevented at locations other than the interface between the adhesive layer 2 and the separator 1.

The above describes the adhesive tape of the present invention, but the present invention is not limited to this.

For example, any component capable of exerting the same function may be added to each layer of the adhesive tape of the present invention, or the substrate may be composed of a single layer as described in the above embodiment, or may be composed of multiple layers. For example, the substrate may be provided with an antistatic layer on the surface opposite the adhesive layer.

Furthermore, the structure of each layer of the adhesive tape can be replaced with any other layer that can perform the same function, or any other layer can be added.

Furthermore, depending on the configuration of the semiconductor device formed using adhesive tape, it may be possible to omit the formation of the molded portion 17 provided on the semiconductor device 10.

The adhesive tape of the present invention can be applied not only to cases where a semiconductor substrate with adhesive tape attached is cut in the thickness direction (diced) to obtain cut pieces, i.e., semiconductor chips as components, but also to various substrate processing applications in which components must be peeled off from the adhesive tape after the substrate is temporarily fixed on the adhesive tape and cut in the thickness direction to obtain components. In addition to the semiconductor substrates (semiconductor wafers) described above, examples of substrates that can be attached with the adhesive tape of the present invention include glass substrates such as soda-lime glass, borosilicate glass, and quartz glass; ceramic substrates such as alumina, silicon nitride, and titanium oxide; resin material substrates such as acrylic, polycarbonate, and rubber; and metal material substrates.

Next, specific examples of the present invention will be described.
However, the present invention is not limited to the descriptions in these examples.

1. Preparation of Raw Materials First, the raw materials used in the production of the pressure-sensitive adhesive tapes of each Example and Comparative Example are shown below.

(Polyolefin Resin 1)
As polyolefin resin 1, polypropylene (PP, manufactured by Sumitomo Chemical Co., Ltd., "FS2011DG3") was prepared.

(Polyolefin resin 2)
As the polyolefin resin 2, low-density polyethylene (LDPE, manufactured by Sumitomo Chemical Co., Ltd., "Sumikathene F200-0", specific gravity: 0.92 g/cm ³ ) was prepared.

(Elastomer)
As the elastomer, a styrene-based elastomer (manufactured by Asahi Kasei Corporation, "Tuftec H1221") was prepared.

(Antistatic Agent 1)
As antistatic agent 1, a polyether-based antistatic agent (manufactured by Sanyo Chemical Industries, Ltd., "Pelectron PVL") was prepared.

(Base resins 1 to 4)
Base resins 1 to 4 were prepared by mixing at least two of butyl acrylate, butyl methacrylate, 2-ethylhexyl acrylate, acrylic acid, 2-hydroxyethyl acrylate, N,N-dimethylacrylamide, and vinyl acetate, and subjecting the mixture to solution polymerization in a toluene solvent by a conventional method.

The glass transition temperatures and weight-average molecular weights of base resins (acrylic copolymers) 1 to 4 were as shown below. Of base resins 1 to 4, base resin 1 is a copolymer having an amide monomer (N,N-dimethylacrylamide) as a structural unit.

Base resin 1 (glass transition temperature: -14°C, weight average molecular weight: 500,000)
Base resin 2 (glass transition temperature: -9°C, weight average molecular weight: 500,000)
Base resin 3 (glass transition temperature: -45°C, weight average molecular weight: 650,000)
Base resin 4 (glass transition temperature: -37°C, weight average molecular weight: 600,000)

(Curable resin 1)
As the curable resin 1, dipentaerythritol hexaacrylate (manufactured by Daicel Allnex Corporation, product number: DPHA), which is an esterification product of (meth)acrylic acid and polyhydric alcohol, was prepared.

(Curable resin 2)
As the curable resin 2, urethane acrylate (manufactured by Miwon Specialty Chemical Co., Ltd., product number: SC2152) was prepared.

(Curable resin 3)
As the curable resin 3, urethane acrylate (product number UX-5000, manufactured by Nippon Kayaku Co., Ltd.) was prepared.

(Curable resin 4)
As the curable resin 4, bis-A type epoxy acrylate (manufactured by Nippon Kayaku Co., Ltd., product number: R-130) was prepared.

(Crosslinking agent 1)
As a crosslinking agent 1, polyisocyanate (manufactured by Tosoh Corporation, product number: Coronate L) was prepared.

(Photopolymerization initiator 1)
As a photopolymerization initiator 1, benzyl dimethyl ketal (manufactured by Tokyo Chemical Industry Co., Ltd.) was prepared.

(Plasticizer 1)
As plasticizer 1, a polyester plasticizer (manufactured by DIC Corporation, product number: W-230H) was prepared.

(Oil repellent 1)
As oil repellent 1, silicone oil (manufactured by Dow Toray Industries, Inc., "SF8419") was prepared.

2. Preparation of adhesive tape [Example 1]
A resin composition containing polyolefin resin 1 (45% by weight), elastomer (40% by weight), and antistatic agent 1 (15% by weight) was extruded using an extruder to prepare a substrate 4 having a thickness of 80 μm.

Next, a liquid material containing a resin composition containing base resin 1 (100 parts by weight), curable resin 2 (30 parts by weight), crosslinker 1 (0.5 parts by weight), oil repellent 1 (0.6 parts by weight), and photopolymerization initiator 1 (7 parts by weight) was prepared. This liquid material was bar-coated onto substrate 4 so that the adhesive layer 2 would be 10 μm thick after drying, and then dried at 80°C for 1 minute to form an adhesive layer 2 with a thickness of 10 μm on the upper surface (one side) of substrate 4, thereby obtaining adhesive tape 100 of Example 1.

[Examples 2 to 6, Comparative Examples 1 to 4]
The adhesive tapes of Examples 2 to 6 and Comparative Examples 1 to 4 were produced in the same manner as in Example 1, except that the constituent materials contained in the resin composition used to form the substrate 4 and the constituent materials contained in the resin composition used to form the adhesive layer 2 were those shown in Table 1, and the contents of the constituent materials were changed as shown in Table 1 to form the substrate 4 and adhesive layer 2 with the thicknesses shown in Table 1.

3. Evaluation The pressure-sensitive adhesive tapes obtained in each of the Examples and Comparative Examples were evaluated by the following methods.

3-1. Tensile Test of Adhesive Layer For each of the pressure-sensitive adhesive tapes 100 of the Examples and Comparative Examples, test pieces of the adhesive layer 2 measuring 0.8 mm thick x 6 mm wide x 20 mm long were prepared. Before applying energy to the test pieces, a tensile test was performed on the test pieces using a precision universal testing machine (Shimadzu Corporation, "Autograph AGS-X") at 25°C with a chuck distance of 10 mm and a tensile speed of 50 mm/min to obtain a displacement-stress curve (S-S curve). Then, based on the obtained displacement-stress curve, the work load [J/m ³ ] required to reach 100% elongation, the breaking elongation W [%], and the stress X [MPa] at 100% elongation were calculated.

3-2. Measurement of peel strength from silicon wafer Each adhesive tape having a width of 20 mm in each example and each comparative example was attached to the surface of a silicon wafer whose surface had been polished with #2000 polishing, and then, after applying energy by irradiating the adhesive layer 2 with ultraviolet light under conditions of ultraviolet illuminance: 55 W/cm ² and ultraviolet exposure dose: 200 mJ/cm ² , one end of the adhesive tape was held in an environment of 25°C and peeled at a speed of 1000 mm/min in a direction of 30°, and the peel strength A (peel strength A) measured when peeled was measured using a peel analyzer (manufactured by Kyowa Interface Science Co., Ltd., "VPA-H100").

3-3. Evaluation of Adhesive Residue on the Side Surface of a Silicon Chip <1> A silicon wafer (manufactured by SUMCO, 625 μm thick) made of silicon was prepared as a silicon substrate. Then, the adhesive tape 100 of each example and comparative example was fixed to the surface of the silicon substrate, with the adhesive layer 2 facing the silicon substrate side. Then, using a 30 μm thick blade, the silicon substrate was cut in the thickness direction until it reached half the thickness of the base material 4 under conditions of a rotation speed of 30,000 rpm and a processing speed of 60 mm/s, thereby obtaining multiple silicon chips measuring 6 mm long x 6 mm wide.

<2> Next, ultraviolet light was irradiated onto the adhesive layer 2 from the substrate 4 side under conditions of ultraviolet illuminance: 55 W/ ^cm2 and ultraviolet exposure dose: 200 mJ/ ^cm2 , and then the silicon chip was pushed up using a needle with a needle push-up amount of 600 μm, and then the silicon chip was picked up by suction with a vacuum collet.

By going through steps <1> to <3> above, 50 silicon chips were picked up by suction for each adhesive tape of each example and comparative example.

Then, for each silicon chip obtained using the adhesive tape of each Example and Comparative Example, the adhesion rate, which indicates the area where the adhesive layer adhered to the side of the picked-up silicon chip, was measured. This adhesion rate measurement was performed on 50 silicon chips for each adhesive tape of each Example and Comparative Example, and the average value was calculated, and then these were evaluated according to the following criteria.

(Evaluation of adhesion rate of adhesive layer)
The average adhesion rate of the adhesive layer on the side of 50 silicon chips is
◎: 0.0% (no adhesion) or more and 2.0% or less ○: More than 2.0% and 5.0% or less △: More than 5.0% and 10.0% or less ×: More than 10.0%

A transparent glass substrate having a plurality of square pillar-shaped protrusions (convex stripes) having a height of 200 μm, a width of 400 μm, and a pitch of 1.0 mm on its surface was prepared, and the pressure-sensitive adhesive tape 100 of each of the Examples and Comparative Examples was attached to the surface of this transparent glass substrate by pressing a roller having a diameter of 35 mm and a width of 400 mm at a pressure of 0.5 MPa with the stage heated to 45°C.

Then, the depth of the recesses formed in the adhesive tape 100 by the adhesive layer 2 following the protrusions was divided by the height of the protrusions to calculate the conformance rate of the adhesive tape 100. The conformance of the protrusions to the adhesive tape 100 was evaluated using this conformance rate based on the following evaluation criteria.

(Evaluation of the follow-up rate of the adhesive layer)
The conformability of the adhesive tape 100 to the protrusion is ⊚: 10% or less.
◯: More than 10% but not more than 30% △: More than 30% but not more than 50% ×: More than 50%

3-5. Evaluation of adhesive residue on the back surface of a silicon chip <1> A silicon wafer (manufactured by SUMCO, thickness 625 μm) made of silicon was prepared as a silicon substrate. Then, the adhesive tape of each example and comparative example was fixed to the silicon substrate with the adhesive layer facing the silicon substrate. Then, using a 100 μm thick blade, while supplying cutting water to the silicon substrate, the silicon substrate was cut in the thickness direction to reach half the thickness of the base material 4 under conditions of a rotation speed of 30,000 rpm and a processing speed of 60 mm/s, thereby obtaining multiple silicon chips measuring 6 mm long x 6 mm wide.

<2> Next, the silicon chip was pushed up using a needle with a needle push-up distance of 600 μm, and then the silicon chip was picked up by suction with a vacuum collet.

By going through steps <1> and <2> above, 50 silicon chips were picked up by suction for each adhesive tape of each example and comparative example.

Then, for each silicon chip obtained using the adhesive tape of each Example and Comparative Example, the presence or absence of contamination due to the adhesion of cutting water on the surface (back surface) of the picked-up silicon chip was confirmed. This confirmation of the presence or absence of contamination was carried out on 50 silicon chips for each adhesive tape of each Example and Comparative Example, and they were evaluated according to the following criteria.

(Evaluation of adhesion rate of adhesive layer)
Of the 50 silicon chips, the number of those found to have surface contamination was:
◎: 0 pieces ○: More than 0 pieces and 5 pieces or less △: More than 5 pieces and 10 pieces or less ×: More than 10 pieces

3-6. Evaluation of Pickup Properties A silicon wafer (manufactured by SUMCO Corporation) made of silicon was prepared and ground in a conventional manner to obtain a silicon wafer with a thickness of 230 μm. After grinding to a thickness of 200 μm with a #2000 wheel, the adhesive tape 100 of each example and each comparative example was fixed to the ground surface with the adhesive layer 2 facing the silicon wafer side. Thereafter, using a dicing blade (manufactured by DISCO Corporation, product number: Z05-SD2000-N1-50, thickness: 30 μm), the silicon wafer was cut in the thickness direction until it reached the middle of the substrate 4 under conditions of rotation speed: 40,000 rpm, processing speed: 60 mm/s, and thus a plurality of silicon chips measuring 6 mm long x 6 mm wide were obtained.
Next, ultraviolet light was irradiated onto the adhesive layer 2 from the substrate 4 side under conditions of ultraviolet illuminance: 55 W/cm ² and ultraviolet exposure dose: 200 mJ/cm ² , and then the surface of the diced silicon chip was adsorbed using a vacuum adsorption collet, and four needles spaced 4 mm apart were raised 200 μm from below the substrate 4 to pick up the silicon chip from the adhesive tape.

The silicon chip pick-up properties of the adhesive tapes of each Example and Comparative Example were checked for the silicon chips picked up in the manner described above, and evaluated according to the following criteria.

(Pickup property evaluation)
◎: 50 out of 50 were successfully picked. 〇: Between 48 and 50 out of 50 were successfully picked.
△: 40 or more but less than 48 out of 50 were picked up.
×: Less than 40 out of 50 were picked up.

As shown in Table 1, the adhesive tape 100 of each Example satisfied the following conditions: the breaking elongation W in the displacement-stress curve before energy application was 200% or more and 1000% or less, and the work required to reach 100% elongation of the test piece was 300 J/m3 or ^more and 2000 J/ ^{m3 or} less, and as a result, it was possible to reliably suppress or prevent a part of the adhesive layer of the adhesive tape from adhering to the side or back surface of the obtained silicon chip.

In contrast, the adhesive tapes of each comparative example did not satisfy at least one of the following conditions: the breaking elongation W in the displacement-stress curve before energy application was 200% or more and 1000% or less; and the workload required for the elongation of the test piece to reach 100% was 300 J/ ^m3 or more and 2000 J/ ^m3 or less. As a result, part of the adhesive layer of the adhesive tape adhered to the side or back surface of the obtained silicon chip.

REFERENCE SIGNS LIST 1 separator 2 adhesive layer 4 substrate 7 semiconductor substrate 9 wafer ring 10 semiconductor device 17 molded portion 20 semiconductor chip 21 terminal 23 semiconductor chip body 30 interposer 41 terminal 70 bump 80 sealing layer 81 connection portion 85 solder bump 100 adhesive tape 121 outer periphery 122 center 200 dicer table 300 table 310 center 320 outer periphery 400 stage 410 ejector head 430 needle

Claims (12)

An adhesive tape used for temporarily fixing a semiconductor wafer as a substrate and a component, the adhesive tape comprising: a base material; and an adhesive layer laminated on one surface of the base material; the adhesive tape is used for temporarily fixing a semiconductor wafer as a substrate and a component, the semiconductor wafer being fixed in place; the adhesive tape is then diced into individual pieces to form semiconductor chips as a plurality of components; and the semiconductor chips are then detached from the adhesive layer;
the adhesive layer contains a base resin having adhesiveness and a curable resin that is cured by the application of energy, and the adhesive strength of the adhesive layer is reduced by applying energy to the adhesive layer to cure the adhesive layer;
a pressure-sensitive adhesive tape characterized in that a test piece made of the pressure-sensitive adhesive layer having dimensions of 0.5 mm thick × 6 mm wide × 20 mm long is prepared, and the test piece is elongated in the longitudinal direction at 25°C, at a chuck distance of 10 mm, and at a tensile speed of 50 mm/min, before the application of energy, and in a displacement-stress curve measured when the test piece is elongated, the breaking elongation W is 200% or more and 1000% or less, and the work required for the elongation of the test piece to reach 100% is 300 J/ ^m3 or more and 2000 J/ ^m3 or less. The adhesive tape according to claim 1, wherein the curable resin is at least one of urethane acrylate and bisphenol A-based epoxy acrylate. The adhesive tape according to claim 1 or 2, wherein the base resin is an acrylic resin. The adhesive tape according to claim 3, wherein the acrylic resin is a copolymer having amide monomers as structural units. The adhesive tape according to any one of claims 1 to 4, wherein the adhesive layer further contains a crosslinking agent, and the crosslinking agent is an isocyanate-based crosslinking agent. The adhesive tape according to any one of claims 1 to 5, wherein the adhesive layer further contains a silicone material as an oil repellent. The adhesive tape according to any one of claims 1 to 6, wherein the substrate contains an antistatic agent. The adhesive tape according to any one of claims 1 to 7 is used when a semiconductor wafer is fixed onto the adhesive layer as the substrate, the semiconductor wafer is cut partway through the thickness of the base material to separate the semiconductor wafer into individual semiconductor chips, and the adhesive tape is then stretched lengthwise while the semiconductor chips are pushed up from the base material side and pulled out from the opposite side of the base material, thereby detaching the semiconductor chips from the adhesive layer. The adhesive tape according to any one of claims 1 to 8, wherein the adhesive layer has a thickness of 5 μm or more and 15 μm or less. The adhesive tape according to any one of claims 1 to 9, wherein the stress X at 100% elongation in the displacement-stress curve is 0.04 MPa or more and 0.4 MPa or less. The adhesive tape according to any one of claims 1 to 10, which satisfies the following requirement A:
Requirement A: The adhesive tape has a peel strength A of 30 cN/20 mm or more and 300 cN/20 mm or less, as measured when a 20 mm wide adhesive tape is attached to a #2000 polished silicon wafer with the adhesive layer facing the silicon wafer, the energy is applied to the adhesive layer, and then one end of the adhesive tape is held and peeled off at a speed of 1000 mm/min in a direction of 30° at 25°C. The adhesive tape according to claim 1 , which satisfies the following requirement B:
Requirement B: When a silicon substrate is fixed to the adhesive tape and then cut into individual pieces using a 30 μm thick blade in the thickness direction until it reaches half the thickness of the base material at a rotation speed of 30,000 rpm and a processing speed of 60 mm/s to obtain silicon chips measuring 4 mm long x 4 mm wide, the adhesion rate of the adhesive layer adhering to the side surface of the silicon chip due to the cutting must be 5.0% or less.