TW201829698A - Adhesive sheet for semiconductor processing - Google Patents

Abstract

The present invention relates to an adhesive sheet for semiconductor processing, the adhesive sheet being provided with: a base material; and an adhesive layer which is disposed on the base material and is formed of an adhesive composition. The adhesive composition includes: a polymer (A) that contains a reactive functional group (A1); a polymer (B) that contains a reactive functional group (B1) which is different from the functional group (A1), and that contains an actinic ray-polymerizable group (B2); a crosslinking agent (C) that reacts with the reactive functional group (A1); and a crosslinking agent (D) that reacts with the reactive functional group (B1).

Description

   Adhesive sheet for semiconductor processing   

The present invention relates to an adhesive sheet for semiconductor processing, and more particularly to an adhesive sheet for semiconductor wafer surface protection used to protect the surface of a semiconductor wafer with bumps.

Information terminal equipment is rapidly becoming thinner, smaller, and more versatile. Semiconductor devices mounted on these devices are also required to be thinner and more dense, and semiconductor wafers are also expected to be thinner. Conventionally, in order to meet this requirement, the back surface of a semiconductor wafer has been ground and thinned. Furthermore, in recent years, semiconductor wafers have formed bumps made of solder or the like on the surface of the wafer with a height of several tens to hundreds of μm. When such a semiconductor wafer with bumps is ground, the surface protection sheet is attached to the surface of the wafer forming the bumps in order to protect the bumps.

As a surface protection sheet, conventionally, as disclosed in Patent Document 1, an adhesive sheet in which an intermediate layer and an adhesive layer are sequentially provided on a substrate is known. In Patent Document 1, a conventional adhesive such as an acrylic adhesive, a silicone adhesive, or a rubber adhesive is used as the adhesive layer. In addition, it has been shown that a cross-linking agent can be blended in the adhesive to introduce a cross-linked structure.

Furthermore, Patent Document 1 discloses that an energy-ray-curable oligomer is formulated in an adhesive, or a carbon-carbon double bond is introduced into a polymer constituting the adhesive to make the adhesive energy-hardenable. The surface protection sheet uses an energy ray-curable adhesive, and the energy of the adhesive layer is reduced by irradiation of the energy ray. Therefore, it is easy to peel off the semiconductor wafer after use.

In addition, an ultra-high-strength gel having a dual network structure has been conventionally known. As shown in Non-Patent Document 1, an ultra-high-strength gel system is obtained by polymerizing poly (2-acrylamidine-2-methylpropanesulfonic acid) to obtain a gel (PAMPS gel), and immersing the PAMPS gel in acrylic resin An amine monomer solution was obtained by polymerizing acrylamide in a PAMPS gel.

    [Prior technical literature]         [Patent Literature]    

Patent Document 1: Japanese Patent No. 4367769

Non-Patent Document 1: Invasive Drugs of Ultra-High-Strength Double Network Gels and Their High-Strength Mechanism, Proceedings of Polymers, Vol.65, No.12, (2008) pp.707-715

In recent years, with the higher density and miniaturization of semiconductor devices, the bump height has tended to increase. However, for semiconductor wafers with increased bump heights, adhesive residues (paste residues) tend to occur on the bumps when the surface protection sheet is peeled off. Although the paste residue tends to decrease due to the energy ray hardening of the adhesive, in recent years, it has been required to further reduce the contamination of semiconductor wafers. Only by making the energy ray hardenability, the paste residue cannot be reduced to the desired level. degree.

Furthermore, in Non-Patent Document 1, no attempt has been made to apply an ultra-high-strength gel having a dual network structure to an adhesive or to change it to energy ray-hardenability.

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide an adhesive sheet for semiconductor processing that is less likely to cause paste residue on the surface of a workpiece such as a semiconductor wafer during peeling.

As a result of active review by the present inventors, it was found that by blending two polymers in the adhesive composition constituting the adhesive layer, the individual polymers were crosslinked by different cross-links, and one polymer was made into energy ray hardening. In order to solve the above problems, the following inventions have been completed. That is, this invention provides the adhesive sheet for semiconductor processing of the following (1)-(10).

(1) An adhesive sheet for semiconductor processing, comprising: a substrate; and an adhesive sheet for semiconductor processing comprising a substrate and an adhesive layer formed on one side of the substrate and formed with an adhesive composition. The aforementioned adhesive composition contains a polymer (A) having a reactive functional group (A1), a reactive functional group (B1) different from the reactive functional group (A1), and an energy ray polymerizable group (B2). Polymer (B), cross-linking agent (C) that reacts with reactive functional group (A1), and cross-linking agent (D) that reacts with reactive functional group (B1).

(2) The adhesive sheet for semiconductor processing according to the above (1), wherein the weight average molecular weight of the polymer (A) is higher than the weight average molecular weight of the polymer (B).

(3) The adhesive sheet for semiconductor processing as described in (1) or (2) above, wherein the polymer (A) and the polymer (B) are both acrylic polymers.

(4) The adhesive sheet for semiconductor processing according to the above (3), wherein the weight average molecular weight of the acrylic polymer constituting the polymer (A) is higher than the weight average molecular weight of the acrylic polymer constituting the polymer (B), and there is a difference More than 200,000.

(5) The adhesive sheet for semiconductor processing according to the above (4), wherein the weight average molecular weight of the acrylic polymer constituting the polymer (A) is 300,000 to 1,000,000, and the weight of the acrylic polymer constituting the polymer (B) The average molecular weight is 5,000 ~ 100,000.

(6) The adhesive sheet for semiconductor processing according to any one of (3) to (5) above, wherein the acrylic polymer constituting the polymer (A) contains a functional group derived from a functional group having a reactive functional group (A1) An acrylic copolymer (A ') of a constituent unit of the body (a1) and a constituent unit derived from an alkyl (meth) acrylate (a2).

(7) The adhesive sheet for semiconductor processing according to any one of (3) to (6) above, wherein the acrylic polymer constituting the polymer (B) is an energy-containing polymer having an energy ray polymerizable group (B2) The linear polymerizable compound (S) and acrylic acid derived from the structural unit containing the functional group monomer (b1) having a reactive functional group (B1) and acrylic acid derived from the structural unit of the alkyl (meth) acrylate (b2) A reactant obtained by partially reacting a reactive functional group (B1) of the polymer (B ').

(8) The adhesive sheet for semiconductor processing according to any one of claims 1 to 7, wherein the reactive functional group (A1) is a carboxyl group and the reactive functional group (B1) is a hydroxyl group.

(9) The adhesive sheet for semiconductor processing according to the above (8), wherein the crosslinking agent (C) is an epoxy-based crosslinking agent, and the crosslinking agent (D) is an isocyanate-based crosslinking agent.

(10) The adhesive sheet for semiconductor processing according to any one of (1) to (9) above, wherein the content of the crosslinking agent (D) in the adhesive composition is more than that of the crosslinking agent on a mass basis. The content of (C) is 2 to 20 parts by mass relative to 100 parts by mass of the polymer (B).

The present invention can provide an adhesive sheet for semiconductor processing, in which paste residues are unlikely to occur on the surface of a workpiece during peeling.

FIG. 1 is a stress-strain curve made by performing a cyclic tensile test on the adhesive layer after the energy ray hardening in Example 1. FIG.

FIG. 2 is a stress-strain curve prepared by performing a cyclic tensile test on the adhesive layer before the energy ray hardening in Example 1. FIG.

FIG. 3 is a stress-strain curve obtained by performing a cyclic tensile test on the adhesive layer after the energy ray hardening of Comparative Example 1. FIG.

In the following description, "weight average molecular weight (Mw)" is a polystyrene-equivalent value measured by a gel permeation chromatography (GPC) method, and specifically a value measured based on the method described in the examples.

In addition, in the description in this specification, the term "(meth) acrylate" is used in terms of both "acrylate" and "methacrylate", and the same applies to other similar terms.

Hereinafter, the present invention will be described in more detail using embodiments.

The adhesive sheet for semiconductor processing of the present invention (hereinafter also simply referred to as "adhesive sheet") includes a base material and an adhesive layer provided on the other surface of the base material. The adhesive sheet may have an intermediate layer between the substrate and the adhesive layer. The adhesive sheet may be composed of two or three layers as described above, and other layers may be further provided. For example, a release material may be further provided on the adhesive layer.

Hereinafter, each member constituting the adhesive sheet will be described in detail.

<Substrate>

The substrate used for the adhesive sheet is not particularly limited, but a resin film is preferred. Resin films are more suitable for processing components of electronic parts because they generate less dust than paper or non-woven fabrics. The substrate may be a single-layer film made of a single resin film, or a multilayer film formed by laminating a plurality of resin films.

Examples of the resin film used as the substrate include polyolefin-based films, halogenated ethylene polymer-based films, acrylic resin-based films, rubber-based films, cellulose-based films, polyester-based films, polycarbonate-based films, and polystyrene. Based films, polyphenylene sulfide based films, cycloolefin polymer based films, and the like.

Among these, based on the viewpoint that the wafer can be stably held when the wafer is ground to a very thin thickness, and the viewpoint of a film having high thickness accuracy, a polyester film is preferred. Among the polyester films, it is easy to obtain and has a thickness. From the viewpoint of high accuracy, a polyethylene terephthalate film is more preferable.

The thickness of the substrate is not particularly limited, but is preferably 10 to 200 μm, more preferably 25 to 150 μm, and still more preferably 25 to 100 μm.

In addition, from the viewpoint of improving the adhesion of the substrate to the adhesive layer or the intermediate layer, it can also be used as a substrate on the surface of a resin film and further provided with an easy-adhesion layer. Furthermore, the base material used in the present invention may contain a filler, a colorant, an antistatic agent, an antioxidant, an organic lubricant, a catalyst, and the like, as long as the effect of the present invention is not impaired. In addition, the base material may be transparent or may be colored as desired, but it is preferably an energy-ray-transmitting person capable of hardening the adhesive layer to a sufficient degree.

<Adhesive layer>

When the adhesive layer is provided on the base material or an intermediate layer is provided, it is the one provided on the intermediate layer. The adhesive layer is formed of an adhesive composition. The adhesive composition contains a polymer (A) and a polymer (B), and a crosslinking agent (C) and a crosslinking agent (D). Hereinafter, each component will be described in detail.

[Polymer (A), (B)]

The polymer (A) has a reactive functional group (A1). The polymer (B) has a reactive functional group (B1) and an energy ray polymerizable group (B2) having a functional group different from the reactive functional group (A1). The reactive functional group (A1) of the polymer (A) does not cross-link with the cross-linking agent (D), but preferentially cross-links with the cross-linking agent (C). In addition, the reactive functional group (B1) of the polymer (B) does not crosslink with the crosslinking agent (C), but preferentially crosslinks with the crosslinking agent (D).

The reactive functional group (A1) and the reactive functional group (B1) are not particularly limited, and may be selected from the group consisting of a hydroxyl group, a carboxyl group, an amino group, an epoxy group, and the like, and among these, a carboxyl group and a hydroxyl group are preferred.

Furthermore, it is more preferable that any one of the reactive functional group (A1) and the reactive functional group (B1) is a carboxyl group, and the other is a hydroxyl group. Thereby, the polymer (A) and the cross-linking agent (C) react, and the polymer (B) and the cross-linking agent (D) react to easily form an individual mesh structure.

Here, the reactive functional group (A1) may be a hydroxyl group, and the reactive functional group (B1) may be a carboxyl group, but more preferably the reactive functional group (A1) is a carboxyl group, and the reactive functional group (B1) is a hydroxyl group. When the reactive functional group (B1) of the polymer (B) is a hydroxyl group, the polymer (B) easily reacts with an energy ray polymerizable group-containing compound (S) described later.

Since the polymer (A) is a compound having no energy ray polymerizable group, it is a non-energy ray-curable compound that does not harden even when irradiated with energy rays. On the other hand, since the polymer (B) has an energy ray polymerizable group (B2), it is an energy ray-curable compound that hardens by irradiating energy rays. When the adhesive layer formed of the adhesive composition is irradiated with energy rays, the polymer (B) hardens and the adhesive force becomes low.

In addition, the energy line means a person having an energy quantum in an electromagnetic wave or a charged particle beam, and as an example thereof, ultraviolet rays, electron beams, and the like are exemplified. Among these, it is preferable to harden the adhesive layer using ultraviolet rays.

In addition, examples of the energy ray polymerizable group (B2) include those having a carbon-carbon double bond such as (meth) acrylfluorenyl, vinyl, and allyl. Among these, (meth) acrylfluorenyl is preferred. .

When the polymer (A) has a reactive functional group (A1) and the polymer (B) has a reactive functional group (B1) different from the reactive functional group (A1), the crosslinking agent (C) ) The polymer (A) is crosslinked, and the polymer (B) is crosslinked with a crosslinking agent (D) different from the crosslinking agent (C). Therefore, in the adhesive layer, a three-dimensional mesh structure (hereinafter also referred to as a "first mesh") composed of the polymer (A) and the crosslinking agent (C) is formed, and the polymer (B) and the crosslinking agent are formed. A three-dimensional mesh structure (hereinafter also referred to as a "second mesh") constituted by the agent (D). In addition, since the polymer (B) contains the energy ray polymerizable group (B2) in the second mesh, it is estimated that by irradiating the energy ray, the mesh becomes denser and has a hard and brittle structure. On the other hand, it is estimated that the first mesh is composed of the polymer (A) and the cross-linking agent (C), and has a structure that is softer and easier to stretch than the second mesh.

After the adhesive layer is irradiated with energy rays, a rigid second mesh is assembled into the soft first mesh as described above to form a so-called double network. Therefore, the breaking properties, elongation at break, and breaking energy of the adhesive layer after energy ray irradiation tend to be good. When the adhesive sheet is peeled off from a workpiece such as a semiconductor wafer, paste residues are unlikely to occur on the workpiece.

The polymer (A) and the polymer (B) are adhesive components (adhesive resins) that make the adhesive layer exhibit adhesiveness, and are selected from, for example, acrylic polymers, urethane polymers, rubber polymers, and Polyolefin. The polymer (A) and the polymer (B) are preferably an acrylic polymer and a urethane polymer selected from these, and more preferably an acrylic polymer.

As the polymer (A) and the polymer (B), from the viewpoint of compatibility and the like, it is preferable to use polymers of the same kind. That is, when the polymer (A) is an acrylic polymer, it is preferable that the polymer (B) is also an acrylic polymer. When the polymer (A) is a urethane polymer, the polymer (B) is preferably also a urethane polymer.

When the weight average molecular weight of the polymer (A) is high in the adhesive layer, the first mesh tends to have a softer and more stretched structure. On the other hand, when the weight average molecular weight of the polymer (B) is low, the second mesh is Eyes tend to be harder and more brittle. In addition, the second mesh easily enters the first mesh, and it is easy to form a dual network. From these viewpoints, it is preferable that the weight average molecular weight of the polymer (A) is higher than the weight average molecular weight of the polymer (B).

The content of the polymer (B) in the adhesive composition is preferably 10 to 100 parts by mass, more preferably 20 to 80 parts by mass, and more preferably 30 parts by mass relative to 100 parts by mass of the polymer (A). ~ 70 parts by mass.

When the content of the polymer (B) is equal to or more than the above-mentioned lower limit value, appropriate energy ray-hardenability can be imparted to the adhesive layer. In addition, when the content is within the above range, the applicability of the adhesive composition, the film-forming property of the adhesive layer, and the like tend to be good. Furthermore, the first mesh and the second mesh are formed in a well-balanced manner, and the fracture characteristics of the adhesive layer tend to be good.

In the adhesive composition, the polymer (A) and the polymer (B) are preferred as main components. The total content of the so-called main component polymer (A) and polymer (B) is 50% by mass or more, more preferably 70 to 98% by mass, based on the total amount of the adhesive composition (based on the solid content). It is more preferably 80 to 95% by mass. The solid content in the present invention means all components other than the organic solvent, and includes those that are liquid at room temperature (25 ° C).

(Acrylic polymer)

Hereinafter, a case where each of the polymer (A) and the polymer (B) is an acrylic polymer will be described in more detail.

The acrylic polymer constituting the polymer (A) is a polymer containing a constituent unit derived from a (meth) acrylate, and preferably contains a functional group monomer (a1) derived from a reactive functional group (A1) (hereinafter The acrylic copolymer (A '), which is sometimes referred to as "functional monomer (a1)", preferably contains the structural unit derived from functional monomer (a1) and the alkyl (meth) acrylate ( a2) Acrylic copolymer (A ') as a constituent unit. When the polymer (A) contains a constituent unit derived from an alkyl (meth) acrylate (a2), it is easy to make a good adhesive property of the adhesive layer.

The functional group monomer (a1) is exemplified by a monomer having the above-mentioned reactive functional group (A1), and a carboxyl group-containing monomer is preferred. Examples of the carboxyl group-containing monomer include carboxylic acids having ethylenically unsaturated bonds such as acrylic acid, methacrylic acid, crotonic acid, maleic acid, itaconic acid, and citraconic acid. These may be used individually by 1 type, and may be used in combination of 2 or more type. Among these, acrylic acid and methacrylic acid are preferred, and acrylic acid is more preferred.

The content of the constituent units derived from the functional group monomer (a1) (for example, a carboxyl group-containing monomer) is preferably 0.5 to 15% by mass, more preferably 1 to 8% by mass based on the acrylic copolymer (A '). It is more preferably 1.5 to 5 mass%. When the content of the functional group monomer (a1) such as a carboxyl group-containing monomer is within the above range, it is easy to impart an appropriate adhesive force to the adhesive layer. In addition, the first mesh can be formed better by the crosslinking of the crosslinking agent (C).

Examples of the alkyl (meth) acrylate (a2) include alkyl (meth) acrylates having 1 to 20 carbon atoms in the alkyl group.

Examples of the alkyl (meth) acrylate having 1 to 20 carbon atoms include, for example, methyl (meth) acrylate, ethyl (meth) acrylate, propyl (meth) acrylate, and (meth) N-butyl acrylate, n-amyl (meth) acrylate, n-hexyl (meth) acrylate, 2-ethylhexyl (meth) acrylate, isooctyl (meth) acrylate, n-decyl (meth) acrylate Esters, n-dodecyl (meth) acrylate, myristyl (meth) acrylate, palmityl (meth) acrylate, stearyl (meth) acrylate, and the like. These may be used individually by 1 type, and may use 2 or more types together.

Among the above, from the viewpoint of appropriately exhibiting adhesiveness, an (meth) acrylic acid alkyl ester having 1 to 8 carbon atoms is preferred, and an (meth) alkyl having 4 to 8 carbon atoms is more preferred. Alkyl acrylate (hereinafter sometimes referred to as "monomer (α)"). Specifically, as the monomer (α), 2-ethylhexyl (meth) acrylate and n-butyl (meth) acrylate are preferable, and n-butyl (meth) acrylate is more preferable.

The content of the constituent unit derived from the (meth) acrylic acid alkyl ester (a2), based on the acrylic copolymer (A '), is preferably 50 to 99.5% by mass, more preferably 60 to 99% by mass, and even more preferably 70-99.8% by mass.

In addition, as the (meth) acrylic acid alkyl ester (a2), the (meth) acrylic acid alkyl ester having a carbon number of 4 to 8 as described above, that is, the monomer (α), but also The (meth) acrylic acid alkyl ester (a2) contained in the acrylic copolymer (A ') may be all monomers (α), or a part of them may be monomers (α).

Specifically, the monomer (α) is preferably 70 to 100% by mass, more preferably 80 to 100% by mass, and still more preferably 90 to 100% with respect to the total amount of the (meth) acrylic acid alkyl ester (a2). quality%.

The acrylic copolymer (A ') used in the polymer (A) may be a copolymer of a functional group monomer (a1) and an alkyl (meth) acrylate (a2), or may be a component (a1), ( a2) A copolymer of the component and the monomer (a3) other than the component (a1) and the component (a2).

The other monomer (a3) means a copolymerizable monomer other than the (a1) component and the (a2) component, and specifically, for example, a (meth) acrylic ring having 3 to 20 carbon atoms in a cycloalkyl group Alkyl esters, benzyl (meth) acrylate, isobornyl (meth) acrylate, and other (meth) acrylates having a cyclic skeleton; vinyl ester compounds such as vinyl acetate and vinyl propionate; ethylene, Olefins such as propylene and isobutylene; halogenated olefins such as vinyl chloride and vinylidene chloride; styrene-based monomers such as styrene and α-methylstyrene; butadiene, isoprene, and chloroprene Diene monomers; nitrile monomers such as acrylonitrile and methacrylonitrile.

The other monomer (a3) may be used alone or in combination of two or more kinds in the acrylic copolymer (A ').

The acrylic polymer constituting the polymer (B) is a polymer containing constituent units derived from a (meth) acrylate, and it is preferred that the energy ray polymerizable group-containing compound (S) has an energy ray polymerizable group (B2). It is obtained by reacting with an acrylic copolymer (B ') containing a constituent unit derived from a functional monomer (b1) having a reactive functional group (B1) (hereinafter sometimes referred to as "functional monomer (b1)"). Reactant. The energy ray polymerizable group-containing compound (S) reacts with a part of the reactive functional group (B1) of the acrylic copolymer (B '). The acrylic copolymer (B ') further preferably contains a structural unit derived from an alkyl (meth) acrylate (b2).

That is, the acrylic polymer constituting the polymer (B) is preferably an energy ray polymerizable group-containing compound (S) having an energy ray polymerizable group (B2) and a functional group derived from a reactive functional group (B1). A reactant obtained by partially reacting a constituent unit of the monomer (b1) with a reactive functional group (B1) of the acrylic copolymer (B ') derived from the constituent unit of the alkyl (meth) acrylate (b2).

The functional group monomer (b1) is a monomer having the above-mentioned reactive functional group (B1), and preferred examples thereof include a hydroxyl group-containing monomer.

Examples of the hydroxyl-containing monomer include 2-hydroxyethyl (meth) acrylate, 2-hydroxypropyl (meth) acrylate, 3-hydroxypropyl (meth) acrylate, and 2-hydroxybutyl (meth) acrylate. Esters, hydroxyalkyl (meth) acrylates, etc., such as 3-hydroxybutyl (meth) acrylate, 4-hydroxybutyl (meth) acrylate, and the like. Among these, from the viewpoints of reactivity with the cross-linking agent (D) and the energy ray polymerizable group-containing compound (S) and copolymerizability with other monomers, 2-hydroxyethyl (meth) acrylate, 4-hydroxybutyl (meth) acrylate, more preferably 2-hydroxyethyl (meth) acrylate.

In the acrylic copolymer (B ') used to form the polymer (B), the content of the constituent units derived from the functional group monomer (b1) (for example, a hydroxyl-containing monomer) is based on the acrylic copolymer (B') basis. It is preferably 10 to 45 mass%, more preferably 15 to 40 mass%, and still more preferably 20 to 35 mass%. When the content of the functional group monomer (b1) such as a hydroxyl-containing monomer is within the above range, it is easy to impart an appropriate adhesive force to the adhesive layer. Furthermore, the cross-linking agent (D) can form a three-dimensional mesh structure formed by the second mesh. Furthermore, by reacting the energy ray polymerizable group-containing compound (S) with the reactive functional group (B1), an appropriate amount of the energy ray polymerizable group (B2) can be introduced into the polymer (B).

Specific examples of the (meth) acrylic acid alkyl ester (b2) are exemplified as those in which the component (a2) is the same. These may be used alone or in combination of two or more. As with the component (a2), an alkyl (meth) acrylate having 1 to 8 carbon atoms is preferred, and the monomer (α) is more preferably used. The preferred compound as the monomer (α) is also preferably n-butyl (meth) acrylate.

The content of the (meth) acrylic acid alkyl ester (b2), based on the acrylic copolymer (B '), is preferably 50 to 90% by mass, more preferably 60 to 85% by mass, and even more preferably 65 to 80. quality%.

Also, the alkyl (meth) acrylate (b2) is the same as the component (a2), and the monomer (α) is preferably used, but it may be an alkyl (meth) acrylate contained in the acrylic copolymer (B '). All of the esters (b2) are monomers (α), and some of them may be monomers (α). The content details are the same as those described in the polymer (A).

The acrylic copolymer (B ') may be a copolymer of a functional monomer (b1) and an alkyl (meth) acrylate (b2), but it may also be a component (b1), a component (b2), and these (b1) ) And a copolymer of monomers (b3) other than (b2). The other monomer (b3) means a copolymerizable monomer other than the components (b1) and (b2), and specifically, it can be appropriately selected and used from those listed as the monomer (a3).

(Energy ray polymerizable group-containing compound (S))

The energy ray polymerizable group-containing compound (S) has an energy ray polymerizable group (B2) and a functional group (B3) capable of reacting with a reactive functional group (B1) (hereinafter sometimes referred to simply as "functional group (B3)") . The functional group (B3) may be a functional group capable of reacting with the reactive functional group (B1), and examples thereof include an isocyanate group, an epoxy group, and a carboxyl group.

As described above, if the reactive functional group (B1) is a hydroxyl group, the functional group (B3) contained in the energy ray-polymerizable group-containing compound (S) is preferably an isocyanate group. When the reactive functional group (B1) is a carboxyl group, the functional group (B3) is preferably an epoxy group. When the reactive functional group (B1) is an epoxy group, the functional group (B3) is preferably a carboxyl group.

Among these, from the viewpoint of reactivity and the like, it is more preferable that the reactive functional group (B1) is a hydroxyl group and the functional group (B3) is an isocyanate group.

The preferable aspect of the energy ray polymerizable group (B2) is as described above. Therefore, as the energy ray polymerizable group-containing compound (S), a compound having an isocyanate group and a (meth) acrylfluorenyl group is preferable.

Specific examples of the energy ray polymerizable group-containing compound (S) include 2-isocyanate ethyl (meth) acrylate, propyl isocyanate (meth) acrylate, and 1,1- (bispropenyloxy) isocyanate. Compounds having isocyanate groups and (meth) acrylfluorenyl groups such as methylmethyl) ethyl esters; compounds having epoxy groups and (meth) acrylfluorenyl groups such as glycidyl (meth) acrylate, etc. Medium is preferably 2-isocyanate ethyl (meth) acrylate.

Here, a part of the reactive functional group (B1) of the acrylic copolymer (B ') is reacted with the energy ray polymerizable group-containing compound (S). Therefore, in the polymer (B), the reactive functional group (B1) which has not reacted with the energy ray polymerizable group-containing compound (S) remains, whereby the polymer (B) becomes a polymer having a reactive functional group (B1) and Both energy ray polymerizable groups (B2).

The addition rate of the energy ray polymerizable group-containing compound (S) is preferably 75 to 97 equivalents, more preferably 75 to 97 equivalents relative to the total amount of reactive functional groups (B1) of the acrylic copolymer (B '). 80 to 95 equivalents, and more preferably 85 to 93 equivalents.

The content of the constituent unit derived from the functional group monomer (b1) is in the above-mentioned preferred range (10 to 45 mass%, more preferably 15 to 40 mass%, and more preferably 20 to 35 mass%), and the addition rate is in these ranges Inside, a certain amount of the reactive functional group (B1) remains in the polymer (B). Therefore, the polymer (B) is appropriately cross-linked by the cross-linking agent (D), and it is easy to adjust the adhesive force of the adhesive layer to an appropriate value. Furthermore, an appropriate amount of the energy ray polymerizable group (B2) may be introduced into the polymer (B).

The acrylic copolymer (A ') and the acrylic copolymer (B') may be a random copolymer or a block copolymer.

The acrylic copolymer (A ') and the acrylic copolymer (B') can be produced by polymerizing a mixture of monomers constituting each copolymer by a usual radical polymerization method. The polymerization can be performed using a polymerization initiator as desired, and can be performed by a solution polymerization method or the like. Examples of the polymerization initiator include conventional azo compounds and organic peroxides.

When both the polymer (A) and the polymer (B) are acrylic polymers, the weight average molecular weight of the acrylic polymer constituting the polymer (A) is preferably higher than the weight average molecular weight of the acrylic polymer constituting the polymer (B). And the difference is more than 200,000. In this way, when the molecular weight difference between the two polymers is increased, it is easy to show the difference in characteristics between the first mesh and the second mesh, and it is also easy to form a dual network. Therefore, the fracture characteristics are improved, and the paste residue is easily reduced. From these viewpoints, the weight average molecular weight difference is more preferably 300,000 or more, and more preferably 400,000 or more.

On the other hand, the upper limit of the weight average molecular weight difference is not particularly limited, and the difference is preferably 850,000 or less, more preferably 750,000 or less, and still more preferably 700,000 or less.

In the adhesive composition, the weight average molecular weight of the acrylic polymer constituting the polymer (A) is 300,000 to 1,000,000, and the weight average molecular weight of the acrylic polymer constituting the polymer (B) is 5,000 to 100,000.

Among them, the weight average molecular weight of the acrylic polymer constituting the polymer (A) is more preferably 350,000 to 850,000, and still more preferably 400,000 to 750,000.

On the other hand, the weight average molecular weight of the acrylic polymer constituting the polymer (B) is more preferably 15,000 to 90,000, and still more preferably 30,000 to 80,000.

When the weight average molecular weight of the polymer (A) is equal to or more than the above-mentioned lower limit value, the first mesh structure is softer and easier to stretch. Moreover, it is easy to make the film forming property of the adhesive layer good, and it is also easy to improve the cohesive force of the adhesive layer, and it is not easy to cause paste residue. On the other hand, when the weight average molecular weight of the polymer (A) is equal to or less than the above-mentioned upper limit value, it is easy to make the coating composition and the like of the adhesive composition good.

In addition, when the weight average molecular weight of the polymer (B) is equal to or less than the above-mentioned upper limit value, the second mesh is likely to have a harder and more brittle structure, and it is easy to form a better dual network. In addition, when it is at least the above-mentioned lower limit value, the cohesive force of the adhesive layer is easy to be made, and paste residue is hardly generated.

(Urethane polymer)

Next, a case where the polymer (A) and the polymer (B) are urethane polymers will be described. The urethane polymer used for the polymer (A) and the polymer (B) is a polymer containing at least one of a urethane bond and a urea bond.

The urethane polymer constituting the polymer (A) is one having the above-mentioned reactive functional group (A1), and examples thereof include a carboxyl group-containing polyurethane.

Examples of the urethane polymer constituting the polymer (B) include those obtained by reacting the above-mentioned energy ray polymerizable group-containing compound (S) with a part of the hydroxyl groups of the hydroxy-containing polyurethane. The hydroxyl-containing polyurethane is preferably one having a hydroxyl group at the terminal. An example of the polyurethane having a hydroxyl group at the terminal is a polyurethane polyol obtained by reacting a polyol with a polyisocyanate compound. As the polyol and the polyisocyanate compound, various compounds conventionally used in urethane-based adhesives can be used.

When the polymer (A) and the polymer (B) are both urethane polymers, the weight average molecular weight of the polymer (A) is preferably higher than the weight average molecular weight of the polymer (B), and the difference is 25,000 Above, more preferably above 50,000. The upper limit of the weight average molecular weight difference is not particularly limited, but the difference is preferably 230,000 or less, and more preferably 120,000 or less.

The weight average molecular weight of the urethane polymer constituting the polymer (A) is preferably from 30,000 to 250,000, more preferably from 40,000 to 150,000.

The weight average molecular weight of the urethane polymer constituting the polymer (B) is preferably 2,000 to 25,000, more preferably 3,000 to 20,000.

[Crosslinking agent (C), (D)]

The cross-linking agent (C) is a cross-linking agent that reacts with the reactive functional group (A1), and is used to cross-link the polymer (A). The crosslinking agent (D) is a crosslinking agent that reacts with the reactive functional group (B1), and is used to crosslink the polymer (B).

Crosslinking by a crosslinking agent (C) and a crosslinking agent (D) is normally performed by heating an adhesive composition. That is, as described later, the adhesive composition is heated in a state of being applied as a film, and becomes an adhesive layer crosslinked by the crosslinking agent (C) and the crosslinking agent (D).

The cross-linking agent (C) and the cross-linking agent (D) are selected from, for example, an isocyanate-based cross-linker, an epoxy-based cross-linker, an amine-based cross-linker, a melamine-based cross-linker, and an aziridine-based cross-linker. , Hydrazine-based crosslinking agent, aldehyde-based crosslinking agent, oxazoline-based crosslinking agent, metal alkyd-based crosslinking agent, metal chelating agent-based crosslinking agent, metal salt-based crosslinking agent, and ammonium salt-based crosslinking agent Agent. Each of the crosslinking agent (C) and the crosslinking agent (D) may be used singly or in combination of two or more kinds.

The crosslinking agent (C) is appropriately selected according to the type of the reactive functional group (A1) possessed by the polymer (A), and the crosslinking agent (D) is corresponding to the reactive functional group possessed by the polymer (B) The type of (B1) is appropriately selected. That is, the cross-linking agent (C) may be selected as long as it does not react with the reactive functional group (B1) but reacts with the reactive functional group (A1). In addition, as the cross-linking agent (D), it may be selected as long as it does not react with the reactive functional group (A1) but reacts with the reactive functional group (B1). Therefore, the crosslinking agent (C) and the crosslinking agent (D) are different from each other.

For example, as described above, when the reactive functional group (A1) is a carboxyl group, the crosslinking agent (C) is preferably selected from an epoxy-based crosslinking agent and a metal chelate-based crosslinking agent, and more preferably an epoxy-based crosslinking agent. Crosslinking agent. When the reactive functional group (A2) is a hydroxyl group, the crosslinking agent (D) is preferably an isocyanate-based crosslinking agent.

Examples of the epoxy-based crosslinking agent include 1,3-bis (N, N'-diglycidylaminomethyl) cyclohexane, N, N, N ', N'-tetraglycidyl-methane -Xylene diamine, ethylene glycol diglycidyl ether, 1,6-hexanediol diglycidyl ether, trimethylolpropane diglycidyl ether, diglycidyl aniline, diglycidylamine, and the like. These may be used individually by 1 type, and may be used in combination of 2 or more type. As the epoxy-based crosslinking agent, among these, 1,3-bis (N, N'-diglycidylaminomethyl) cyclohexane is preferred.

Examples of the metal chelate-based cross-linking agent include, for example, acetylacetone, ethyl, and the like on a polyvalent metal such as aluminum, iron, copper, zinc, tin, titanium, nickel, antimony, magnesium, vanadium, chromium, zirconium, and the like.醯 Ethyl acetate, tris (2,4-glutarate), and other compounds. These may be used individually by 1 type, and may be used in combination of 2 or more type.

Examples of the isocyanate-based crosslinking agent include polyisocyanate compounds. Specific examples of the polyisocyanate compound include aromatic polyisocyanates such as toluene diisocyanate, diphenylmethane diisocyanate, and xylene diisocyanate; aliphatic polyisocyanates such as hexamethylene diisocyanate; isophorone diisocyanate, Cycloaliphatic polyisocyanates and the like such as hydrogenated diphenylmethane diisocyanate and the like. In addition, the ureton bodies and isocyanurate bodies are also exemplified, and the low molecular weight active hydrogen compounds containing ethylene glycol, propylene glycol, neopentyl glycol, trimethylolpropane, and castor oil are also exemplified. Adducts of reactants and the like.

These may be used individually by 1 type, and may be used in combination of 2 or more type. Among the above, a polyhydric alcohol (for example, trimethylolpropane) adduct of an aromatic polyisocyanate such as toluene diisocyanate is preferred.

The cross-linking agent (C) tends to make the first mesh soft and stretchable by suppressing the content, and it is easy to reduce paste residue. Therefore, the content of the crosslinking agent (C) is preferably smaller. Specifically, the content of the cross-linking agent (C) of the adhesive composition also depends on the type and molecular weight of the polymer (A), but is preferably 0.05 to 5 parts by mass relative to 100 parts by mass of the polymer (A). It is preferably 0.1 to 3 parts by mass, and more preferably 0.1 to 0.3 parts by mass.

On the other hand, since a large amount of the cross-linking agent (D) is contained in the adhesive composition, the second mesh tends to have a hard and brittle structure. Therefore, the content of the cross-linking agent (D) is more preferable, and the content of the cross-linking agent (D) of the adhesive composition is preferably more than the content of the cross-linking agent (C) on a mass basis.

The content of the cross-linking agent (D) in the adhesive composition also depends on the type and molecular weight of the polymer (B), but specifically, it is preferably 2 to 20 masses relative to 100 parts by mass of the polymer (B). It is preferably 4 to 16 parts by mass, and more preferably 5 to 12 parts by mass.

[Photopolymerization initiator (E)]

The adhesive composition preferably contains a photopolymerization initiator (E). When the adhesive layer contains a photopolymerization initiator (E), the adhesive layer can be easily hardened by energy rays such as ultraviolet rays.

Examples of the photopolymerization initiator (E) include acetophenone, 2,2-diethoxybenzophenone, 4-methylbenzophenone, and 2,4,6-trimethyldiphenyl. Methyl ketone, Michler's ketone, Benzoin, Benzoin methyl ether, Benzoin ether, Benzoin isopropyl ether, Benzoin isobutyl ether, Benzyl diphenyl sulfide, Tetramethylthiuram Sulfide, benzyl dimethyl acetal, biphenyl hydrazone, biacetamidine, 1-chloroanthraquinone, 2-chloroanthraquinone, 2-ethylanthraquinone, 2,2-dimethoxy-1,2 -Diphenylethane-1-one, 1-hydroxycyclohexylphenyl ketone, 2-methyl-1- [4- (methylthio) phenyl] -2-morpholine acetone-1, 2-benzyl Methyl-2-dimethylamino-1- (4-morpholinylphenyl) butanone-1, 2-hydroxy-2-methyl-1-phenyl-propane-1-one, diethylthioxanthone Low molecular weight polymerization initiators such as isopropyl thioxanthone, 2,4,6-trimethylbenzylidene diphenyl-phosphine oxide; oligomeric {2-hydroxy-2-methyl-1- [4- (1methylvinyl) phenyl] acetone} and other oligomerized polymerization initiators. These can be used alone or in combination of two or more.

The content of the photopolymerization initiator (E) is preferably 0.1 to 20 parts by mass, more preferably 0.5 to 15 parts by mass, relative to 100 parts by mass of the total amount of the polymer (A) and the polymer (B). It is preferably 2 to 12 parts by mass.

In addition, the adhesive layer may contain an adhesion-imparting agent, a dye, a pigment, a deterioration preventing agent, an antistatic agent, a flame retardant, a silane coupling agent, a chain transfer agent, a plasticizer, etc., as long as the effect of the present invention is not impaired. Fillers, resin components other than the polymer (A) and the polymer (B), and the like are components other than the above.

The thickness of the adhesive layer can be appropriately adjusted corresponding to the height of the bumps on the wafer surface, etc., and the surface state to which the adhesive sheet is attached, but it is preferably 2 to 150 μm, more preferably 5 to 100 μm, and even more preferably 8 to 50 μm.

[Fracturing characteristics of the adhesive layer]

The adhesive layer is hardened by irradiating energy rays as described above, and the fracture characteristics after hardening of the energy rays are preferably as follows.

That is, the adhesive layer system with a better energy ray hardening has a breaking stress of 1.5 MPa or more, a breaking elongation of 80% or more, and a breaking energy of 1.0 MJ / m ³ or more. When the breaking stress, elongation at break, and breaking energy are at such high values, the breaking strength of the adhesive layer becomes good, and it is difficult to generate paste residue. In addition, from the viewpoint of more preventing paste residue, it is better that the above-mentioned breaking stress is 1.8 MPa or more, the breaking elongation is 100% or more, the breaking energy is 1.4 MJ / m ³ or more, and the breaking stress is more than 2.0 MPa, and the fracture is more preferable. The elongation is 180% or more, and the breaking energy is 1.8MJ / m ³ or more.

In addition, these upper limits are not particularly limited, but in practice, it is preferable that the breaking stress is 10 MPa or less, the breaking elongation is 400% or less, the breaking energy is 5.0 MJ / m ³ or less, and the more preferable breaking stress is 6 MPa or less, and the breaking elongation It is 300% or less and the breaking energy is 3.5MJ / m ³ or less.

The breaking stress, breaking elongation, and breaking energy mean values measured by a tensile test in accordance with JIS K7127: 1999, and specifically, values obtained by measuring by a method described in Examples described later.

(Hysteresis)

When the adhesive layer after the energy ray hardening has the above-mentioned double network, when a certain strain is applied, the second mesh is destroyed, and on the other hand, the first mesh is not damaged and remains. Therefore, when the adhesive layer after energy ray hardening is reapplied after a certain strain is applied, the second mesh is damaged and the stress-strain characteristics are different from those in the initial stage. These properties are called hysteresis, but the presence and magnitude of the hysteresis can be confirmed by the cyclic tensile test shown below.

That is, the elongation (%) increases with each increase in the number of times of elongation, and the cyclic tensile test is performed by repeating the tensile (strain) and release of the sample until the sample breaks, as shown in Fig. 1. stress-strain curve. Therefore, by detecting the maximum value of the stress difference (DSmax) between the curves with the same elongation in a plurality of stress-strain curves, for example, the presence and magnitude of hysteresis can be confirmed.

FIG. 1 shows a stress-strain curve when a cyclic tensile test is performed on a sample after the energy ray hardening of the adhesive layer used in Example 1 described later. Figure 1 shows an example where the maximum elongation at each elongation is 50% (first time) after each increase of 50%, and when the elongation and release are repeated, the sample breaks at 233% at the fifth elongation. Here, FIG. 1 shows the stress-strain curve at each elongation. From the complex stress-strain curve, the maximum value of the stress difference (DSmax) is obtained as shown in FIG. 1. In the example of FIG. 1, the maximum value of the stress difference (DSmax) is calculated from the curve made twice from the fourth and fifth consecutive times. However, it is also possible to calculate the maximum value of the discontinuity from the second and third consecutive times such as the third and fifth. Calculate the curve made twice. The term "elongation" refers to the ratio of the length of the increase amount when the sample is stretched divided by the original length and expressed in%.

If a double network is properly formed, the higher the hysteresis, the more the stress-strain curves of the plural are separated from each other, and the maximum value of the stress difference (DSmax) becomes larger. Based on the viewpoint of proper formation of a double network, the maximum value of the stress difference (DSmax) is preferably 20% or more, more preferably 25% or more, and more preferably 20% or more relative to the stress (BS) at break in a cyclic tensile test. More than 35%. From the standpoint of ease of production, etc., the maximum value of the stress difference (DSmax) is preferably 90% or less, and more preferably 60% or less.

[Peeling force of adhesive sheet]

Since the adhesive layer is hardenable by energy rays, it is relatively soft before the energy rays are irradiated, so that the adhesive layer can easily follow the unevenness formed on the surface of the workpiece. In addition, the adhesive sheet is hardened by irradiating energy rays to reduce the adhesive force, and is easily peeled from the workpiece.

The adhesive force after the energy ray irradiation of the adhesive sheet is preferably 1,700 mN / 25 mm or less. When the adhesive sheet is bonded to a workpiece having a protrusion such as a bump on the surface, the protrusion is usually in a state where the protrusion is embedded by the adhesive layer or the adhesive layer and the intermediate layer of the adhesive sheet. Therefore, paste residue is liable to occur when the adhesive sheet is later peeled off. However, by setting the adhesive force to 1,700 mN / 25 mm or less, it is easy to prevent such paste residue from occurring. In addition, the adhesive sheet is easily peeled from the workpiece. The adhesive force after the energy ray irradiation of the adhesive sheet is preferably 50 to 1,500 mN / 25 mm, more preferably 100 to 1,300 mN / 25 mm.

In addition, although the adhesive force before the energy ray irradiation of the adhesive sheet is greater than, for example, 1,700 mN / 25 mm, it is preferably 1,800 to 20,000 mN / 25 mm, and more preferably 1,800 to 9,000 mN / 25 mm. When the adhesive force before the energy ray irradiation falls within this range, the adhesion to the surface of the workpiece becomes better, and the protective performance of the adhesive sheet to the workpiece is improved.

In addition, the adhesive force of the adhesive sheet is measured when the adhesive layer of the adhesive sheet is attached to a silicon mirror wafer and peeled at a peeling angle of 180 ° and a peeling speed of 300 mm / min in an environment of 23 ° C. Specifically, Are those measured by the method described in the examples described later.

Adhesion can be changed by appropriately changing the types of polymer (A) and polymer (B), the blending amount of these polymers, the types of cross-linking agent (C) and cross-linking agent (D), and the cross-linking agents. The amount of adjustment is adjusted. For example, when the polymer (A) and the polymer (B) are acrylic polymers as described above, an adhesive sheet having the above-mentioned adhesive force is easily obtained. In addition, by increasing the blending amount of the cross-linking agent (C) and the cross-linking agent (D), it is easy to reduce the adhesive force.

The adhesive force after energy ray irradiation can also be adjusted by the amount of the energy ray polymerizable group (B2) and the blending amount of the polymer (B). The adhesive force after the energy ray irradiation has, for example, a tendency that the amount of the energy ray polymerizable group (B2) contained in the adhesive composition increases, decreases, and decreases.

<Middle layer>

The adhesive sheet of the present invention may be provided with an intermediate layer on one side of the substrate. The adhesive sheet has an intermediate layer, and even if the unevenness of the surface of the workpiece provided with bumps or the like on the workpiece is large, the convex portion can be embedded in the adhesive layer and the intermediate layer. Therefore, the surface of the adhesive sheet which is opposite to the surface attached to the workpiece is easily kept flat.

The thickness of the intermediate layer can be appropriately adjusted according to the state of the adhered surface to which the adhesive sheet is attached, but from the viewpoint that relatively high bumps can also absorb, it is preferably 10 to 600 μm, more preferably 25 to 550 μm, and It is more preferably 35 to 500 μm.

The intermediate layer is formed of a resin composition for an intermediate layer. The resin composition for the intermediate layer preferably contains a urethane (meth) acrylate (X).

(Urethane (meth) acrylate (X))

The urethane (meth) acrylate (X) is a compound having at least a (meth) acrylfluorenyl group and a urethane bond, and has a property of being polymerized by irradiating energy rays. The number of (meth) acrylic acid groups in the urethane (meth) acrylate (X) may be monofunctional, bifunctional, or trifunctional, but the resin composition for the intermediate layer preferably contains a monofunctional amine. Carbamate (meth) acrylate. Since the monofunctional urethane (meth) acrylate does not participate in the formation of the three-dimensional mesh structure in the polymerization structure, the intermediate layer is not easy to form the three-dimensional mesh structure, and it is easy to follow the unevenness of the surface of the workpiece.

As the urethane (meth) acrylate (X), for example, a compound (x3) having a (meth) acrylfluorenyl group can be obtained by reacting a polyol compound (x1) with a polyisocyanate compound (x2). It is obtained by reacting a terminal isocyanate urethane prepolymer.

The urethane (meth) acrylate (X) can be used singly or in combination of two or more kinds.

The polyol compound (x1) for forming a urethane (meth) acrylate (X) is not particularly limited as long as it is a compound having two or more hydroxyl groups. Specific examples of the polyol compound (x1) include alkanediol, polyether polyol, polyester polyol, polycarbonate polyol, and the like. Among these, polyether polyols are preferred.

The polyol compound (x1) may be any of a bifunctional polyol, a trifunctional polyol, and a tetrafunctional or higher polyhydric alcohol. However, from the viewpoints of availability, wide useability, reactivity, and the like, It is preferably a bifunctional polyol, more preferably a polyether polyol. Examples of the polyether polyol include polyethylene glycol, polypropylene glycol, and polytetramethylene glycol.

The polyester polyol is obtained by polycondensing a polyol component and a polyacid component. Examples of the polyol component include various alkanediols such as ethylene glycol, diethylene glycol, and butanediol (preferably alkanediols having about 2 to 10 carbon atoms), various diols, and the like.

As the polybasic acid component used in the production of the polyester polyol, compounds known as polybasic acid components of polyesters can be used. Specifically, examples include adipic acid, sebacic acid, and the like having aliphatic carbons having 4 to 20 carbon atoms, aromatic dibasic acids such as terephthalic acid, and aromatic polybasic acids such as trimellitic acid. 、 The corresponding acid anhydrides, their derivatives, dimer acids, hydrogenated dimer acids, etc.

The polycarbonate polyol is not particularly limited, and examples thereof include reactants of glycols and alkylene carbonate.

Examples of the polyisocyanate compound (x2) include, for example, aliphatic polyisocyanates, alicyclic polyisocyanates, and aromatic polyisocyanates. More specifically, they can be used as the crosslinking agent (C) and the crosslinking agent (D). Various polyisocyanate compounds exemplified.

Examples of the compound (x3) having a (meth) acrylfluorenyl group include a (meth) acrylate having a hydroxyl group. The (meth) acrylate having a hydroxyl group is not particularly limited, and for example, a hydroxyalkyl (meth) acrylate is preferred. As the hydroxyalkyl (meth) acrylate, the same ones as those exemplified for the above-mentioned hydroxyl-containing monomer can be used.

The weight average molecular weight of the urethane (meth) acrylate (X) for the resin composition for the intermediate layer is preferably 1,000 to 100,000, more preferably 3,000 to 80,000, and still more preferably 5,000 to 65,000. When the weight average molecular weight is 1,000 or more, a polymer of a urethane (meth) acrylate (X) and a polymerizable monomer (Z) described later can impart moderate hardness to the intermediate layer.

The blending amount of the urethane (meth) acrylate (X) in the resin composition for the intermediate layer is preferably 10 to 70 mass based on the total amount of the resin composition (solid content basis) for the intermediate layer. %, More preferably 20 to 70% by mass, still more preferably 25 to 60% by mass, and even more preferably 30 to 50% by mass. If the blending amount of the urethane (meth) acrylate (X) is within this range, the intermediate layer can easily follow the unevenness on the surface of the workpiece.

In addition to the urethane (meth) acrylate (X), the resin composition for the intermediate layer preferably further contains, for example, one selected from a thiol group-containing compound (Y) and a polymerizable monomer (Z). One or more of the group preferably contains both of them.

(Thiol group-containing compound (Y))

The thiol group-containing compound (Y) is not particularly limited as long as it is a compound having at least one thiol group in the molecule, but it is preferably a polyfunctional thiol group-containing compound, more preferably a 4-functional thiol-containing compound. Based compound.

Specific examples of the thiol group-containing compound (Y) include nonylthiol, 1-dodecylthiol, 1,2-ethanedithiol, 1,3-propanedithiol, and triazinethiol. Alcohol, triazine dithiol, triazine trithiol, 1,2,3-propane trithiol, tetraethylene glycol-bis (3-mercaptopropionate), trimethylolpropane tri (3-mercapto Propionate), pentaerythritol tetrakis (3-mercaptopropionate), pentaerythritol tetraglycolate, dipentaerythritol hexa (3-mercaptopropionate), tris [(3-mercaptopropionyloxy) ethyl] isocyanate Urate, 1,4-bis (3-mercaptobutyryloxy) butane, pentaerythritol tetrakis (3-mercaptobutyrate), 1,3,5-tris (3-mercaptobutyryloxyethyl) -1,3,5-triazine-2,4,6- (1H, 3H, 5H) -trione and the like.

These thiol group-containing compounds (Y) can be used singly or in combination of two or more kinds.

The blending amount of the thiol group-containing compound (Y) is preferably 1.0 to 4.9 mass based on 100 mass parts of the total of the urethane (meth) acrylate (X) and the polymerizable monomer (Z) described later. Parts, more preferably 1.5 to 4.8 parts by mass.

(Polymerizable monomer (Z))

The resin composition for an intermediate layer preferably contains a polymerizable elastomer (Z) from the viewpoint of improving film forming properties. The polymerizable elastomer (Z) is a polymerizable compound other than the aforementioned urethane (meth) acrylate (X), and is a compound that can be polymerized by irradiation with energy rays. It should be noted that the polymerizable monomer (Z) means a resin component. The polymerizable monomer (Z) is preferably a compound having at least one (meth) acrylfluorenyl group.

In the present specification, the "resin component" means an oligomer or a high molecular weight body having repeating units in the structure, and a compound having a weight average molecular weight of 1,000 or more.

Examples of the polymerizable monomer (Z) include (meth) acrylic acid alkyl esters having an alkyl group having 1 to 30 carbon atoms, and (a) a functional group having a hydroxyl group, a amine group, an amine group, and an epoxy group. Group) acrylate, (meth) acrylate having an alicyclic structure, (meth) acrylate having an aromatic structure, (meth) acrylate having a heterocyclic structure, other vinyl compounds, and the like.

Examples of the (meth) acrylate having a functional group include hydroxyalkyl (meth) acrylate and the like. As the hydroxyalkyl (meth) acrylate, the same ones as those exemplified for the above-mentioned hydroxyl-containing monomer can be used.

Examples of the (meth) acrylate having an alicyclic structure include, for example, isobornyl (meth) acrylate, dicyclopentenyl (meth) acrylate, dicyclopentyl (meth) acrylate, and (meth) Dicyclopentenyl acrylate, cyclohexyl (meth) acrylate, adamantane (meth) acrylate, and the like.

Examples of the (meth) acrylate having an aromatic structure include, for example, phenylhydroxypropyl (meth) acrylate, benzyl (meth) acrylate, and 2-hydroxy-3-phenoxypropyl (meth) acrylate Wait.

Examples of the (meth) acrylate having a heterocyclic structure include tetrahydrofurfuryl (meth) acrylate, morpholine (meth) acrylate, and the like.

These may be used individually by 1 type, and may use 2 or more types together.

As the polymerizable monomer (Z), (meth) acrylate having at least an alicyclic structure is preferably used, and (meth) acrylate having a functional group and (meth) acrylic acid having an alicyclic structure are more preferable. As the ester, both hydroxyalkyl (meth) acrylate and isobornyl (meth) acrylate are more preferably used.

The blending amount of the polymerizable monomer (Z) in the resin composition for the intermediate layer is preferably 20 to 80% by mass, and more preferably 30 to the total amount of the resin composition for the intermediate layer (based on the solid content). 80% by mass, more preferably 40 to 75% by mass, and even more preferably 50 to 70% by mass. If the blending amount of the polymerizable monomer (Z) is within this range, the mobility of the portion polymerized by the polymerizable monomer (Z) in the intermediate layer is high, so the intermediate layer tends to be soft, and the intermediate layer tends to be soft. Follow the unevenness on the surface of the workpiece.

The blending amount of the (meth) acrylate having an alicyclic structure with respect to the total amount of the polymerizable monomer (Z) contained in the resin composition for the intermediate layer is preferably 52 to 87% by mass. It is more preferably 55 to 85% by mass, and still more preferably 60 to 80% by mass. When the compounding amount of the (meth) acrylate having an alicyclic structure is within this range, the intermediate layer can easily follow the unevenness on the surface of the workpiece.

From the same viewpoint, the mass ratio of the urethane (meth) acrylate (X) to the polymerizable monomer (Z) in the resin composition for an intermediate layer [urethane (meth) acrylic acid The ester (X) / polymerizable monomer (Z)] is preferably 20/80 to 60/40, more preferably 30/70 to 50/50, and still more preferably 35/65 to 45/55.

(Photopolymerization initiator)

The resin composition for an intermediate layer preferably further contains a photopolymerization initiator. By containing a photopolymerization initiator, the resin composition for an intermediate layer can be easily hardened by energy rays such as ultraviolet rays.

The photopolymerization initiator can be appropriately selected and used from those exemplified for the photopolymerization initiator (E). The photopolymerization initiator may be used singly or in combination of two or more kinds.

The blending amount of the photopolymerization initiator is preferably 0.05 to 15 parts by mass relative to 100 parts by mass of the total of the urethane (meth) acrylate (X) and the polymerizable monomer (Z), and more preferably It is 0.1 to 10 parts by mass, and more preferably 0.3 to 5 parts by mass.

(Other additives)

The resin composition for an intermediate layer may contain other additives as long as the effect of the present invention is not impaired. Examples of the other additives include, for example, a cross-linking agent, an antioxidant, a softener (plasticizer), a filler, a rust inhibitor, a pigment, a dye, and the like. When blending these additives, the blending amount of other additives is preferably 0.01 to 6 parts by mass based on 100 parts by mass of the total of the urethane (meth) acrylate (X) and the polymerizable monomer (Z). , More preferably 0.1 to 3 parts by mass.

The resin composition for the intermediate layer may contain a urethane (meth) acrylate in addition to the urethane (meth) acrylate (X), as long as the effect of the present invention is not impaired. (X) Other resin components.

The intermediate layer may be formed of a resin composition for an intermediate layer containing other resin components such as an olefin resin instead of the urethane (meth) acrylate (X).

<Peeling Material>

As the release material provided on the adhesive layer and the release material used in the steps of the manufacturing method described later, a release sheet subjected to a single-sided release treatment, a release sheet subjected to a double-sided release treatment, and the like are used. Those who apply a release agent to the material.

Examples of the substrate for the release material include polyester resin films such as polyethylene terephthalate resin, polybutylene terephthalate resin, polyethylene naphthalate resin, and the like; polypropylene resin, and polyethylene Polyolefin resin films such as resin, plastic films, etc.

Examples of the release agent include rubber-based elastomers such as silicone resins, olefin-based resins, isoprene-based resins, and butadiene-based resins, long-chain alkyl resins, alkyd resins, and fluorine-based resins. Wait.

The thickness of the release material is not particularly limited, but is preferably 5 to 200 μm, and more preferably 10 to 120 μm.

[Manufacturing method of adhesive sheet]

The manufacturing method of the adhesive sheet of the present invention is not particularly limited, and can be manufactured according to a conventional method.

The intermediate layer can be formed by, for example, directly coating a resin composition for an intermediate layer on a surface of one side of a substrate to form a coating film, and then drying and hardening it if necessary. In addition, the intermediate layer may be formed by directly applying the resin composition for the intermediate layer to form a coating film by directly applying the resin composition for the intermediate layer on the release-treated surface of the release material, drying it as necessary, and performing a semi-hardening treatment to form a semi-hardened layer on the release material. This semi-hardened layer is bonded to a substrate, and the semi-hardened layer is completely hardened and formed. In this case, the peeling material may be appropriately removed before or after the semi-hardened layer is completely cured. In addition, the hardening of the intermediate layer is preferably polymerized and hardened by irradiating the coating film with energy rays. The energy ray is preferably ultraviolet. When the intermediate layer is formed using an olefin resin, the intermediate layer may be formed by extrusion molding or the like.

In addition, the adhesive layer is preferably formed by applying the adhesive composition, heating the adhesive composition, crosslinking it, and drying it if necessary. At this time, the adhesive composition may be directly applied to the intermediate layer or the substrate, or may be applied to the release-treated surface of the release material to form an adhesive layer, and then, the adhesive layer may be adhered to the intermediate layer or the substrate. form. The release material disposed on the adhesive layer may be peeled as required.

In addition, if the heating temperature and heating time of the adhesive composition are the temperature at which the polymer (A) is crosslinked by the crosslinking agent (C) and the polymer (B) is crosslinked by the crosslinking agent (D) The time is sufficient, and the heating temperature is usually 80 to 110 ° C, preferably 90 to 100 ° C. The heating time is usually 1 to 5 minutes, preferably 2 to 3 minutes.

When the intermediate layer or the adhesive layer is formed, an organic solvent may be further formulated in the resin composition or the adhesive composition for the intermediate layer to prepare a diluent of the resin composition or the adhesive composition for the intermediate layer. Examples of the organic solvent include methyl ethyl ketone, acetone, ethyl acetate, tetrahydrofuran, dioxane, cyclohexane, n-hexane, toluene, xylene, n-propanol, and isopropanol.

In addition, as these organic solvents, the organic solvents used in the synthesis of each component contained in the resin composition for the intermediate layer or the adhesive composition may be used directly, or one or more other organic solvents may be added thereto.

The resin composition or the adhesive composition for the intermediate layer can be applied by a conventional coating method. Examples of the coating method include a spin coating method, a spray coating method, a bar coating method, a doctor blade coating method, a roll coating method, a blade coating method, a die coating method, and a gravure coating method.

[How to use the adhesive sheet]

The adhesive sheet of the present invention is applied to various workpieces, and users who process semiconductor wafers and the like are preferably applied to workpiece surfaces having unevenness, protrusions, and the like.

Furthermore, it is better to adhere to the surface of a semiconductor wafer, especially to the surface of a wafer on which bumps are formed, and to use it as an adhesive sheet for protecting the surface of a semiconductor wafer. In addition, the adhesive sheet is better attached to the surface of the semiconductor wafer, and is used as a back surface abrasive tape for protecting circuits formed on the surface of the wafer during subsequent wafer back grinding. When the adhesive sheet of the present invention has an intermediate layer, even if the wafer surface has a step difference due to bumps or the like, its embedding property is also good, so the wafer surface protection performance is good.

The adhesive layer of the present invention is of an energy ray hardening type, and is an adhesive sheet attached to the surface of a workpiece such as a semiconductor wafer, which is irradiated with energy ray for energy ray hardening, and then peeled off from the surface of the workpiece. Therefore, since the adhesive sheet is peeled after the adhesive force is reduced, its peelability is good. In addition, when the adhesive sheet after the hardening is peeled off as described above, it is difficult for the paste to remain.

Moreover, when an adhesive sheet is used for a semiconductor wafer, it is not limited to a back-grinding sheet, It can also be used for other uses. For example, the adhesive sheet may be attached to the back of the wafer and used as a dicing sheet holding the wafer when dicing the wafer. The wafer in this case may be a through-electrode or the like, or a protrusion or a bump having a bump or the like on the back surface of the wafer.

    Examples    

Hereinafter, the present invention will be described in more detail based on examples, but the present invention is not limited to these examples.

The measurement method and evaluation method in the present invention are as follows.

[Weight average molecular weight (Mw), number average molecular weight (Mn)]

The measurement was performed using a gel permeation chromatography device (product name "HLC-8220", manufactured by TOSOH Co., Ltd.) under the following conditions, and the value measured in terms of standard polystyrene conversion was used.

(Measurement conditions)

Columns: "TSK protective column HLX-H", "TSK gel GMHXL (× 2)", "TSK gel G2000MHXL" (all manufactured by TOSOH Corporation)

Column temperature: 40 ℃

Development solvent: tetrahydrofuran

Flow rate: 1.0mL / min

[Stretching test]

The tensile test was performed in accordance with JIS K7127: 1999 by the method shown below.

In addition, a measurement sample used in the tensile test was prepared as follows, and the values measured using the measurement sample were used as the breaking stress, elongation at break, and breaking energy of the adhesive layer.

(Production of measurement samples)

In the same manner as in Example 1, an adhesive layer having a polyethylene terephthalate (PET) -based release film (manufactured by LINTEC Corporation, product name "SP-PET381031", and thickness of 38 µm) was prepared on both sides. (40 μm thick). In addition, in the same procedure, five adhesive layers were sandwiched between the same release films. Next, two adhesive layers were prepared by peeling a release film and exposing them, so that the surfaces of the adhesive layers were laminated facing each other. This sequence was repeated, thereby laminating five adhesive layers to obtain an adhesive layer having a thickness of 200 μm sandwiched between two release films.

The obtained laminated body was irradiated with ultraviolet rays using a UV irradiation device (manufactured by LINTEC Corporation, product name "RAD-2000m / 12") under conditions of an irradiation speed of 15 mm / second, an illuminance of 220 mW / cm ² and a light amount of 500 mJ / cm ² . To harden the adhesive layer. The hardened material of the obtained adhesive layer was cut into 15 mm × 140 mm to obtain a measurement sample.

(Measurement of breaking stress, breaking elongation and breaking energy of the adhesive layer)

Labels for sample stretching were affixed to 20 mm portions of the two ends of the above-mentioned measurement samples, and samples with a measurement target portion of 15 mm × 100 mm were produced. A tensile tester (manufactured by Shimadzu Corporation, trade name "AUTOGRAPH AG-IS 1kN") was used for this sample, and the breaking stress and elongation at break were measured under conditions of 100 mm between fixtures and a tensile speed of 200 mm / min. . In addition, a stress-strain curve was prepared when measuring the breaking stress and elongation at break, and the area under the curve was calculated to obtain the breaking energy.

[Cyclic tensile test]

In the same manner as the above-mentioned tensile test, a measurement sample was prepared. The tensile test in the cyclic tensile test was performed using the measurement sample, using a tensile tester (manufactured by Shimadzu Corporation, trade name "AUTOGRAPH AG-IS 1kN") at a tensile speed of 200 mm / min and a release speed of 600 mm / min conditions.

The cyclic tensile test is to increase the elongation (%) each time the number of elongations is increased and repeat the elongation (strain) and release of the sample until the sample breaks. At this time, the elongation of each sample is set to increase by a certain percentage every time from 0% elongation. Specifically, the increase in elongation is from 3%, 5%, 8%, 10%, 20%, 30%, 50%, 100%, (100 + 100n)% (n is an integer of 1 or more) For one, the sample is broken at 4 to 6 cycles. That is, for example, when the increase in elongation is 50%, the increase in elongation is 50%, 100%, 150%, 200%, or 250%.

In the cyclic tensile test, a stress-strain curve is prepared for each elongation. The prepared stress-strain curve is recorded on the same tracing paper. From the obtained multiple stress-strain curves, the maximum stress difference between the curves with the same elongation is detected. Value (DSmax).

[Adhesion after energy ray irradiation]

The adhesive sheets of the examples and comparative examples were evenly cut to a width of 25 mm, and were temporarily placed on the silicon mirror wafer of the adherend so that the adhesive layer became the adherend side. The temporarily placed adhesive sheet is reciprocated once with a roller weighing 1 kg, and a load is applied to the adhesive sheet to attach the adhesive sheet to the adherend. After attaching, store at 23 ° C and 50% relative humidity for 20 minutes. Use a UV irradiation device (manufactured by LINTEC Co., Ltd., product name "RAD-2000m / 12") at an illuminance of 220mW / cm ² and a light amount of 500mJ / Under conditions of cm ² and irradiation speed of 15 mm / sec, ultraviolet rays are irradiated from the side of the adhesive sheet. Next, after leaving it at 23 ° C and 50% relative humidity for 5 minutes, a tensile tester (manufactured by ORIENTEC, product name "TENSILON") was used to measure at 23 ° C and 50% relative humidity at a peeling angle of 180 °, peeling speed of 300mm / min, the adhesive force when peeling the adhesive sheet.

[Adhesion before energy ray irradiation]

The measurement was performed in the same manner as described above, except that the ultraviolet irradiation and the subsequent 5 minutes of standing were omitted.

[Evaluation of paste residue]

Prepare wafers with spherical bumps with a bump height of 250 μm, a pitch of 500 μm, and a diameter of 300 μm in plan view (8-inch wafers made by Waltz, bump specifications Sn / Ag / Cu = 96.5 / 3 / 0.5% by mass, The material of the wafer surface is SiO ₂ ) as an adherend.

The adhesive sheet produced in the examples and comparative examples was placed in a state in which the adhesive layer of the adhesive sheet and the bumps of the wafer faced each other, and a laminator (manufactured by LINTEC Corporation, product name "RAD-3510F / 12") was used. ) To attach the adhesive sheet to the wafer. In addition, the laminating table and laminating roll of the laminator were set at 60 ° C. during the application.

After lamination using a UV irradiation apparatus (of LINTEC Co., product name "RAD-2000m / 12"), to ^2, light quantity of 560mJ / cm ^2, the irradiation illuminance speed condition 220mW / cm 15mm / sec, the thin self-adhesive The sheet side is irradiated with ultraviolet rays.

Next, at 50 ° C and 50% relative humidity, use a tensile tester (manufactured by Shimadzu Corporation, product name "AUTOGRAPH AG-IS 1KN") at a tensile speed of 120 mm / min. Peel off the adhesive sheet.

After peeling, the exposed bump formation surface of the wafer was observed with a digital microscope (manufactured by KYENCE Corporation, product name "VHX-1000"), and it was confirmed whether or not a paste remained. Further, the bump portion of the wafer was observed using a scanning electron microscope (manufactured by KYENCE Corporation, product name "VE-9800"), and it was confirmed whether or not a paste remained. In addition, compared with a digital microscope, a scanning electron microscope can observe finer paste residues.

The paste residue was evaluated by the following evaluation criteria.

A: No paste residue was observed in any of the microscopes.

B: No paste residue was observed with a digital microscope, but a slight paste residue was observed with a scanning electron microscope.

C: Residual paste was observed in any microscope.

Next, a substrate with an intermediate layer and an adhesive sheet were produced by the following procedure. In addition, each mass part in the following description is expressed in terms of solid content when it is diluted with a diluent such as an organic solvent.

[Production of substrate with intermediate layer]

40 parts by mass of monofunctional urethane acrylate, 45 parts by mass of isobornyl acrylate (IBXA), 15 parts by mass of 2-hydroxypropyl acrylate (HPA), and pentaerythritol tetrakis (3-mercaptobutyrate) (Showa Electric Co., Ltd., product name "CALENDS MT PE1", class 2 4-functional thiol-containing compound, solid content concentration 100% by mass) 3.5 parts by mass, 1.8 parts by mass of a crosslinking agent, and 2- as a photopolymerization initiator 1.0 mass part of hydroxy-2-methyl-1-phenyl-propane-1-one (manufactured by BASF, product name "DAROCURE 1173", solid content concentration: 100% by mass), to prepare a resin composition for an intermediate layer. This resin composition for an intermediate layer was applied to a PET-based release film (product name: "SP-PET381031", thickness: 38 µm) by a fountain nozzle method to form a coating film.

Next, ultraviolet rays were irradiated from the coating film side to form a semi-hardened layer. For the ultraviolet irradiation, a belt-conveying ultraviolet irradiation device (manufactured by EYE GRAPHICS Co., Ltd., product name "ECS-401GGX") was used as the ultraviolet irradiation device, and a high-pressure mercury lamp (manufactured by EYE GRAPHICS Co., Ltd., product name "K04-L41"") As an ultraviolet source, the irradiation conditions were performed under the conditions of an illuminance of 112 mW / cm ² at a light wavelength of 365 nm and an amount of light of 177 mJ / cm ² (manufactured by EYE GRAPHICS Corporation, product name" UVPF-A1 ").

On the formed semi-hardened layer, a base material made of a PET-based film (manufactured by Toyobo Co., Ltd., product name "COSOMO SHINE A4100", 50 µm thick) was laminated, and ultraviolet radiation was applied from the PET-based film side (using the above) The ultraviolet irradiation device and ultraviolet source were irradiated at 271 mW / cm ² and the light amount was 1,200 mJ / cm ² ), and were completely hardened to form an intermediate layer with a thickness of 300 μm on the PET film of the substrate. An intermediate layer-attached substrate was obtained.

[Example 1]

An acrylic polymer (weight average molecular weight: 600,000) obtained by copolymerizing 97 parts by mass of n-butyl acrylate (BA) and 3 parts by mass of acrylic acid (AA) was prepared as the polymer (A).

For an acrylic copolymer (B ') obtained by copolymerizing 70 parts by mass of n-butyl acrylate (BA) and 30 parts by mass of 2-hydroxyethyl acrylate (2HEA), the hydroxyl copolymer (100 equivalents) derived from 2HEA An acrylic polymer (weight average molecular weight: 50,000) obtained by adding 2-isocyanate ethyl methacrylate (manufactured by Showa Denko Corporation, product name "CALENDS MOI") to a 90 equivalent weight ratio as a polymer ( B).

To a mixture of 100 parts by mass of the polymer (A) and 50 parts by mass of the polymer (B), 2,2-dimethoxy-1,2-diphenylethyl as a photopolymerization initiator (E) was added. 14.9 parts by mass of alkane-1-one (manufactured by BASF, product name "Irgacure651"), trimethylolpropane addition toluene diisocyanate (manufactured by TOSOH Corporation, product name "CORONATE L") as a crosslinking agent (D) ") 4.2 parts by mass of 1,3-bis (N, N-diglycidylaminomethyl) cyclohexane as a crosslinking agent (C) (manufactured by Mitsubishi Gas Chemical Co., Ltd., product name" TETRAD- C ″) 0.19 parts by mass, diluted with an organic solvent (methyl ethyl ketone) to a solid content concentration of 20% by mass, and stirred for 30 minutes to prepare a diluent of the adhesive composition.

Next, the diluted solution of the prepared adhesive composition was coated on a PET-based release film (product of "LINTEC Corporation, product name" SP-PET381031 ", thickness: 38 µm), heated at 100 ° C for 2 minutes to dry, and peeled off. A 10 μm thick adhesive layer was formed on the film.

After removing the release film on the substrate with an intermediate layer previously made, and bonding the exposed intermediate layer to the adhesive layer on the release film, the unnecessary portion in the width direction is cut and removed to obtain the substrate / intermediate layer / adhesion. Adhesive sheet made of agent layer / release sheet. The obtained adhesive sheet and the adhesive layer used in the adhesive sheet were evaluated for breaking stress, breaking elongation, breaking energy, adhesive force, and paste residue according to the above-mentioned evaluation method. The results are shown in Table 1.

The adhesive layer after the energy ray hardening used in Example 1 was subjected to a cyclic tensile test. The complex stress-strain curve obtained in the cyclic tensile test is shown in FIG. 1. The elongation increase in this test is 50%, the elongation (%) is 50% at the first time, 100% at the second time, 150% at the third time, 200% at the fourth time, and 250 at the fifth time. %.

As shown in FIG. 1, in the cyclic tensile test, at the fifth elongation, the specimen at 233% elongation breaks, and the stress at this time is 3.26 MPa.

Further, the maximum value of the stress difference between the curves (DSmax) at the same elongation read from FIG. 1 was 1.55 MPa. DSmax is 48% relative to the stress at break of the cyclic tensile test.

Similarly, the adhesive layer before the energy ray hardening used in Example 1 was subjected to a cyclic tensile test. The complex stress-strain curve obtained in this test is shown in FIG. 2. The increase in elongation in this test is 100%, and the elongation (%) is set to 100% at the first time, 200% at the second time, 300% at the third time, and 400% at the fourth time.

As shown in FIG. 2, before the energy ray hardens, the specimen with a elongation of 321% is broken, and the stress at this time is 1.41 MPa.

In addition, the maximum value (DSmax) of the stress difference between the curves at the same elongation read from FIG. 2 was 0.20 MPa. The DSmax is 14% relative to the stress at break of the cyclic tensile test.

[Example 2]

Except for 1,3-bis (N, N-diglycidylaminomethyl) cyclohexane (Mitsubishi Gas Chemical Co., Ltd., product name "TETRAD-C") as a cross-linking agent (C) An adhesive sheet was produced in the same procedure as in Example 1 except that the amount used was changed to 0.38 parts by mass.

[Example 3]

Except for 1,3-bis (N, N-diglycidylaminomethyl) cyclohexane (Mitsubishi Gas Chemical Co., Ltd., product name "TETRAD-C") as a cross-linking agent (C) An adhesive sheet was produced in the same procedure as in Example 1 except that the amount used was changed to 0.57 parts by mass.

[Comparative Example 1]

For an acrylic copolymer obtained by copolymerizing 90 parts by mass of 2-hydroxyethyl acrylate (2EHA) and 10 parts by mass of 4-hydroxybutyl acrylate (4HBA), the addition ratio to the hydroxyl group (100 equivalents) derived from 4HBA 2-isocyanatoethyl methacrylate (manufactured by Showa Denko Corporation, product name "CALENDS MOI") was added in an amount of 65 equivalents to obtain an acrylic polymer (weight average molecular weight: 1,000,000). To 100 parts by mass of the acrylic polymer, 3 parts by mass of 1-hydroxycyclohexylphenyl ketone (manufactured by BASF, product name "Irgacure184") was added as a photopolymerization initiator, and trimethylol as a crosslinking agent was added. 1.1 parts by mass of methylpropane addition toluene diisocyanate (manufactured by TOSOH Corporation, product name "CORONATE L"), diluted with an organic solvent (methyl ethyl ketone) to a concentration of 20% by mass, and stirred for 30 minutes to prepare an adhesive Diluent of composition. Next, a diluted solution of the prepared adhesive composition was applied to a PET-based release film (product of "LINTEC Corporation, product name" SP-PET381031 ", thickness: 38 µm), heated at 100 ° C for 2 minutes to dry, and peeled A 10 μm thick adhesive layer was formed on the film.

After removing the release film on the substrate with an intermediate layer produced previously, bonding the exposed intermediate layer to the adhesive layer, cutting and removing the unnecessary portion in the width direction, and obtaining the substrate / intermediate layer / adhesive layer / peeling Adhesive flakes formed by flakes. The obtained adhesive sheet and the adhesive layer used in the adhesive sheet were evaluated according to the above-mentioned evaluation method. The results are shown in Table 1.

The adhesive layer after the energy ray hardening used in Comparative Example 1 was subjected to a cyclic tensile test. The complex stress-strain curve obtained in the cyclic tensile test is shown in FIG. 3. The% increase in elongation in this test is 5%, and the elongation setting (%) is 5% in the first time, 10% in the second time, 15% in the third time, and 20% in the fourth time.

As shown in Fig. 3, in the cyclic tensile test, the sample was broken at the elongation of 20% during the fourth elongation, and the stress at this time was 0.94 MPa.

As shown in FIG. 3, the stress-strain curves at the first to fourth elongations all overlap. Therefore, the maximum value of the stress difference between the curves (DSmax) at the same elongation read from the stress-strain curve is 0 MPa, which is 0% relative to the stress at break in the cyclic tensile test.

The adhesive composition of Examples 1 to 3 contains a polymer (A) and a polymer (B), and a cross-linking agent (C) and a cross-linking agent (D), respectively, and the polymer (B) It has an energy ray polymerizable group (B2), so an appropriate double network is formed after energy ray irradiation. Therefore, as is clear from FIG. 1, the adhesive layer exhibits specific hysteresis, and the fracture strength is also good. When the adhesive sheet is peeled off from the workpiece (that is, the bump formation surface of the wafer), it can effectively prevent the paste residue on the surface of the workpiece . The adhesive force values before and after the energy ray irradiation are both appropriate. As is clear from FIG. 2, the adhesive composition of the example showed insufficient hysteresis before energy ray irradiation, and did not form a double network properly.

On the other hand, in Comparative Example 1, since only one polymer and the cross-linking agent for cross-linking the polymer were prepared, the fracture characteristics were not good, and the paste residue could not be prevented properly.

Claims (10)

    An adhesive sheet for semiconductor processing, comprising: a substrate; and an adhesive sheet for semiconductor processing, which is provided on a surface of one side of the substrate and has an adhesive layer formed by an adhesive composition. The adhesive sheet is characterized by the aforementioned adhesive. The agent composition contains a polymer (A) having a reactive functional group (A1), a polymer having a reactive functional group (B1) and an energy ray polymerizable group (B2) different from the reactive functional group (A1). (B), a crosslinking agent (C) that reacts with the reactive functional group (A1), and a crosslinking agent (D) that reacts with the reactive functional group (B1).         The adhesive sheet for semiconductor processing according to claim 1, wherein the weight average molecular weight of the polymer (A) is higher than the weight average molecular weight of the polymer (B).         For example, the adhesive sheet for semiconductor processing of claim 1 or 2, wherein both the polymer (A) and the polymer (B) are acrylic polymers.         The adhesive sheet for semiconductor processing of claim 3, wherein the weight average molecular weight of the acrylic polymer constituting the polymer (A) is higher than the weight average molecular weight of the acrylic polymer constituting the polymer (B), and the difference is 200,000 or more.         For example, the adhesive sheet for semiconductor processing of claim 4, wherein the weight average molecular weight of the acrylic polymer constituting the polymer (A) is 300,000 to 1,000,000, and the weight average molecular weight of the acrylic polymer constituting the polymer (B) is 5,000 to 100,000.         The adhesive sheet for semiconductor processing according to any one of claims 3 to 5, wherein the acrylic polymer constituting the polymer (A) has a composition containing a functional group monomer (a1) having a reactive functional group (A1) Unit, acrylic copolymer (A ') with constituent units derived from alkyl (meth) acrylate (a2).         The adhesive sheet for semiconductor processing according to any one of claims 3 to 6, wherein the acrylic polymer constituting the polymer (B) is an energy ray polymerizable group-containing compound having an energy ray polymerizable group (B2) ( S) Acrylic copolymer (B ') with a structural unit containing a functional unit monomer (b1) having a reactive functional group (B1) and a structural unit derived from an alkyl (meth) acrylate (b2) Part of the reactive functional group (B1) is reacted.         The adhesive sheet for semiconductor processing according to any one of claims 1 to 7, wherein the reactive functional group (A1) is a carboxyl group and the reactive functional group (B1) is a hydroxyl group.         The adhesive sheet for semiconductor processing according to claim 8, wherein the crosslinking agent (C) is an epoxy-based crosslinking agent, and the crosslinking agent (D) is an isocyanate-based crosslinking agent.         The adhesive sheet for semiconductor processing according to any one of claims 1 to 9, wherein, in the adhesive composition, the content of the crosslinking agent (D) is more than the content of the crosslinking agent (C) on a mass basis, The content of the cross-linking agent (D) in the adhesive composition is 2 to 20 parts by mass based on 100 parts by mass of the polymer (B).