WO2022172927A1

WO2022172927A1 - Method of manufacturing member having solder bump, member having solder bump, and member for forming solder bump

Info

Publication number: WO2022172927A1
Application number: PCT/JP2022/004958
Authority: WO
Inventors: 純一欠畑; 邦彦赤井; 歩未葭葉; 芳則江尻
Original assignee: 昭和電工マテリアルズ株式会社
Priority date: 2021-02-10
Filing date: 2022-02-08
Publication date: 2022-08-18
Also published as: CN117099190A; TW202236440A; KR20230141813A; JPWO2022172927A1

Abstract

A method of manufacturing a member having a solder bump, comprising: a preparation step for preparing a base body with a plurality of recessed portions having roughness on a bottom surface; a disposition step for disposing solder particles in the recessed portions; and a pressing step for pressing the base body and a substrate having an electrode in a state in which the solder particles and the electrode face each other, bringing the solder particles and the electrode into contact with each other, and forming, on the electrode, a solder bump having a recess at least in part of a surface.

Description

Method for manufacturing member with solder bumps, member with solder bumps, and member for forming solder bumps

The present invention relates to a method for manufacturing a member with solder bumps, a member with solder bumps, and a member for forming solder bumps.

When a semiconductor chip is mounted on a circuit board by a flip-chip method, a process of transferring flux to the adhesive surface of solder bumps formed on the semiconductor chip with the circuit board and the vicinity thereof, and a semiconductor chip to which the flux is transferred. A method for mounting a semiconductor chip is known, characterized by comprising a step of flip-chip connecting to a circuit board (for example, Patent Document 1).

JP-A-1995-078847

In general, the solder bump surface has a smooth curved surface. Therefore, when using the technique disclosed in Patent Document 1, when the solder bumps are pressed against the electrodes on the circuit board, the flux may be pushed away from the surfaces of the solder bumps. If a sufficient amount of flux is not secured at the time of joining, it becomes difficult for the solder to wet and spread appropriately, and there is a possibility that there will be places where the electrodes are not properly connected.

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a method of manufacturing a member with solder bumps capable of realizing better connection between electrodes. Another object of the present invention is to provide a member with solder bumps and a member for forming solder bumps.

One aspect of the present invention includes a preparation step of preparing a base having a plurality of recesses having unevenness on the bottom surface, an arrangement step of arranging solder particles in the recesses, a base and a substrate having electrodes, and solder particles and electrodes. A method of manufacturing a member with solder bumps, comprising: .

In general, when performing solder bump bonding, liquid flux is applied to a wiring substrate having metal electrodes such as Au and Cu, and the solder bumps are pressed and bonded. From the viewpoints of shortening the process, reducing raw material costs, suppressing metal electrode corrosion, etc., it is desirable that the amount of flux to be applied is small. On the other hand, if the amount of flux is small, the flux may be pushed away from the surface of the solder bumps, or the flux may flow at the bonding temperature, making it impossible to secure the amount of flux required to bond the solder bumps to the electrodes. . Accordingly, the inventors conducted studies to allow the solder bumps themselves to capture more flux while reducing the amount of flux used, and completed the present invention. In the present invention, a solder bump forming member is produced using a substrate having an uneven bottom surface, and is pressed against an electrode to form a solder bump. A solder bump having a depression in at least a portion of the electrode can be formed on the electrode. In the member with solder bumps obtained in this way, the depressions on the surfaces of the solder bumps capture the flux when the electrodes are connected, so that the electrodes can be connected more satisfactorily. This makes it possible to fabricate a connection structure that achieves both excellent insulation reliability and conduction reliability even with a very small amount of flux.

The solder particles may be heated in the pressing step in one aspect of the method for manufacturing a member with solder bumps.

One aspect of the method for manufacturing a member with solder bumps may further include a reducing step of exposing the solder particles to a reducing atmosphere before the arranging step.

One aspect of the method for manufacturing a member with solder bumps may further include a reducing step of exposing the solder particles to a reducing atmosphere after the arranging step and before the pressing step.

In the pressing step in one aspect of the method for manufacturing a member with solder bumps, the solder particles may be heated under a reducing atmosphere.

One aspect of the method for manufacturing a member with solder bumps may further include a removing step of removing the base from the substrate after the pressing step.

One aspect of the method for manufacturing a member with solder bumps may further include a cleaning step for removing solder particles that are not bonded to the electrodes after the removing step.

One aspect of the present invention relates to a member with solder bumps, which includes a substrate having electrodes, solder bumps on the electrodes, and recesses formed in at least part of the surfaces of the solder bumps.

In one aspect of the member with solder bumps, the recess depth of the solder bumps may be 25% or less of the solder bump height.

In one aspect of the member with solder bumps, adjacent solder bumps may be independent of each other.

In one aspect of the member with solder bumps, the height of the solder bumps may be smaller than the diameter of the solder bumps in the planar direction.

One aspect of the present invention relates to a solder bump-forming member comprising a substrate having a plurality of recesses having unevenness on the bottom surface and solder particles in the recesses.

In one aspect of the solder bump-forming member, the height difference between the recesses and protrusions of the unevenness may be 20% or less of the average particle diameter of the solder particles.

In one aspect of the solder bump forming member, the average particle diameter of the solder particles is 1 to 35 μm, and C.I. V. The value may be 20% or less.

According to the present invention, it is possible to provide a method of manufacturing a member with solder bumps capable of achieving better connection between electrodes. Further, according to the present invention, it is possible to provide a member with solder bumps obtained by the manufacturing method, and a member for forming solder bumps for obtaining the member with solder bumps.

FIG. 1 is a cross-sectional view schematically showing a solder bump forming member according to one embodiment. FIG. 2 is a cross-sectional view schematically showing an example of a substrate. 3(a) is a plan view schematically showing an example of the substrate 60, and FIG. 3(b) is a cross-sectional view taken along line Ib--Ib of FIG. 3(a). FIGS. 4A to 4D are cross-sectional views schematically showing examples of cross-sectional shapes of concave portions of the substrate. FIG. 5 is a cross-sectional view schematically showing a state in which solder particles 111 are accommodated in concave portions 62 of substrate 60. As shown in FIG. 6(a) and 6(b) are cross-sectional views schematically showing an example of the manufacturing process of a member with solder bumps. 7A and 7B are cross-sectional views schematically showing an example of the manufacturing process of the connection structure. 8 is a SEM photograph of the solder bump forming member obtained in Production Example 1. FIG.

The present embodiment will be described below. This embodiment is not limited to the following aspects. The materials exemplified below may be used singly or in combination of two or more unless otherwise specified. The content of each component in the composition means the total amount of the plurality of substances present in the composition unless otherwise specified when there are multiple substances corresponding to each component in the composition. A numerical range indicated using "-" indicates a range including the numerical values before and after "-" as the minimum and maximum values, respectively. In the numerical ranges described stepwise in this specification, the upper limit value or lower limit value of the numerical range at one step may be replaced with the upper limit value or lower limit value of the numerical range at another step. In the numerical ranges described herein, the upper or lower limits of the numerical ranges may be replaced with the values shown in the examples.

<Member for solder bump formation>
FIG. 1 is a cross-sectional view schematically showing a solder bump forming member according to one embodiment. The solder bump forming member 10 includes a substrate 60 having a plurality of recesses having unevenness on the bottom surface, and solder particles 1 in the recesses 62 . A concave portion having an uneven bottom surface can also be referred to as a concave portion having a protrusion on the bottom surface. In FIG. 1, the top of the protrusion is schematically shown as a curved surface, and the number of protrusions is drawn uniformly in each recess. It may be more pointed and the number of protrusions may differ between recesses. In a predetermined longitudinal section of the solder bump forming member 10, one solder particle 1 is arranged in a horizontal direction (horizontal direction in FIG. 1) while being separated from an adjacent solder particle 1. The solder particles 1 may be in contact with the side surfaces and/or the bottom surface within the recesses 62 . The solder bump forming member may be film-shaped (solder bump forming film), sheet-shaped (solder bump forming sheet), or substrate-shaped (solder bump forming substrate).

In the solder bump forming member 10, a part of the solder particles 1 may or may not protrude from the recess.

When part of the solder particles 1 protrude from the recesses, it means that at least the tops of the solder particles 1 protrude from the recesses 62 of the solder bump forming member 10 (protrude from the main surface of the substrate 60). can. In a cross-sectional view perpendicular to the main surface of the solder bump forming member ₁₀ , when the depth of the concave portion 62 is H1 and the height from the substrate surface to the _top of the solder particle ₁ is H2, H1 < (H ₁ + H ₂ ), ie 0<H ₂ . The height H2 of the solder particles ₁ is the height from the surface of the substrate 60 to the apex of the solder particles 1 in a cross-sectional view. The depth H1 of the recess 62 is measured based on the _average height of the protrusions present on the bottom surface. In FIG. 1, the height of the projections is schematically drawn to be constant, so H1 is the depth from the _top of the projection to the surface of the substrate. However, the height of the protrusion does not need to be constant (it is not necessary for the _bottom of the recess to have uniform unevenness). Measured relative to a value. Although the degree of protrusion of solder particles _{1 is not particularly limited, the upper limit of the ratio of H2 to H1 (H 2 /H 1} ₎ _can be set to 2.00 from the viewpoint of suppressing falling off of solder particles ₁ .

The depth H ₁ of the recesses 62 and the height H ₂ from the base surface to the top of the solder particles 1 can be measured with a laser microscope.

(solder particles)
Solder particles 1 may have an average particle size of 1 to 35 μm. The average particle size of the solder particles 1 may be 30 μm or less, 25 μm or less, 20 μm or less, or 15 μm or less. Moreover, the average particle diameter of the solder particles 1 may be 2 μm or more, 3 μm or more, or 5 μm or more.

The average particle size of solder particles 1 can be measured using various methods according to the size. For example, methods such as dynamic light scattering method, laser diffraction method, centrifugal sedimentation method, electrical detection band method, and resonance mass measurement method can be used. In addition, methods for measuring particle size from images obtained by optical microscopy, electron microscopy, etc. are available. Specific devices include flow-type particle image analyzers, microtracks, coulter counters, and the like. The average particle diameter of the solder particles 1 is the projected area circle equivalent diameter (the diameter of a circle having an area equal to the projected area of the particles) when the solder particles 1 are observed from the direction perpendicular to the main surface of the solder bump forming member 10. diameter).

The height difference (height of protrusions) between the recesses and protrusions of the unevenness on the bottom surface may be 20% or less of the average particle size of the solder particles 1 . From the viewpoint of maintaining the shape of the solder bumps, the height difference between the recesses and protrusions of the unevenness may be 20% or less, 15% or less, or 10% or less of the average particle size of the solder particles 1 . The lower limit of the height difference between the recesses and protrusions of the unevenness can be, for example, 2% or more of the average particle diameter of the solder particles 1 . From this point of view, for example, when the solder particle size is 4 μm, the height difference between the concave portion and the convex portion of the unevenness may be 0.8 μm or less, 0.6 μm or less, or even 0.4 μm or less. good. The lower limit of the height difference between the concave portion and the convex portion of the unevenness can be, for example, 0.08 μm or more.

The difference in height (height of protrusion) between the concave portion and the convex portion of the unevenness on the bottom surface can be measured with a laser microscope.
The height difference (height of the projection) between the recesses and projections of the unevenness on the bottom surface is calculated as the average value of the height differences (height of the projection) present on the bottom surface. In this specification, the height difference (height of protrusion) between the recesses and protrusions of the unevenness formed on the bottom surface of the recess of the solder bump forming member 10 is calculated as follows. For 100 concave portions of an arbitrary solder bump forming member 10, the protrusion height is measured by observation using a laser microscope, and the protrusion height for each concave portion is calculated. Further, the average value of the protrusion heights of the 100 recesses is calculated and used as the height difference (protrusion height) between the recesses and protrusions on the bottom surface of the recesses of the solder bump forming member 10 .

"C. of Solder Particle 1" V. The value may be 20% or less. C. of solder particle 1; V. The value may be 20% or less, 10% or less, or 7% or less from the viewpoint of achieving better electrical conductivity reliability and insulation reliability. Also, the C.I. V. The lower limit of the value is not particularly limited. For example, it may be 1% or more, or 2% or more.

"C. of Solder Particle 1" V. The value is calculated by dividing the standard deviation of the particle size measured by the method described above by the average particle size and multiplying by 100.

The solder particles 1 may contain tin or tin alloys. As the tin alloy, for example, In—Sn alloy, In—Sn—Ag alloy, Sn—Au alloy, Sn—Bi alloy, Sn—Bi—Ag alloy, Sn—Ag—Cu alloy, Sn—Cu alloy, etc. are used. be able to. Specific examples of these tin alloys include the following examples.
・In-Sn (In52% by mass, Bi48% by mass, melting point 118° C.)
・In-Sn-Ag (In20% by mass, Sn77.2% by mass, Ag2.8% by mass, melting point 175° C.)
・Sn-Bi (43% by mass of Sn, 57% by mass of Bi, melting point 138°C)
・Sn-Bi-Ag (Sn 42% by mass, Bi 57% by mass, Ag 1% by mass, melting point 139°C) ・Sn-Ag-Cu (Sn 96.5% by mass, Ag 3% by mass, Cu 0.5% by mass, melting point 217°C)
・Sn-Cu (Sn 99.3% by mass, Cu 0.7% by mass, melting point 227°C)
・Sn-Au (Sn21.0% by mass, Au79.0% by mass, melting point 278°C)

The solder particles may contain indium or an indium alloy. As the indium alloy, for example, an In--Bi alloy, an In--Ag alloy, or the like can be used. Specific examples of these indium alloys include the following examples.
・In-Bi (In66.3% by mass, Bi33.7% by mass, melting point 72° C.)
・ In-Bi (In 33.0% by mass, Bi 67.0% by mass, melting point 109 ° C.)
・In-Ag (97.0% by mass of In, 3.0% by mass of Ag, melting point 145°C)

The above-mentioned tin alloy or indium alloy can be selected according to the use of the solder particles 1 (temperature during connection). For example, when the solder particles 1 are used for fusion bonding at a low temperature, an In--Sn alloy or a Sn--Bi alloy may be used. In this case, fusion bonding can be performed at 150.degree. When a material with a high melting point such as Sn--Ag--Cu alloy or Sn--Cu alloy is used, high reliability can be maintained even after being left at a high temperature.

The solder particles 1 may contain one or more selected from Ag, Cu, Ni, Bi, Zn, Pd, Pb, Au, P and B. Among these elements, Ag or Cu may be included from the following viewpoints. That is, since the solder particles 1 contain Ag or Cu, the melting point of the solder particles 1 can be lowered to about 220° C., and the bonding strength with the electrodes is further improved, so that better conduction reliability is achieved. easier to obtain.

The Cu content of the solder particles 1 is, for example, 0.05 to 10% by mass, and may be 0.1 to 5% by mass or 0.2 to 3% by mass. When the Cu content is 0.05% by mass or more, it becomes easier to achieve better solder connection reliability. Further, when the Cu content is 10% by mass or less, the solder particles 1 tend to have a low melting point and excellent wettability, and as a result, the connection reliability of the joints by the solder particles 1 tends to be good.

The Ag content of the solder particles 1 is, for example, 0.05 to 10% by mass, and may be 0.1 to 5% by mass or 0.2 to 3% by mass. When the Ag content is 0.05% by mass or more, it becomes easier to achieve better solder connection reliability. Further, when the Ag content is 10% by mass or less, the solder particles 1 tend to have a low melting point and excellent wettability, and as a result, the connection reliability of the joints by the solder particles 1 tends to be good.

(substrate)
Examples of materials that can be used to form the substrate 60 include inorganic materials such as silicon, various ceramics, glass, metals such as stainless steel, and organic materials such as various resins. Among these, the substrate 60 may be made of a heat-resistant material that does not deteriorate at the melting temperature of the solder particles. Also, the substrate 60 may be made of a heat-resistant material that does not deform at the temperature at which the solder particles are melted. Also, the substrate 60 may be made of a material that does not change by alloying with or reacting with the material forming the solder particles. The recesses 62 of the substrate 60 can be formed by known methods such as cutting, photolithography, imprinting, and the like. In particular, when imprinting is used, it is possible to form the concave portion 62 with an accurate size in a short process.

The surface of the substrate 60 may have a coating layer. From the viewpoint of widening the selectivity of materials that can be used for the substrate 60, the coating layer may be made of a material that is difficult or not to be alloyed with the material that constitutes the solder particles, and that is a material that is heat-resistant and does not change in quality at the melting temperature of the solder particles. may be . An inorganic material or an organic material can be used as the coating layer. As the coating layer, inorganic substances having a strong oxide layer on the surface such as aluminum and chromium, oxides such as titanium oxide, nitrides such as boron nitride, diamond-like carbon (DLC), diamond, carbon materials such as graphite, Highly heat-resistant resin such as fluororesin and polyimide can be used. The coating layer may play a role in adjusting wettability with solder. By providing a coating layer on the surface of the substrate 60, wettability with solder can be appropriately adjusted according to the purpose of use.

Lamination, solution dipping, coating, painting, impregnation, sputtering, plating, etc. can be used as methods for forming the coating layer.

From the viewpoint of facilitating the setting of conditions for the transfer process, the material of the substrate 60 may be a material having physical properties similar to or the same as the electrodes for transferring the solder particles and the substrate on which the electrodes are formed. For example, if the materials have similar coefficients of thermal expansion (CTE) or are of the same material, misalignment is less likely to occur when solder particles are transferred.

An alignment mark may be provided on the substrate 60 . Alignment marks may also be present on the substrate side having the electrodes. Alignment marks should be readable by a camera. When transferring the solder particles onto the electrodes, a camera mounted on an alignable device reads the alignment marks on the substrate 60 and the alignment marks of the substrate having the electrodes, and determines the positions of the recesses 62 having the solder particles and the solder. It becomes possible to accurately grasp the position of the electrode for transferring the particles. By providing alignment marks for the substrate having the substrate 60 and the electrodes, the solder particles can be transferred onto the electrodes with high positional accuracy.

It suffices that there be one or more alignment marks on the substrate 60 . If there are two or more alignment marks, the positional accuracy will be high.

A specific configuration of the base 60 will be described below.

(Organic material single layer)
Substrate 60 may be composed of an organic material. The organic material may be a polymeric material such as a thermoplastic material, a thermosetting material, a photocurable material, or the like. By using an organic material, the range of selection of physical properties is widened, so that it is easy to form the substrate 60 according to the purpose. For example, if it is an organic material, the substrate 60 (including the concave portion 62) can be easily bent or stretched. Various techniques can be used to form the recesses 62 as long as they are organic materials. Imprinting, photolithography, cutting, laser processing, and the like can be used as a method for forming the concave portion 62 . According to the imprint method, a desired shape can be formed on the surface by pressing a mold having a desired shape against the substrate 60 made of an organic material. A concave portion 62 having a desired pattern can be formed by forming a convex pattern on a mold and pressing it against a substrate 60 made of an organic material. When a photocurable resin is used to form the concave portions 62, the substrate 60 having the concave portions 62 can be formed by applying the photocurable resin to a mold, exposing the mold, and then peeling off the mold. Moreover, in the case of cutting, the concave portion 62 can be formed with a drill or the like.

(Organic material, multiple layers)
The substrate may be composed of multiple organic materials. The substrate may have multiple layers, and the multiple layers may be composed of different organic materials. The organic material may be a polymeric material such as a thermoplastic material, a thermosetting material, a photocurable material, or the like. The substrate may have two layers made of an organic material, and the concave portion may be formed in the organic material layer on one side. By forming multiple layers, it is possible to select the material for each layer by dividing the function, such as selecting a material having suitable wettability with solder for the material of the recesses that come into contact with the solder. For example, FIG. 2 is a cross-sectional view schematically showing an example of a substrate. Substrate 600 comprises base layer 601 and recess layer 602 . The base layer 601 is a layer that supports the concave layer 602, and the concave layer 602 is a layer in which the concave portions 62 are formed by processing. A resin material excellent in heat resistance and dimensional stability can be used for the base layer 601 , and a material excellent in workability of the recesses 62 can be selected for the recess layer 602 . For example, a thermoplastic resin such as polyethylene terephthalate or polyimide can be used for the base layer 601, and a thermosetting resin capable of forming the recesses 62 by imprint molding can be used for the recess layer 602. FIG. For example, by sandwiching a thermosetting resin between polyethylene terephthalate and an imprint mold and applying heat and pressure, a substrate 600 (including recesses 62) having excellent flatness can be obtained. In addition, when the concave portion 62 is formed using a photo-curing material, a material with high light transmittance may be used for the base layer 601 . Examples of materials with high light transmittance include polyethylene terephthalate, transparent (colorless type) polyimide, and polyamide. When forming the concave portions 62 using a photocurable material, for example, an appropriate amount of the photocurable material is applied to the surface of the imprint mold, a polyethylene terephthalate film is placed thereon, and the polyethylene terephthalate film is pressed with a roller from the polyethylene terephthalate side. Ultraviolet light is irradiated while pressing. After the photocurable material is cured, the imprint mold is peeled off to obtain a substrate 600 having a layer of polyethylene terephthalate and a layer of photocurable material and having concave portions 62 formed of the photocurable material. be able to. The material composition of the inner walls and bottom surface of the recess 62 can vary. For example, the inner wall and the bottom surface of the recess 62 can be made of the same resin material. Also, the inner wall and the bottom surface of the recess 62 can be made of different resin materials (for example, a thermosetting material and a thermoplastic material).

A photosensitive material may be used as the organic material. Photosensitive materials include positive photosensitive materials and negative photosensitive materials. For example, the concave portions 62 can be formed by forming a photosensitive material with a uniform thickness on the surface of a thermoplastic polyethylene terephthalate film, and performing exposure and development (photolithography). The method using exposure and development is widely used in the manufacture of semiconductors, wiring boards, etc., and is a highly versatile method. As the exposure method, in addition to exposure using a mask, it is also possible to use a direct drawing method such as direct laser exposure.

By making the material of the base layer 601 thicker than the material forming the concave layer 602 , the physical properties of the entire substrate 600 can be dominated by the material properties of the base layer 601 . As a result, for example, even if the material forming the concave layer 602 has a weak point, the weak point can be compensated for by the material of the base layer 601 . For example, even if the material of the concave layer 602 is a material that easily shrinks due to heat, a material that does not easily shrink due to heat is selected as the material of the base layer 601, and the thickness of the base layer 601 is made thicker than the thickness of the material forming the concave layer 602. By doing so, deformation during heating can be suppressed.

A combination of a resin material with excellent heat resistance or dimensional stability and a material with little component elution at the melting temperature of solder particles, or a resin material with excellent heat resistance or dimensional stability and wettability with solder is appropriate. The organic material can be appropriately selected according to the purpose, such as combination with other materials.

As described above, the substrate may be the substrate 600 composed of the base layer 601 and the concave layer 602 . For example, by using a photosensitive material for the recess layer 602, the recesses 62 can be formed by photolithography. By using a photocurable or thermosetting material, a thermoplastic material, or the like for the recessed portion layer 602, the recessed portions 62 can be formed by an imprint method. By changing the thickness of the base layer 601, it is possible to adjust the properties of the entire base, so that a base having desired properties can be manufactured.

(Inorganic material single layer (opaque))
The substrate 60 may be composed of an inorganic material. For example, silicon (silicon wafer), stainless steel, aluminum, etc. can be used as the inorganic material from the viewpoint of easily controlling the elution of components and the generation of foreign matter to a low level. When these materials are used in a semiconductor mounting process or the like, contamination countermeasures are easy, and high yields and stable production can be easily achieved. For example, when transferring solder particles formed in the recesses 62 to electrodes on a silicon wafer, if the base 60 is made of a silicon wafer, the base and the substrate should have similar CTEs or be made of the same material. become. As a result, misregistration, warping, etc. are less likely to occur, and transfer to an accurate position becomes possible. As a method for forming the concave portion 62, processing by laser, cutting, etc., dry etching or wet etching, electron beam drawing (for example, FIB processing), or the like can be used. Dry etching is widely used in the production of semiconductors, MEMS, etc., and can process inorganic materials with high precision on the micron order to the nano order.

(Inorganic material single layer (transparent))
Glass, quartz, sapphire, or the like can be used as the substrate 60 . Since these materials are transparent, it is easy to align when transferring the solder particles in the recesses 62 to another substrate on which electrodes are formed. As a method for forming the concave portion 62, processing by laser, cutting, etc., dry etching or wet etching, electron beam drawing (for example, FIB processing), or the like can be used.

The advantage of using inorganic materials is that they have excellent dimensional stability compared to organic materials. When transferring the solder particles in the recesses 62 onto the electrodes, the transfer can be performed with high positional accuracy. For example, when transferring solder particles to multiple electrodes with a size and pitch on the order of micrometers, if an inorganic material with excellent dimensional stability is used, the solder particles can be transferred to the same position on any electrode.

(Organic-inorganic composite materials)
The substrate may be composed of multiple materials. The substrate may have multiple layers, and the multiple layers may be made of different materials. As the organic-inorganic composite material, for example, a combination of an inorganic material and an organic material can be used. The combination of the inorganic material and the organic material facilitates achieving both dimensional stability and workability of the recess 62 . As a substrate having a combination of an inorganic material and an organic material, for example, there is a substrate having a base layer 601 made of an inorganic material such as silicon, various ceramics, glass, metal such as stainless steel, and a concave layer 602 made of an organic material. mentioned. Such a substrate can be obtained, for example, by forming a film of a photosensitive material on the surface of a silicon wafer and forming recesses by exposure and development. The inner wall and bottom surface of the recess 62 may be composed of a photosensitive material, or the inner wall of the recess 62 may be composed of a photosensitive material and the bottom surface of the recess 62 may be composed of a silicon wafer. The configuration of the recesses 62 can be appropriately selected according to purposes such as wettability with the solder particles in the recesses 62 and ease of transfer to the electrodes. When the inner wall and the bottom surface of the concave portion 62 are made of a photosensitive material, a film of the photosensitive material is formed on the surface of the silicon wafer and hardened to form a layer of the photosensitive material on the surface of the silicon wafer. A method of providing the concave portion 62 by forming a film of a photosensitive material again and performing exposure and development can be used. In this case, the photosensitive material on the surface side of the silicon wafer and the photosensitive material provided on the outermost layer may have different compositions. The photosensitive material can be appropriately selected in consideration of the wettability of the solder particles, staining property, and the like. When the solder particles formed in the recesses 62 are transferred onto the electrodes, the surface of the outermost photosensitive material layer may come into contact with the electrodes or the surface of the substrate having the electrodes. Therefore, it is possible to appropriately select a photosensitive material that does not damage the electrodes and the substrate or contaminate the electrodes and the substrate. The photosensitive material may be a material that prevents elution of uncured components and contamination with halogen-based materials, silicone-based materials, and the like. The photosensitive material may be a material that is highly resistant to a reducing atmosphere, flux, etc. when transferring the solder particles to the electrode. For example, the photosensitive material may be a material that is resistant to reducing atmospheres such as formic acid, hydrogen, hydrogen radicals, and the like. Additionally, the photosensitive material may be a material that is highly resistant to the temperatures at which the solder particles are transferred to the electrodes. Specifically, the photosensitive material may be a material that is resistant to temperatures between 100°C and 300°C. Since the melting point of the solder particles differs depending on the constituent material, the heat-resistant temperature of the photosensitive material can also be selected according to the solder material to be used. When using tin-silver-copper-based solder (e.g. SAC305 (melting point: 219°C)), which is a lead-free solder widely used in electronic devices, heat resistance of 220°C or higher, especially 260°C or higher used in the reflow process of heat-resistant materials can be used. When using tin-bismuth solder (eg SnBi58 (melting point 139°C)), a material with heat resistance of 140°C or higher can be used, especially if it is a material with heat resistance of 160°C or higher, industrially will be more likely to be used. When indium solder (melting point 159° C.) is used, a material having heat resistance of 170° C. or more can be used. When indium-tin solder (eg, melting point 120° C.) is used, a material having heat resistance of 130° C. or higher can be used.

Other substrates include substrates having recesses 62 formed of a thermosetting or thermoplastic resin on a stainless steel plate. The substrate can be obtained by sandwiching a thermosetting material (resin) between a stainless steel plate and an imprint mold, applying pressure and heating, and then peeling off the imprint mold. Other substrates include substrates having recesses 62 formed of a photocurable material on a glass plate. The substrate can be obtained by a method of applying a photocurable material onto a glass plate, exposing to light while pressing an imprint mold to cure the photocurable material, and peeling off the imprint mold. When forming the concave portion 62 using an imprint mold, the material composition of the inner wall and the bottom surface of the concave portion 62 can be changed depending on the pressurizing conditions. For example, if the pressurizing condition is relaxed, the inner wall and the bottom surface of the recess 62 can be made of the same resin material. On the other hand, when the pressure condition is increased, the inner wall of the recess 62 can be made of a resin material, and the bottom surface can be made of an inorganic material.

As the material of the base layer 601, a composite material containing glass fiber, filler, etc., and a resin component can also be used. Composite materials include copper-clad laminates for wiring boards and the like. The concave portions 62 can be formed as described above by applying a photosensitive material, a thermosetting resin, a photosetting resin, or the like to the surface of the copper-clad laminate. Although the copper-clad laminate mainly contains a large amount of resin material, it can be combined with glass fiber, various fillers, and the like to achieve a low CTE, so that the aforementioned dimensional stability can be ensured. When the electrodes are formed on the copper-clad laminate, forming the recesses 62 on the same copper-clad laminate makes the CTEs of the base and the substrate the same or close to each other. This facilitates alignment during transfer of the solder particles in the recesses 62, and prevents misalignment.

A package sealing material can also be used as the material of the concave layer 602 . Any of solid, liquid and film forms can be used as the sealing material. The concave portion 62 can be formed by laminating the sealing material in a thin layer on glass, a silicon wafer, or the like, and applying pressure and heat using an imprint mold.

The uneven shape on the bottom surface of the concave portion of the substrate is obtained by pressing a mold having a desired shape (mold: for example, a mold in which a plurality of convex portions having unevenness at the tip are arranged) against a substrate made of an organic material, as in imprint technology. Alternatively, the mold may be formed by applying a photocurable resin to the mold, exposing the mold, and then peeling off the mold.

The concave-convex shape may be formed by post-processing a substrate having a flat concave bottom surface.
For example, a wet etching process, a dry etching process, or the like can be used to form an uneven shape on the bottom surface of the recess. Also, a blasting method in which an abrasive containing hard fine particles is sprayed at high pressure can be used to form an uneven shape on the bottom surface of the recess.

An uneven shape may be formed in the concave portion other than the bottom surface. The uneven shape other than the bottom surface may be of a degree that can maintain the shape of the recess and does not hinder the formation of bumps on the electrodes.

<Members with solder bumps>
The member with solder bumps includes a substrate having electrodes, solder bumps on the electrodes, and depressions formed in at least a portion of the surfaces of the solder bumps. The member with the solder bumps may have a depression on at least the top of the surface of the solder bumps so that the amount of flux necessary for bonding the solder bumps to the electrodes can be easily secured. The member with solder bumps can be called an electrode substrate with solder bumps.

Specific examples of metal electrodes of members with solder bumps include copper, copper/nickel, copper/nickel/gold, copper/nickel/palladium, copper/nickel/palladium/gold, copper/nickel/gold, copper/palladium, copper /palladium/gold, copper/tin, copper/silver, and indium tin oxide. The electrodes can be formed by electroless or electrolytic plating or sputtering or etching of metal foils.

The depth of the depression on the solder bump surface may be 25% or less of the solder bump height. From the viewpoint of maintaining the shape of the solder bump, the depth of the recess on the surface of the solder bump may be 25% or less, 15% or less, or 10% or less of the solder bump height. The lower limit of the recess depth of the solder bump surface can be, for example, 2% or more of the solder bump height. From this point of view, for example, the recess depth of a solder bump having a bump height of 3.6 μm may be 0.9 μm or less, 0.54 μm or less, or 0.36 μm or less. The lower limit of the depth of the depression on the surface of the solder bump can be, for example, 0.07 μm or more. The solder bump height is the height from the electrode surface to the top of the solder bump when the solder bump does not have a depression (assumed to be spherical).

The depth of the depression on the surface of the solder bump and the height of the solder bump can be measured with a laser microscope.

Adjacent solder bumps formed on the member with solder bumps may be independent of each other.

From the viewpoint of connectivity when a connection structure is obtained using a member with bumps, the bump pitch may be the same as the pitch of the substrate recesses. It does not have to be the same as the pitch.

A part of the solder bump formed on the electrode may be joined to the electrode (metal electrode).
Depending on the material combination, the electrode may have an alloy layer on its surface (at the solder bump-electrode interface).

The diameter of the alloy layer may be larger than the diameter of the solder bump in the planar direction. As long as the solder bumps are independent, the alloy layers formed under the solder bumps may be in contact with each other.

In the member with solder bumps, the height of the solder bumps may be smaller than the diameter of the solder bumps in the planar direction (electrode planar direction). The diameter of the solder bump in the plane direction is the maximum diameter (maximum width) of the solder bump in the plane direction. The height of the solder bump is the height from the electrode surface to the top of the solder bump. Since the solder bumps are flat in this way, the contact area between the electrodes on the members facing each other and the solder bumps increases when used for mounting, and more stable connection becomes possible.

<Method for manufacturing members with solder bumps>
A method for manufacturing a member with solder bumps includes a preparation step of preparing a substrate having a plurality of recesses having unevenness on the bottom surface, an arrangement step of arranging solder particles in the recesses, a substrate having an electrode, and a substrate having electrodes. and a pressing step of pressing the electrodes while facing each other to bring the solder particles and the electrodes into contact with each other, thereby forming solder bumps having depressions on at least part of the surfaces of the electrodes.

A solder bump forming member is obtained through the preparation process and the placement process. A method of manufacturing a solder bump forming member can be said to include a preparation step of preparing a substrate having a plurality of recesses having unevenness on the bottom surface, and an arrangement step of arranging solder particles in the recesses.

A method of manufacturing the solder bump forming member 10 will be described with reference to FIGS.
According to the procedure described below, the solder particles can be arranged in the recesses by fusing the solder particles in the recesses of the prepared substrate.

A substrate 60 for containing solder particles and solder particles are prepared. 3(a) is a plan view schematically showing an example of the substrate 60, and FIG. 3(b) is a cross-sectional view taken along line Ib--Ib of FIG. 3(a). A substrate 60 shown in FIG. 3A has a plurality of recesses 62 . The plurality of recesses 62 may be regularly arranged in a predetermined pattern. The positions and number of the recesses 62 may be set according to the shape, size and pattern of the electrodes to be connected.

The distance L between adjacent recesses is not particularly limited, but it can be 0.1 times or more the average particle diameter of the solder particles to be accommodated, and may be 0.2 times or more. The distance between recesses is the edge-to-edge distance of the recess openings, not the center-to-center distance of the recesses.

The concave portion 62 of the base 60 may be tapered such that the opening area increases from the bottom surface 62a side of the concave portion 62 toward the surface 60a side of the base 60 . That is, as shown in FIGS. 3A and 3B, the width of the bottom surface 62a of the recess 62 (the width a in FIGS. 3A and 3B) is the opening of the surface 60a of the recess 62. (width b in FIGS. 3(a) and 3(b)). The size (width a, width b, volume, taper angle, depth, etc.) of the recess 62 may be set according to the desired size of the solder particles.

The shape of the concave portion 62 of the substrate 60 can be freely designed by lithography, machining, imprint technology, or the like. Since the size of solder particles 1 depends on the amount of solder particles 111 accommodated in recesses 62 , the size of solder particles 1 can be freely designed by designing recesses 62 .

The shape of the recess 62 may be a shape other than the shape shown in FIGS. 3(a) and 3(b). For example, the shape of the opening in the surface 60a of the recess 62 may be oval, triangular, quadrangular, polygonal, etc., other than circular as shown in FIG. 3(a).

The shape of the recess 62 in the cross section perpendicular to the surface 60a may be, for example, the shape shown in FIG. FIGS. 4A to 4D are cross-sectional views schematically showing examples of cross-sectional shapes of concave portions of the substrate. 4A to 4D, the width (width b) of the opening on the surface 60a of the recess 62 is the maximum width in the cross-sectional shape. As a result, the solder particles formed in the recesses 62 can be easily taken out, improving workability. In addition, since the width (width b) of the opening is the maximum width in the cross-sectional shape, when the solder particles 1 are transferred onto the electrode, the solder particles 1 easily escape from the recesses 62, improving the transfer rate. can be expected. Further, by appropriately adjusting the width (width b) of the opening, misalignment is less likely to occur when the solder particles 1 are transferred onto the electrodes, making it easier to form solder bumps at accurate positions.

The solder particles may contain particles having a particle diameter smaller than the width (width b) of the opening in the surface 60a of the recess 62, but may contain more particles having a particle diameter smaller than the width b. For example, the D10 particle size of the particle size distribution of the solder fine particles may be smaller than the width b, the D30 particle size of the particle size distribution may be smaller than the width b, and the D50 particle size of the particle size distribution may be smaller than the width b.

The particle size distribution of solder particles can be measured using various methods according to their size. For example, methods such as dynamic light scattering method, laser diffraction method, centrifugal sedimentation method, electrical detection band method, and resonance mass measurement method can be used. In addition, methods for measuring particle size from images obtained by optical microscopy, electron microscopy, etc. are available. Specific devices include flow-type particle image analyzers, microtracks, coulter counters, and the like.

C.I. of solder particles V. Although the value is not particularly limited, C.I. V. Value should be high. For example, C.I. V. The value may be above 20%, 25% or more, or 30% or more.

C.I. of solder particles V. The value is calculated by dividing the standard deviation of the particle size measured by the method described above by the average particle size (D50 particle size) and multiplying by 100.

However, even solder fine particles with a large variation in particle size distribution or an irregular shape can be used as a raw material if they can be accommodated in the recess 62 .

The solder particulates may contain tin or tin alloys. As the tin alloy, for example, In—Sn alloy, In—Sn—Ag alloy, Sn—Au alloy, Sn—Bi alloy, Sn—Bi—Ag alloy, Sn—Ag—Cu alloy, Sn—Cu alloy, etc. are used. be able to. Specific examples of these tin alloys include the following examples.
・In-Sn (In52% by mass, Bi48% by mass, melting point 118° C.)
・In-Sn-Ag (In20% by mass, Sn77.2% by mass, Ag2.8% by mass, melting point 175° C.)
・Sn-Bi (43% by mass of Sn, 57% by mass of Bi, melting point 138°C)
・Sn-Bi-Ag (Sn 42% by mass, Bi 57% by mass, Ag 1% by mass, melting point 139°C) ・Sn-Ag-Cu (Sn 96.5% by mass, Ag 3% by mass, Cu 0.5% by mass, melting point 217°C)
・Sn-Cu (Sn 99.3% by mass, Cu 0.7% by mass, melting point 227°C)
・Sn-Au (Sn21.0% by mass, Au79.0% by mass, melting point 278°C)

The tin alloy or indium alloy can be selected according to the application of the solder particles (temperature during use). For example, when it is desired to obtain solder particles for low-temperature fusion bonding, an In--Sn alloy or a Sn--Bi alloy may be used. In this case, solder particles that can be fusion-bonded at 150.degree. When a material with a high melting point such as Sn--Ag--Cu alloy or Sn--Cu alloy is used, it is possible to obtain solder particles capable of maintaining high reliability even after being left at high temperature.

The solder fine particles may contain one or more selected from Ag, Cu, Ni, Bi, Zn, Pd, Pb, Au, P and B. Among these elements, Ag or Cu may be included from the following viewpoints. That is, by including Ag or Cu in the solder fine particles, the melting point of the obtained solder particles can be lowered to about 220 ° C., and solder particles with excellent bonding strength with the electrodes can be obtained, resulting in better conduction reliability. The effect of obtaining sexuality is exhibited.

The Cu content of the solder fine particles is, for example, 0.05 to 10% by mass, and may be 0.1 to 5% by mass or 0.2 to 3% by mass. When the Cu content is 0.05% by mass or more, it becomes easier to obtain solder particles capable of achieving good solder connection reliability. Further, when the Cu content is 10% by mass or less, solder particles having a low melting point and excellent wettability are easily obtained, and as a result, the connection reliability of electrodes with solder bumps is likely to be improved.

The Ag content of the solder fine particles is, for example, 0.05 to 10% by mass, and may be 0.1 to 5% by mass or 0.2 to 3% by mass. When the Ag content is 0.05% by mass or more, it becomes easier to obtain solder particles capable of achieving good solder connection reliability. Further, when the Ag oil content is 10% by mass or less, solder particles having a low melting point and excellent wettability are easily obtained, and as a result, the connection reliability of electrodes with solder bumps is likely to be improved.

The solder fine particles prepared as described above are accommodated in each of the concave portions 62 of the substrate 60 . Here, all of the solder particles may be accommodated in the recesses 62 , or part of the solder particles (for example, solder particles smaller than the width b of the opening of the recesses 62 ) may be accommodated in the recesses 62 .

FIG. 5 is a cross-sectional view schematically showing a state in which the solder particles 111 are accommodated in the concave portions 62 of the substrate 60. As shown in FIG. As shown in FIG. 5 , a plurality of solder particles 111 are accommodated in each of the plurality of recesses 62 .

By adjusting the amount of the solder particles 111 accommodated in the recesses 62, the degree of projection of the solder particles 1 can be adjusted. The amount of solder particles 111 accommodated in recesses 62 may be, for example, 20% or more, 30% or more, 50% or more, or 60% or more of the volume of recesses 62 . This allows a part of the solder particles to protrude from the recesses 62 . In addition, variation in the amount contained is suppressed, making it easier to obtain solder particles with a smaller particle size distribution.

In general, solder materials have the property of forming a spherical shape due to their own surface tension when they are melted in an environment above their melting point.

The solder particles 111 accommodated in the concave portions 62 are united by a fusion process described below to form the solder particles 1 . The height of the obtained solder particles 1 is higher than the depth of the recesses 62 , and the solder particles 1 protrude from the recesses 62 . Therefore, if the diameter of solder particle 1 is greater than the depth of recess 62 , solder particle 1 protrudes from recess 62 . Since the diameter of the solder particles 1 can be adjusted by adjusting the shape of the recess 62 and the amount of the solder particles 111 accommodated in the recess 62, the degree of projection from the recess 62 can be adjusted accordingly.

When the solder particles 111 are melted by the fusion process described later, wetting and spreading occurs on the bottom and inner walls of the recesses 62 depending on the material of the recesses 62, and at least part of the solder particles 1 come into contact with the bottom and/or the inner walls of the recesses 62. part occurs. As a result, at least part of the solder particles 1 may have flat portions. The size of the plane portion differs depending on the combination of the surface material of the concave portion 62 and the composition of the solder forming the solder particles 111 . Accordingly, the shape of the solder particles 1 may be a true sphere, an ellipsoid, a flattened sphere, a shape partially having a flat portion, or the like. Inorganic substances such as glass and silicon, or organic substances such as plastics and resins, can be used as the substrate 60. Such materials generally tend to have low wettability with solder, and the solder particles 1 are nearly spherical. It tends to be spherical. Therefore, it is also possible to assume that the solder particles 1 are nearly spherical and approximate the height of the solder particles 1 to the diameter of the solder particles 1 . Since the diameter of the solder particles 1 can be calculated from the total volume of the solder particles 111 filled in the recesses 62, the amount of the solder particles 111 required for the solder particles 1 to protrude from the recesses 62 can be calculated.

All the solder particles 111 filled in the recesses 62 are melted and coalesced to form the solder particles 1. Assuming that the solder particles 1 are spherical, the solder particles 111 necessary for the solder particles 1 to protrude from the recesses 62 are can be quantified.

When the upper diameter (opening width b) of the recess 62 is L and the depth of the recess 62 is D, the aspect ratio of the recess is represented by L/D. At this time, the filling rate of the solder fine particles 111 into the concave portions 62 is 66 volume % or more when the aspect ratio is 1, 38 volume % or more when the aspect ratio is 0.75, and 17 volume % when the aspect ratio is 0.5. % or more, and may be 5 volume % or more when the aspect ratio is 0.25.

In order to suppress variations in the amount of storage, the average particle size, particle size, etc. of the solder fine particles 111 can be selected according to the size of the concave portion 62 and the ratio of the diameter to the depth (aspect ratio). For example, when the diameter of the concave portion 62 is 4 μm and the depth is 4 μm (the aspect ratio is 1), by using the solder fine particles 111 having an average particle size of 1 to 2 μm or less, the unevenness in filling amount of the concave portion 62 can be reduced. In addition, variations in the diameter of the obtained solder particles 1 are also suppressed, and variations in the amount (height) of protrusion from the concave portion 62 are easily suppressed. When the variation in the protrusion amount (height) from the recess 62 is suppressed, the contact between the solder particle 1 and the electrode is stabilized when the solder particle 1 is pressed against the electrode, and the solder bump formation variation is easily suppressed.

The method of accommodating the solder particles in the recesses 62 is not particularly limited. The storage method may be either dry or wet. For example, by placing solder particles on the substrate 60 and rubbing the surface 60a of the substrate 60 with a squeegee, it is possible to remove excess solder particles while accommodating sufficient solder particles in the recesses 62. If the width b of the opening of the recess 62 is larger than the depth of the recess 62 , solder particles may fly out of the opening of the recess 62 . Using a squeegee removes the solder particles protruding from the openings of the recesses 62 . Methods for removing excess solder particles include blowing compressed air, rubbing the surface 60a of the substrate 60 with a non-woven fabric or fiber bundles, and the like. Since these methods require less physical force than squeegees, they are easy to handle deformable solder fine particles. With these methods, the solder particles protruding from the opening of the recess 62 can be left in the recess.

Next, the solder particles 111 accommodated in the recesses 62 are fused (for example, by heating to 130 to 260° C.) to form the solder particles 1 in the recesses 62 partially protruding from the recesses 62 . The solder particles 111 accommodated in the recesses 62 are melted to be united and spherically shaped by surface tension. At this time, the molten solder may follow the bottom surface 62a at the contact portion with the bottom surface 62a of the recess 62, and the flat portion 11 may be formed on a part of the solder particle surface. Thus, the solder bump forming member 10 shown in FIG. 1 is obtained.

As a method of melting the solder particles 111 accommodated in the recesses 62, there is a method of heating the solder particles 111 to the melting point of solder or higher. The solder fine particles 111 may not melt even when heated at a temperature equal to or higher than the melting point due to the influence of the oxide film, or may not wet and spread, and may not be coalesced. Therefore, by exposing the solder particles 111 to a reducing atmosphere and removing the surface oxide film of the solder particles 111, the solder particles 111 are heated to a temperature equal to or higher than the melting point of the solder particles 111, thereby melting the solder particles 111, wetting and spreading, and coalescing. can be unified.
The melting of the solder particles 111 may be performed in a reducing atmosphere. By heating the solder particles 111 to a temperature higher than the melting point of the solder particles 111 and creating a reducing atmosphere, the oxide film on the surface of the solder particles 111 is reduced, and the solder particles 111 melt, spread, and coalesce efficiently. easier to proceed.

The method of creating a reducing atmosphere is not particularly limited as long as the above effects can be obtained, and examples include methods using hydrogen gas, hydrogen radicals, formic acid gas, and the like. For example, by using a hydrogen reduction furnace, a hydrogen radical reduction furnace, a formic acid reduction furnace, or a conveyor furnace or continuous furnace thereof, the solder particles 111 can be melted in a reducing atmosphere. These devices may be equipped with a heating device, a chamber filled with an inert gas (nitrogen, argon, etc.), a mechanism for evacuating the chamber, etc. in the furnace, which makes it easier to control the reducing gas. Become. Further, if the chamber can be evacuated, voids can be removed by reducing the pressure after the solder particles 111 are melted and coalesced, and solder particles 1 with even better connection stability can be obtained.

Profiles such as reduction of solder particles 111, melting conditions, temperature, furnace atmosphere adjustment, etc. may be appropriately set in consideration of the melting point, particle size, recess size, material of substrate 60, etc. of solder particles 111. FIG. For example, solder particles can be obtained as follows.
The substrate 60 with the solder particles 111 filled in the recesses is inserted into the furnace, and the furnace is evacuated.
A reducing gas is introduced to fill the inside of the furnace with the reducing gas, thereby removing the surface oxide film of the solder fine particles 111 .
The reducing gas is removed by vacuuming.
The solder particles 111 are heated to a melting point or higher to melt and coalesce, forming solder particles in the recesses 62 .
After filling nitrogen gas, the temperature inside the furnace is returned to room temperature, and solder particles 1 can be obtained.

Alternatively, for example, solder particles may be obtained as follows. In the following method, heating the solder fine particles in a reducing atmosphere increases the reducing power and has the advantage of facilitating the removal of the surface oxide film of the solder fine particles.
The substrate 60 with the solder particles 111 filled in the recesses is inserted into the furnace, and the furnace is evacuated.
A reducing gas is introduced to fill the furnace with the reducing gas.
Solder particles 111 are heated by a furnace heating heater to remove surface oxide films of solder particles 111 .
The reducing gas is removed by vacuuming.
The solder particles 111 are heated to a melting point or higher to melt and coalesce, forming solder particles in the recesses 62 .
After filling nitrogen gas, the temperature inside the furnace is returned to room temperature, and solder particles 1 can be obtained.

Further, for example, solder particles may be obtained as follows. The following method has the advantage that the temperature in the furnace can be adjusted in a short time because it is only necessary to adjust the temperature in the furnace once.
The substrate 60 with the solder particles 111 filled in the recesses is inserted into the furnace, and the furnace is evacuated.
A reducing gas is introduced to fill the furnace with the reducing gas.
The solder particles 111 are heated to a temperature higher than the melting point of the solder particles 111 by the heater in the furnace, and the oxide films on the surfaces of the solder particles 111 are removed by reduction. At the same time, the solder particles are melted and coalesced to form solder particles within the recesses 62 .
Evacuation removes the reducing gas and further reduces voids in the solder particles.
After filling nitrogen gas, the temperature inside the furnace is returned to room temperature, and solder particles 1 can be obtained.

After forming the solder particles in the recesses 62 described above, a step of removing the unremoved surface oxide film by making the inside of the furnace a reducing atmosphere again may be added. As a result, it is possible to reduce residues such as solder fine particles remaining without being fused and a portion of the oxide film remaining without being fused.

When an atmospheric pressure conveyor furnace is used, the substrate 60 with the solder fine particles 111 filled in the recesses is placed on the conveyer and continuously passed through a plurality of zones to obtain the solder particles 1 . For example, solder particles can be obtained as follows.
The substrate 60 with the solder particles 111 filled in the recesses is placed on a conveyor set at a constant speed.
The solder particles 111 are passed through a zone filled with an inert gas such as nitrogen or argon at a temperature lower than the melting point of the solder particles 111 .
Solder particles 111 are passed through a zone in which a reducing gas such as formic acid gas having a temperature lower than the melting point of solder particles 111 exists to remove surface oxide films of solder particles 111 .
Solder particles 111 are melted and coalesced by passing through a zone filled with an inert gas such as nitrogen or argon at a temperature higher than the melting point of solder particles 111 .
Solder particles 1 can be obtained by passing through a cooling zone filled with an inert gas such as nitrogen or argon.

Solder particles may also be obtained in the following manner using an atmospheric pressure conveyor furnace.
The substrate 60 with the solder particles 111 filled in the recesses is placed on a conveyor set at a constant speed.
The solder particles 111 are passed through a zone filled with an inert gas such as nitrogen or argon at a temperature higher than the melting point of the solder particles 111 .
Solder particles 111 are passed through a zone in which reducing gas such as formic acid gas having a temperature higher than the melting point of solder particles 111 is present to remove surface oxide films of solder particles 111 . At the same time, the solder fine particles are melted and coalesced.
Solder particles 1 can be obtained by passing through a cooling zone filled with an inert gas such as nitrogen or argon.

Since the conveyor furnace is capable of processing at atmospheric pressure, it is also possible to continuously process film-like materials in a roll-to-roll manner. For example, solder particles can be fused as follows.
A continuous roll product of the substrate 60 in which the solder particles 111 are filled in the recesses is produced.
A roll unwinder is installed on the inlet side of the conveyor furnace, and a roll winder is installed on the outlet side of the conveyor furnace to transport the substrate 60 at a constant speed.
By passing through each zone in the conveyor furnace, the solder fine particles 111 filled in the recesses can be fused.

By the above method, solder particles 1 of uniform size can be formed regardless of the material and shape of solder particles 111 . For example, indium-based solder can be deposited by plating, but it is difficult to deposit in the form of particles, and it is soft and difficult to handle. However, in the above method, indium solder particles having a uniform particle size can be easily produced by using fine indium solder particles as raw materials. Further, the formed solder particles 1 can be handled while being accommodated in the concave portions 62 of the substrate 60 . Therefore, the solder particles 1 can be transported and stored without being deformed. Furthermore, the formed solder particles 1 are in a state of being accommodated in the recesses 62 of the substrate 60 . Therefore, the solder particles can be brought into contact with the electrodes without being deformed.

In the above method, the solder particles 111 are fused in the recesses 62 of the substrate 60, and the solder particles 1 are arranged in the recesses 62. A solder bump-forming member may be obtained by arranging separately prepared solder particles such as solder particles or solder particles having a desired particle size distribution by purchasing industrial products in the recesses of the substrate.

As a method for accommodating separately prepared solder particles in the concave portions of the substrate, for example, the above-described method of accommodating fine solder particles in the concave portions can be used.

Next, a process of preparing a member with solder bumps by pressing the solder bump forming member prepared as described above against an electrode will be described.

Specific examples of substrates (circuit members) having a plurality of electrodes on the surface include IC chips (semiconductor chips), resistor chips, capacitor chips, chip parts such as driver ICs, and rigid package substrates. Typically, these circuit members have multiple circuit electrodes. Other examples of the substrate having a plurality of electrodes on its surface include a wiring substrate such as a flexible tape substrate having metal wiring, a flexible printed wiring board, and a glass substrate on which indium tin oxide (ITO) is vapor-deposited.

Electrodes include copper, copper/nickel, copper/nickel/gold, copper/nickel/palladium, copper/nickel/palladium/gold, copper/nickel/gold, copper/palladium, copper/palladium/gold, copper/tin, Electrodes made of copper/silver, indium tin oxide, and the like can be mentioned. Electrodes can be formed by electroless plating, electrolytic plating, sputtering, etching of metal foil, or the like.

6(a) and 6(b) are cross-sectional views schematically showing an example of the manufacturing process of a member with solder bumps (electrode substrate with solder bumps). A substrate 60 shown in FIG. 6A is in a state in which one solder particle 1 is accommodated in each of recesses 62 having unevenness on the bottom surface. On the other hand, the substrate 2 has a plurality of electrodes 3 on its surface. With the electrode 3 side surface of the substrate 2 facing the surface of the substrate 60 on the opening side of the recess 62 , the substrate 60 is held until the solder particles 1 accommodated in the recess 62 of the substrate 60 and the electrode 3 come into contact with each other. and the substrate 2 (arrows A and B in FIG. 6A). From the viewpoint of suitably bonding the solder particles 1 and the electrodes 3, the pressing means that the solder bump forming member 10 and the substrate 2 are pressed against each other in the directions of arrows A and B in FIG. It is in a state of being pressed with the force of Thereby, a solder bump having a depression on at least part of the surface can be formed on the electrode. FIG. 6(b) is a schematic diagram of the member 20 with solder bumps thus obtained. The number of solder particles 1 to be brought into contact with each electrode 3 is not particularly limited, and may be one particle per electrode, or multiple particles per electrode. Moreover, the solder particles 1 may be brought into contact with only a specific electrode among the plurality of electrodes. Since the force acting between the solder particles 1 and the recesses 62 (for example, intermolecular force such as van der Waals force) is greater than the gravity acting on the solder particles 1, even if the main surface of the substrate 60 faces downward, Solder particles 1 remain in recesses 62 without falling off. Moreover, when at least some of the solder particles 1 have flat portions that contact the bottom surface and/or the inner wall of the recesses 62 , the solder particles 1 are less likely to fall out of the recesses 62 .

If there are alignment marks on the surfaces of the base body and the substrate, respectively, it is easy to align the two. For example, alignment marks are provided in advance on the surface of the substrate and the substrate so that the positions of the recesses and the positions of the electrodes are opposed to each other when the solder bump forming member and the substrate with electrodes are opposed to each other. After that, when the solder bump forming member and the substrate with electrodes are actually opposed to each other, the positions are adjusted using the alignment marks, so that the solder bumps can be formed on the electrodes with high position accuracy.

In the pressing step, solder bumps may be formed by pressing the substrate 60 against the substrate 2 pre-coated with a resin material, flux material, or the like having a tack force. In this case, the solder particles 1 are transferred onto the electrodes 3, and a member with solder bumps in which the solder bumps 1A and the electrodes 3 are not metallically bonded can be obtained. By manufacturing a connection structure using such a member with solder bumps, it is possible to perform more suitable bonding between electrodes. The solder bumps are melted by heating when manufacturing the connection structure, and the upper and lower electrodes are joined together.

In the pressing step, the solder particles may be heated while the solder particles and the electrodes are brought into contact with each other under pressure from the viewpoint of more suitable bonding between the solder particles and the electrodes.
From the viewpoint of shortening the manufacturing time, the solder particles may be heated to a temperature equal to or higher than the melting point of the solder particles in the pressing step. When the solder particles are heated to the melting point of the solder particles or higher, the solder particles are melted and the solder particles are suitably joined to the electrode surface in a short time. Further, when the solder particles are heated to the melting point or higher of the solder particles, the fluidity of the solder particles is increased, so that the solder particles are easily brought into contact with the electrode surface, and the reliability of the connection with the electrode is enhanced. Specifically, for example, when solder particles are collectively bonded to a plurality of electrodes, the number of unbonded electrodes can be reduced and the yield can be increased. Moreover, since the reliability of bonding between the solder particles and the electrodes is enhanced, the pressing time can be shortened as a result.
In addition, in the pressing step, the solder particles may be heated to a temperature below the melting point of the solder particles from the viewpoint of making the bonding between the solder particles and the electrodes more uniform. When the solder particles are heated to a temperature below the melting point of the solder particles, the solder particles do not melt, but interdiffusion occurs between the metal elements in the solder particles and the metal elements in the electrodes at the contact surface between the solder particles and the electrodes. joins the two. By setting the heating temperature below the melting point, this interdiffusion occurs slowly, so the solder composition is easily maintained even after heating for a long time (for example, 10 minutes or longer). Specifically, when solder particles are jointed to multiple electrodes at once, even in situations where joint unevenness is likely to occur due to pressure unevenness, electrode surface height unevenness, etc., the pressing time is set long to alleviate pressure unevenness. Thus, the bonding between the solder particles and the electrodes can be performed more reliably.

Due to the oxide film, the solder particles 1 may not melt even when heated, or may not spread. Therefore, the solder particles 1 can be melted by exposing the solder particles 1 to a reducing atmosphere, removing the surface oxide film of the solder particles 1, and then heating the solder particles 1. Also, the solder particles 1 can be melted in a reducing atmosphere. By heating the solder particles 1 and creating a reducing atmosphere, the oxide film on the surface of the solder particles 1 is reduced, and further the oxide film on the surface of the electrode is reduced, so that the melting and spreading of the solder particles 1 efficiently proceed. becomes easier. That is, the method for manufacturing a member with solder bumps further comprises a reducing step of exposing the solder particles (and/or electrodes) to a reducing atmosphere before the placing step, or after the placing step and before the pressing step. good. Moreover, in the pressing step of the method for manufacturing a member with solder bumps, the solder particles may be heated in a reducing atmosphere. In the pressing process for forming the solder bumps on the electrodes, the electrodes and the openings of the solder bump forming members are brought into close contact with each other (with heating if necessary), so that the solder bumps are formed only on the electrodes, and the gap between the adjacent electrodes is reduced. solder bridging is easily suppressed.

For details of the reducing atmosphere, the description of the manufacturing method of the solder bump forming member can be referred to as appropriate.

When heating is performed during the pressing process, by cooling the whole after heating, the top of the electrodes 3 and the solder bumps 1A formed by melting the solder particles 1 are fixed.

After the solder bumps 1A are formed on the electrodes 3, the members 20 with solder bumps can be obtained by removing the solder bump forming members 10 from the substrate 2. That is, the method of manufacturing a member with solder bumps may further include a removing step of removing the base from the substrate after the pressing step.

Solder particles 1 that are detached from recesses 62 but are not used for bonding with electrodes 3 may exist on the resulting solder bumped member 20 . Therefore, the method of manufacturing a member with solder bumps may further include a cleaning step for removing solder particles that are not bonded to the electrodes. Examples of cleaning methods include blowing compressed air and rubbing the surface of the substrate with a nonwoven fabric or a fiber bundle.

According to the manufacturing method of the member with solder bumps, it is possible to obtain the member with solder bumps 20 which comprises the substrate 2, the electrodes 3, and the solder bumps 1A in this order, and in which at least a part of the solder bump surface is formed with a depression. .

<Method for manufacturing connection structure>
7A and 7B are cross-sectional views schematically showing an example of the manufacturing process of the connection structure. A method for manufacturing the connection structure will be described with reference to FIGS. 7(a) and 7(b). First, a member 20 with solder bumps shown in FIG. 6B is prepared in advance. Also, another substrate 4 having a plurality of other electrodes 5 is prepared. Then, both are arranged so that the solder bump 1A and the other electrode 5 face each other. After that, the solder bump 1A is melted between the electrode 3 and the other electrode 5 by heating while the solder bump 1A and the other electrode 5 are kept in contact with each other. After that, by cooling the whole, a solder layer 1B is formed between the electrode 3 and another electrode 5, and the electrodes are electrically connected. At this time, by heating in an oxygen-free atmosphere, oxidation of the solder bumps 1A and the electrodes 5 can be suppressed. For example, heating in an inert gas atmosphere such as nitrogen can be mentioned, and specifically, a vacuum reflow furnace, a nitrogen reflow furnace, or the like can be used.

Heating can be performed in a reducing atmosphere in order to melt the solder bumps 1A by heating and more suitably join the

electrodes

3 and 5 facing each other. Hydrogen gas, hydrogen radicals, formic acid, and the like can be used to create a reducing atmosphere. Specifically, hydrogen reduction furnaces, hydrogen reflow furnaces, hydrogen radical furnaces, formic acid furnaces, vacuum furnaces of these, continuous furnaces, and conveyor furnaces can be used. By creating a reducing atmosphere, the oxide film on the surface of the solder bump 1A and the oxide film on the surface of the electrode 5 can be reduced and removed. A more stable bond is achieved between 3 and electrode 5 .

You may apply pressure to achieve a stable connection. A member 20 with solder bumps shown in FIG. 6B is prepared in advance. Also, another substrate 4 having a plurality of other electrodes 5 on its surface is prepared. Then, both are arranged so that the solder bump 1A and the other electrode 5 face each other. Thereafter, pressure is applied in the thickness direction of the laminate of these members (directions of arrows A and B shown in FIG. 7(a)). The solder bumps 1A are melted between the electrode 3 and the other electrode 5 by heating the whole when pressurized. After that, by cooling the whole, a solder layer 1B is formed between the electrode 3 and another electrode 5, and the electrodes are electrically connected. Also in this case, in order to suppress oxidation of the surfaces of the solder bumps 1A, the

electrodes

5 and 3, the above process may be performed under vacuum, under an inert gas atmosphere such as nitrogen, or under a reducing atmosphere. Examples of the method for creating a reducing atmosphere include the aforementioned hydrogen gas, hydrogen radicals, and formic acid. Specifically, hydrogen reduction furnaces, hydrogen reflow furnaces, hydrogen radical furnaces, formic acid furnaces, vacuum furnaces thereof, continuous furnaces, conveyer furnaces, and the like can be used.

A flux material having a reducing action can be placed near the solder bumps 1A or the

electrodes

5 and 3. First, a member 20 with solder bumps shown in FIG. 6B is prepared in advance. A flux material is placed over the entire surface of the member 20 on which the solder bumps 1A are formed or in the vicinity of the solder bumps 1A and the electrodes 3 including the solder bumps 1A. Also, another substrate 4 having a plurality of other electrodes 5 on its surface is prepared. Then, both are arranged so that the solder bump 1A and the other electrode 5 face each other. Thereafter, the solder bump 1A is melted between the electrode 3 and the other electrode 5 by heating the solder bump 1A and the other electrode 5 while they are in contact with each other through, for example, a flux material. After that, by cooling the whole, a solder layer 1B is formed between the electrode 3 and another electrode 5, and the electrodes are electrically connected. After that, when the flux component is removed by washing, corrosion of the solder layer 1B and the

electrodes

3 and 5 can be suppressed by the residual flux.

As another method, a member 20 with solder bumps shown in FIG. 6(b) is prepared in advance. Also, another substrate 4 having a plurality of other electrodes 5 on its surface is prepared, and the flux material is arranged on the entire surface of the substrate 4 having the electrodes 5 or near the surface of the electrodes 5 . Then, both are arranged so that the solder bump 1A and the other electrode 5 face each other. Thereafter, the solder bump 1A is melted between the electrode 3 and the other electrode 5 by heating the solder bump 1A and the other electrode 5 while they are in contact with each other through, for example, a flux material. After that, by cooling the whole, a solder layer 1B is formed between the electrode 3 and another electrode 5, and the electrodes are electrically connected.

In any of the above methods, the flux material is captured in the depressions on the solder bump surface, and a sufficient amount of flux necessary for bonding the solder bump to the electrode is ensured.

As a heating method for melting the solder bumps 1A, in a vacuum, for example, a heating plate in a reflow furnace is heated, and the solder bumps 1A are heated through the

substrates

2 and 4 in contact with the heating plate. A method using radiation such as Also, in addition to or in combination with the heating method using a heating plate or infrared rays, a heated gas and a method of heating the solder bumps 1A via the gas can be used. Specifically, the solder bumps 1A can be heated by heating an inert gas such as nitrogen, hydrogen, hydrogen radicals, formic acid, or the like. The flux material may contain at least one selected from the group consisting of succinic acid, glutaric acid, adipic acid, pimelic acid, suberic acid, benzoic acid, and malic acid.

Another method is to use electromagnetic waves such as microwaves. For example, a specific electromagnetic wave can be applied from the outside that heats the components of

electrodes

3, 5 and solder bumps 1A. For example, when the substrate 4 and the substrate 2 are resin substrates, when a specific electromagnetic wave is irradiated from the outside of the substrate 4 and the substrate 2, the electromagnetic wave penetrates the substrate 4 and the substrate 2, and the electrode 3 and the solder bump 1A or the electrode 5 are formed. Heated by electromagnetic waves. In the case of this method, it is possible to selectively heat the parts to be joined, so there is an advantage that unnecessary heat history does not remain. For example, even if the

substrates

2 and 4 are made of a material with low heat resistance, the solder bumps 1A can be melted to reliably bond the

electrodes

3 and 5 together. In addition, there is an advantage in that warping and decomposition after bonding are easily suppressed because the heat history is less likely to remain in the entire system to be bonded. When using microwaves, the solder bumps 1A can be melted in a shorter time than when using a heating plate, infrared rays, heated gas, etc., so there is the advantage that the heat history of the entire system to be bonded can be reduced. effect is easily obtained. Furthermore, by using microwaves, it is possible to locally heat only the portions of the electrodes 3, the solder bumps 1A and the electrodes 5 desired to be bonded or melted. Therefore, there is no need to heat the entire system, and the solder bumps 1A can be melted and joined even if there is a material with low heat resistance, other electronic components, or the like near the

electrodes

3 and 5 to which heat should not be applied. be able to.

Another method is to use ultrasound. For example, when an ultrasonic vibrator is arranged on the opposite side of the substrate 2 from the electrode 3 and ultrasonic waves are applied, the solder bumps 1A are melted by the vibrational energy of the ultrasonic waves. As a result, the electrode 3 and the electrode 5, which has been arranged in advance to face the electrode 3, are joined via the solder layer 1B. Bonding by ultrasonic waves can melt the solder bumps 1A in a short period of time, so there is no need to apply heat to the

entire substrates

2 and 4, and even if the

substrates

2 and 4 are made of materials with low heat resistance, the electrodes can be reliably bonded. 3 and the electrode 5 can be joined.

FIG. 7(b) is a schematic diagram of the connection structure 30 thus obtained. That is, FIG. 7B schematically shows a state in which an electrode 3 of the substrate 2 and another electrode 5 of another substrate 4 are connected via a solder layer 1B formed by fusion bonding. is shown. As used herein, the term “fusion” means that at least a part of the electrode is joined with solder (solder bump 1A) melted by heat, and then solder is joined to the surface of the electrode by going through a process of solidifying it. means state. The connection structure 30 includes a first circuit member having a substrate and a plurality of electrodes on its surface, a second circuit member having another substrate and a plurality of other electrodes on its surface, a plurality of electrodes and a plurality of electrodes. and a solder layer between other electrodes. From the viewpoints of maintaining the strength of the connection structure and blocking deterioration factors (moisture, oxygen, etc.) of the connection portion, the circuit members may be sealed by a method such as mold underfill, capillary underfill, or edge bond. For example, the space between the first circuit member and the second circuit member can be filled with an underfill material containing epoxy resin as a main component.

The connection structures are applied to semiconductor memories, semiconductor logic chip connections, primary and secondary mounting connections of semiconductor packages, CMOS image elements, laser elements, LED light emitting elements, and other joints. Devices such as cameras, sensors, liquid crystal displays, personal computers, mobile phones, smart phones, and tablets used.

The contents of this embodiment are listed below.
A preparation step of preparing a substrate having a plurality of recesses having unevenness on the bottom surface, an arrangement step of arranging solder particles in the recesses, and pressing the substrate and a substrate having electrodes in a state in which the solder particles and the electrodes face each other. and a pressing step of bringing the solder particles and the electrodes into contact with each other to form solder bumps having depressions on at least part of the surface of the electrodes.
The above manufacturing method, wherein the solder particles are heated in the pressing step.
The above manufacturing method, further comprising a reducing step of exposing the solder particles to a reducing atmosphere before the arranging step.
The above manufacturing method, further comprising a reducing step of exposing the solder particles to a reducing atmosphere after the arranging step and before the pressing step.
The above manufacturing method, wherein the solder particles are heated in a reducing atmosphere in the pressing step.
The above manufacturing method, further comprising a removing step of removing the base from the substrate after the pressing step.
The above manufacturing method, further comprising a cleaning step for removing solder particles not bonded to the electrode after the removing step.
A member with solder bumps, comprising: a substrate having electrodes; and solder bumps on the electrodes, wherein depressions are formed in at least part of the surfaces of the solder bumps.
The above solder bumped member, wherein the depth of the solder bump depression is 25% or less of the solder bump height.
A solder bumped component as above wherein adjacent solder bumps are independent of each other.
The above solder bumped member, wherein the height of the solder bumps is smaller than the diameter of the solder bumps in the planar direction.
A member for forming solder bumps, comprising: a substrate having a plurality of recesses having unevenness on the bottom; and solder particles in the recesses.
The member for forming solder bumps, wherein the height difference between the recesses and protrusions of the unevenness is 20% or less of the average particle diameter of the solder particles.
The average particle size of the solder particles is 1 to 35 μm, and the C.I. V. The above solder bump forming member, wherein the value is 20% or less.

The present embodiment will be described in more detail below with reference to examples, but the present embodiment is not limited to these examples.

<Fabrication of bump forming member>
(Production example 1)
Step a1: Classification of Solder Fine Particles 100 g of Sn—Bi solder fine particles (manufactured by 5N Plus, melting point 139° C., Type 8) are immersed in distilled water, ultrasonically dispersed, left to stand, and the solder fine particles floating in the supernatant. recovered. This operation was repeated to collect 10 g of solder fine particles. The average particle diameter of the obtained solder fine particles was 1.0 μm, and the C.I. V. The value was 42%.
Step b1: Arrangement on Substrate As shown in Table 1, the opening diameter is 6.2 μm, the bottom diameter is 4.8 μm, the depth is 3.7 μm, and the bottom surface unevenness (difference in height between concave and convex portions. Formed by dry etching. ) A substrate 1 (polyimide film, thickness 100 μm) having a plurality of recesses of 0.4 μm was prepared. A plurality of concave portions were arranged regularly at a pitch of 6.4 μm. The solder fine particles (average particle diameter 1.0 μm, CV value 42%) obtained in step a1 were placed in the recesses of the substrate 1 . The side of the substrate 1 on which the recesses were formed was rubbed with a slightly adhesive roller to remove excess solder particles, and the solder particles were arranged only in the recesses.
Step c1: Formation of Solder Particles The substrate 1 with the solder particles arranged in the recesses in step b1 is placed in a hydrogen radical reduction furnace (plasma reflow device manufactured by Shinko Seiki Co., Ltd.), and after evacuation, hydrogen gas is introduced into the furnace. The furnace was filled with hydrogen gas. After that, the inside of the furnace was adjusted to 120° C., and hydrogen radicals were irradiated for 5 minutes. After that, the hydrogen gas in the furnace is removed by vacuuming, and after heating to 170° C., nitrogen is introduced into the furnace to restore the atmospheric pressure, and then the temperature in the furnace is lowered to room temperature to remove the solder particles. formed. As a result, a solder bump forming member (film) having solder particles in the concave portions was obtained.

<Evaluation of Solder Particles>
A part of the solder bump forming member obtained through step c1 was fixed to the surface of a pedestal for observation with a scanning electron microscope (SEM), and the surface was subjected to platinum sputtering. The diameter of 300 solder particles was measured by SEM, and the average particle diameter was calculated. Table 2 shows the results. Further, the surface shape of a part of the solder bump forming member obtained through step c1 is measured using a laser microscope (manufactured by Olympus Corporation, LEXT OLS5000-SAF), and the height of the solder particles from the substrate surface is measured. It was measured and the average value of 300 was calculated. Table 2 shows the results.

(Production Examples 2 to 4)
Substrates holding solder particles in recesses were produced and evaluated in the same manner as in Production Example 1, except that the size of the recesses was changed as shown in Table 1. Table 2 shows the results.

<Fabrication of evaluation chip with solder bumps>
Step d1: Preparation of Evaluation Chip Four types of chips with gold bumps (5×5 mm, thickness: 0.5 mm) shown below were prepared.
Chip C1: electrode size: 24 μm×12 μm, pitch: 48 μm in X direction, 24 μm in Y direction, number of bumps: 15,000 Chip C2: electrode size: 72 μm×36 μm, pitch: 144 μm in X direction, 72 μm in Y direction, number of bumps : 3400 Chip C3: Electrode size: 96 µm x 48 µm, Pitch: X direction: 192 µm, Y direction: 96 µm, Number of bumps: 850 Chip C4: Electrode size: 140 µm x 70 µm, Pitch: X direction: 280 µm, Y direction: 140 µm, Number of bumps : 420 pieces

Step e1: Solder Bump Formation Solder bumps were formed on the chip C1 prepared in step d1 using the solder bump forming member 1 produced in step c1 according to the following procedures i) to ii).
i) Place a glass plate (thickness 0.3 mm) on the lower hot plate of a formic acid reflow furnace (manufactured by Shinko Seiki Co., Ltd., batch type vacuum soldering device), and place the evaluation chip C1 on the glass plate with the Au electrode facing upward. arranged to be A solder bump forming member 1 adjusted to a size of 7×7 mm was placed on the evaluation chip C1 so that the opening side of the concave portion faced the Au electrode. Further, a glass plate (thickness: 0.3 mm) and a SUS weight were placed thereon in that order.
ii) After evacuating, the chamber was filled with formic acid gas, the temperature of the lower hot plate was raised to 150° C., and the chamber was heated for 1 minute.
After that, after discharging formic acid gas by vacuuming, nitrogen substitution was performed, the lower hot plate was returned to room temperature, the inside of the furnace was opened to the atmosphere, solder particles were transferred onto the electrodes of the evaluation chip C1, and solder bumps were formed. . Thus, an evaluation chip with solder bumps (member with solder bumps) was obtained. A depression of a predetermined depth was formed on the top of the solder bump.

<Evaluation of evaluation chips with solder bumps>
The height and diameter of the solder bumps on the evaluation chip obtained through step e1 and the depth of the recesses at the top of the solder bumps are measured using a laser microscope (manufactured by Olympus Corporation, LEXT OLS5000-SAF), and the average of 300 value was calculated. Table 3 shows the results.

(Production Examples 2 to 4)
Solder bumps were formed in the same manner as step e1, except that solder bump forming members 2 to 4 and evaluation chips C2 to C4 were used. Furthermore, 300 solder bumps on the electrodes were evaluated to calculate the average values of the height and diameter of the solder bumps and the depth of the depressions at the top of the solder bumps. Table 3 shows the results.

FIG. 8 is a SEM photograph of the solder bump forming member obtained in Production Example 1. FIG. It can be seen that a recess is formed at the top of the solder bump.

<Production of connection structure>
Step f1: Preparation of Evaluation Substrate Four kinds of evaluation substrates with gold bumps (70×25 mm, thickness: 0.5 mm) shown below were prepared. The gold bumps are arranged at positions facing the gold electrodes of the aforementioned evaluation chips C1 to C4, and alignment marks are provided on the substrate. In addition, lead wiring for resistance measurement is formed on a part of the gold bump.
Substrate D1: area 24 μm×12 μm, pitch: 48 μm in X direction, 24 μm in Y direction, height: 3 μm, number of bumps: 15,000 Substrate D2: area 72 μm×36 μm, pitch: 144 μm in X direction, 72 μm in Y direction, height Thickness: 3 μm, Number of bumps: 3400 Substrate D3: Area 96 μm×48 μm, Pitch: X direction 192 μm, Y direction 96 μm, Height: 3 μm, Number of bumps: 850 Substrate D4: Area 140 μm×70 μm, Pitch: X direction 280 μm , Y direction 140 μm, height: 3 μm, number of bumps: 420

Step g1: Electrode bonding According to the procedures i) to iv) shown below, the evaluation chip with solder bumps prepared in step e1 and the evaluation substrate with gold bumps prepared in step f1 were connected via solder bumps. .
i) The gold-bumped evaluation substrate D1 was fixed on the stage of a spin coater SC-308S (manufactured by Oshikane Co., Ltd.) using a polyimide tape. Liquid flux WHS-003C (manufactured by Arakawa Chemical Industries, Ltd.) was added drop by drop at intervals of 1 cm, totaling 9.6 g, and then the stage was rotated at 500 rpm for 10 seconds and 1000 rpm for 3 seconds in a two-step process. Thus, an evaluation substrate D1 with gold bumps on which the liquid flux was uniformly applied was obtained. The thickness of the flux layer measured after natural drying was 1 μm or less.
ii) An evaluation board D1 with gold bumps coated with flux is placed on the stage of FC3000W (manufactured by Toray Engineering Co., Ltd.), the evaluation chip C1 with solder bumps is picked up by the head, and gold electrodes are opposed to each other using alignment marks. Then, the evaluation chip C1 with solder bumps was placed on the evaluation board D1 with gold bumps coated with flux. Then, while pressing both with a tool, they were thermocompression bonded at a tool temperature of 220° C. for 6 seconds to obtain a bonded sample 1.
iii) The flux remaining between the evaluation chip and the evaluation substrate was removed by cleaning the bonding sample 1 with isopropyl alcohol.
iv) Put an appropriate amount of underfill material (manufactured by Hitachi Chemical Co., Ltd., CEL series) with adjusted viscosity between the evaluation chip of bonding sample 1 from which residual flux was removed and the evaluation substrate, and after filling by vacuuming, 125 ° C. and cured for 4 hours to produce a connection structure 1 between the evaluation chip and the evaluation substrate.

(Production Examples 2 to 4)
Connection structures 2 to 4 were produced in the same manner as in step g1, except that evaluation chips C2 to 4 with solder bumps and evaluation substrates D2 to D4 with gold bumps were used. The combination of each material used for manufacturing the connection structure is as follows.
Connection structure (1): chip C1/solder bump forming member 1/substrate D1
Connection structure (2): chip C2/solder bump forming member 2/substrate D2
Connection structure (3): chip C3/solder bump forming member 3/substrate D3
Connection structure (4): chip C4/solder bump forming member 4/substrate D4

<Evaluation of connection structure>
A conduction resistance test and an insulation resistance test were performed on a part of the obtained connection structure. The results are shown in Tables 4, 5 and 6.

(Continuity resistance test - Moisture absorption and heat resistance test)
Regarding the conduction resistance between the chip with gold bumps and the substrate with gold bumps, the initial value of the conduction resistance and the value after the moisture absorption and heat resistance test (100, 500, 1000 hours under the conditions of temperature 85 ° C and humidity 85%) were measured for 20 samples. were measured and their average values were calculated. Conduction resistance was evaluated from the obtained average values according to the following criteria. Table 4 shows the results. Incidentally, when the following criterion A or B is satisfied after 1000 hours of the moisture absorption and heat resistance test, it can be said that the conduction resistance is good.
A: Average conduction resistance less than 2 Ω B: Average conduction resistance 2 Ω or more and less than 5 Ω C: Average conduction resistance 5 Ω or more and less than 10 Ω D: Average conduction resistance 10 Ω or more and less than 20 Ω E: Conduction resistance average value of 20Ω or more

(Continuity resistance test - High temperature exposure test)
Regarding the conduction resistance between the chip with gold bumps (bumps) and the substrate with gold bumps (bumps), the initial value of the conduction resistance and the value after the high temperature storage test (left at 100°C for 100, 500, and 1000 hours) are Twenty samples were measured. After being left at a high temperature, a drop impact was applied, and the conduction resistance of the sample after the drop impact was measured. The drop impact was generated by fixing the connection structure to a metal plate with screws and dropping it from a height of 50 cm. After the drop, the DC resistance value was measured at the solder joints (four points) at the chip corners where the impact was the greatest, and when the measured value increased by 5 times or more from the initial resistance, it was considered that a fracture had occurred, and evaluation was performed. A total of 80 measurements were made, four on each sample. Table 5 shows the results. Solder connection reliability was evaluated to be good when the following criterion A or B was satisfied after 20 drops.
A: There were no breakage locations.
B: There were 1 or more and 5 or less fracture locations.
C: There were 6 or more and 20 or less places where breakage occurred.
D: There were 21 or more breakage locations.

(insulation resistance test)
Regarding the insulation resistance between the chip electrodes, the initial value of the insulation resistance and the value after the migration test (100, 500, 1000 hours under the conditions of temperature 60 ° C., humidity 90%, 20 V application) were measured for 20 samples. The percentage of samples with an insulation resistance value of 10 ⁹ Ω or more among the 20 samples was calculated. The insulation resistance was evaluated from the obtained ratio according to the following criteria. Table 6 shows the results. In addition, when the following criteria A or B are satisfied after 1000 hours of the migration test, it can be said that the insulation resistance is good.
A: 100% of the insulation resistance value is 10 ⁹ Ω or more
B: 90% or more and less than 100% of the insulation resistance value is 10 ⁹ Ω or more C: 80% or more and less than 90% of the insulation resistance value is 10 ⁹ Ω or more D: 50% of the insulation resistance value is 10 ⁹ Ω or more More than 80% E: Less than 50% of the insulation resistance value is 10 ⁹ Ω or more

DESCRIPTION OF SYMBOLS 1... Solder particle 1A... Solder bump 1B... Solder layer 2... Substrate 3... Electrode 4... Other substrate 5... Other electrode 10... Solder bump forming member 20... Member with solder bump, DESCRIPTION OF SYMBOLS 30... Connection structure, 60... Base|substrate, 62... Concave part, 111... Solder particle, 600... Substrate, 601... Base layer, 602... Concave part layer.

Claims

A preparation step of preparing a substrate having a plurality of recesses having unevenness on the bottom;
an arranging step of arranging solder particles in the recess;
The substrate and the substrate having electrodes are pressed in a state in which the solder particles and the electrodes face each other to bring the solder particles and the electrodes into contact with each other, thereby forming solder bumps having depressions on at least part of their surfaces. A pressing step of forming on the electrode;
A method of manufacturing a member with solder bumps, comprising:
The manufacturing method according to claim 1, wherein the solder particles are heated in the pressing step.
The manufacturing method according to claim 1 or 2, further comprising a reducing step of exposing the solder particles to a reducing atmosphere before the arranging step.
The manufacturing method according to any one of claims 1 to 3, further comprising a reducing step of exposing the solder particles to a reducing atmosphere after the arranging step and before the pressing step.
The manufacturing method according to any one of claims 1 to 4, wherein in the pressing step, the solder particles are heated under a reducing atmosphere.
The manufacturing method according to any one of claims 1 to 5, further comprising a removing step of removing the base from the substrate after the pressing step.
The manufacturing method according to claim 6, further comprising a cleaning step of removing the solder particles not bonded to the electrodes after the removing step.
a substrate having electrodes; and solder bumps on the electrodes;
A member with solder bumps, wherein a depression is formed in at least a part of the surface of the solder bumps.
The member with solder bumps according to claim 8, wherein the recess depth of the solder bumps is 25% or less of the height of the solder bumps.
The member with solder bumps according to claim 8 or 9, wherein adjacent solder bumps are independent of each other.
The member with solder bumps according to any one of claims 8 to 10, wherein the height of the solder bumps is smaller than the diameter of the solder bumps in the planar direction.
A member for forming solder bumps, comprising a substrate having a plurality of recesses having unevenness on the bottom surface, and solder particles in the recesses.
13. The member for forming solder bumps according to claim 12, wherein the height difference between the recesses and protrusions of the unevenness is 20% or less of the average particle diameter of the solder particles.
C. the solder particles have an average particle size of 1 to 35 μm; V. 14. The member for forming solder bumps according to claim 12, wherein the value is 20% or less.