Metal paste

Abstract

To provide a conductor that has sufficient conductivity and whose resistance does not easily increase even when subjected to temperature changes, a substrate having a through electrode that has sufficient conductivity and excellent connection reliability, a method for manufacturing the same, and a metal paste used to form the through electrode. [Solution] The metal paste is used to form through electrodes and contains copper particles and solder particles, wherein the copper particles include first copper particles with a particle size of 0.8 μm or more and second copper particles with a particle size of 0.5 μm or less.

Description

[Technical Field]

[0001] The present invention relates to a metal paste, a conductor, a substrate having through electrodes, and a method for manufacturing the same. [Background technology]

[0002] In recent years, in order to miniaturize, enhance the functionality, and integrate electronic devices and components, three-dimensional packaging technology has attracted attention. This technology involves electrically connecting silicon substrates arranged vertically and horizontally via electrodes called through-silicon vibrators (TSVs) to create a high-density stack of semiconductor chips in the vertical direction (height direction).

[0003] As a method for forming through-silicon electrodes, for example, Patent Document 1 discloses a method for manufacturing a semiconductor device having through-silicon electrodes, which includes a step of copper plating non-through vias formed on a silicon substrate by electroplating using a specific copper plating solution. [Prior art documents] [Patent Documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2019-16712 [Overview of the project] [Problems that the invention aims to solve]

[0005] However, the method described in Patent Document 1 requires a longer working time because it is necessary to suppress the deposition rate of the copper film during plating, which presents challenges in terms of productivity.

[0006] On the other hand, substrates with through-silicon electrodes are required to have not only sufficient conductivity but also excellent connection reliability, such as resistance not increasing easily even when subjected to temperature changes.

[0007] Therefore, one aspect of the present invention aims to provide a conductor that has sufficient conductivity and whose resistance value does not easily increase even when subjected to temperature changes, a substrate having a through electrode that has sufficient conductivity and excellent connection reliability, a method for manufacturing the same, and a metal paste used for forming the through electrode. [Means for solving the problem]

[0008] As a result of diligent research to achieve the above objective, the inventors have found that in a conductor comprising a copper sintered body having a porous structure, a conductor that is less prone to cracking can be formed by scattering a specific metal within the copper sintered body and providing a metal body with voids in specific areas. Furthermore, the inventors have found that a substrate having through electrodes with through holes filled with such a conductor exhibits a sufficiently low initial resistance value and that the resistance value does not increase easily even in temperature cycle connection reliability tests, thus completing the present invention.

[0009] In other words, one aspect of this disclosure provides the following invention. [1] A conductor comprising a copper sintered body having a porous structure and solder scattered on the copper sintered body, the conductor comprising a metal body having voids, wherein the metal body includes voids as voids, which are voids present inside the solder and / or voids present between the solder and the copper sintered body. [2] The conductor described in [1], wherein the solder is tin or a tin alloy. [3] The conductor according to [1], wherein the solder is an alloy of In-Sn, In-Sn-Ag, Sn-Bi, Sn-Bi-Ag, Sn-Ag-Cu, or Sn-Cu. [4] The conductor according to any one of [1] to [3], further comprising a resin cured material present in the pores of the metal body. [5] The conductor according to [4], wherein the porosity of the metal body is 1 to 15 volume percent based on the volume of the metal body, and the content of the resin cured material in the conductor is 80 volume percent or more based on the total volume of the internal space of the pores in the metal body.

[0010] In addition, another aspect of the present disclosure provides the following inventions. [6] A substrate having a through electrode, comprising an insulating substrate provided with through holes, the substrate having through holes communicating with both main surfaces, and a conductor filling the through holes, wherein the conductor is the conductor according to any one of [1] to [5] above. [7] The substrate having a through electrode according to [6], wherein the substrate includes a metal film provided at least on the wall surface of the through hole. [8] The substrate having a through electrode according to [6] or [7], wherein the ratio L / D of the length L to the aperture diameter D of the through electrode is 10 or more. [9] The substrate having a through electrode according to any one of [6] to [8], wherein the conductor covers at least a part of the main surface of the substrate.

[10] The substrate having a through electrode according to any one of [6] to [9], wherein the insulating substrate is a silicon wafer and the through electrode is a silicon through electrode.

[0011] In addition, another aspect of the present disclosure provides the following inventions.

[11] A method for manufacturing a substrate having a through electrode, comprising a preparation step of preparing a substrate including an insulating substrate provided with through holes and having through holes communicating with both main surfaces, and a conductor forming step of forming a conductor in the through holes, wherein the conductor forming step includes a metal body forming step of forming a metal body having pores, containing a copper sintered body having a porous structure and solder so as to fill at least the through holes.

[12] The method according to

[11] , wherein the conductor forming step further includes a resin impregnation step of impregnating the metal body with a curable resin composition, and a resin curing step of curing the curable resin composition impregnated in the metal body.

[13] A manufacturing method of a substrate having through electrodes, comprising: a preparation step of preparing an insulating substrate provided with non-through holes, wherein the non-through holes open on one main surface; a conductor formation step of forming a conductor in the non-through holes; and a grinding step of providing through electrodes by grinding the side opposite to the surface where the non-through holes of the substrate on which the conductor is formed open, wherein the conductor formation step includes a metal body formation step of forming a metal body having a porous structure and containing solder so as to at least fill the non-through holes.

[14] The method according to

[13] , wherein the conductor formation step further includes a resin impregnation step of impregnating the metal body with a curable resin composition, and a resin curing step of curing the curable resin composition impregnated in the metal body.

[15] The method according to any one of

[11] to

[14] , wherein the solder contains tin or a tin alloy.

[16] The method according to any one of

[11] to

[14] , wherein the solder is an alloy of In-Sn, In-Sn-Ag, Sn-Bi, Sn-Bi-Ag, Sn-Ag-Cu, or Sn-Cu system.

[17] The method according to any one of

[11] to

[16] , wherein the porosity of the metal body is 1 to 15% by volume based on the volume of the metal body.

[18] The method according to

[12] or

[14] , wherein the content of the cured resin in the conductor is 80% by volume or more based on the total volume of the internal space of the pores of the metal body.

[19] The method according to any one of

[11] to

[18] , wherein in the metal body formation step, the metal body is formed so as to cover at least a part of the main surface of the substrate.

[20] The method according to

[19] , further comprising a conductor removal step of removing at least a part of the conductor formed on the main surface of the substrate.

[21] The method according to any one of

[11] to

[20] , wherein the ratio L / D of the length L to the pore diameter D of the through electrode is 10 or more.

[22] The method according to

[11] or

[12] , wherein the metal body forming step comprises a paste filling step of filling through holes in a substrate with a metal paste containing copper particles and solder particles, and a paste firing step of firing the metal paste to form the metal body.

[23] The method according to

[13] or

[14] , wherein the metal body forming step comprises a paste filling step of filling non-through holes of a substrate with a metal paste containing copper particles and solder particles, and a paste firing step of firing the metal paste to form the metal body.

[24] The method according to

[22] or

[23] , wherein the metal paste comprises, as copper particles, first copper particles having a particle size of 0.8 μm or more and second copper particles having a particle size of 0.5 μm or less.

[25] The method according to

[24] , wherein the first copper particle is flattened.

[26] The method according to any one of

[22] to

[25] , wherein the solder particles described above include tin or a tin alloy.

[27] The method according to any one of

[22] to

[25] , wherein the solder particles are In-Sn, In-Sn-Ag, Sn-Bi, Sn-Bi-Ag, Sn-Ag-Cu, or Sn-Cu alloy.

[28] The method according to any one of

[22] to

[27] , wherein the metal paste is fired under pressure of 0.1 MPa or higher.

[29] The method according to any one of

[22] to

[28] , wherein the metal paste is fired in an atmosphere containing nitrogen, hydrogen, or formic acid.

[30] The method according to any one of

[11] to

[29] , wherein the insulating substrate is a silicon wafer and the through electrode is a silicon through electrode.

[0012] Furthermore, other aspects of this disclosure provide the following inventions.

[31] A metal paste used for forming through electrodes, comprising copper particles and solder particles, wherein the copper particles consist of first copper particles having a particle size of 0.8 μm or more and second copper particles having a particle size of 0.5 μm or less.

[32] The metal paste according to

[31] , wherein the first copper particles are flattened.

[33] The metal paste according to

[31] or

[32] , wherein the solder particles described above include tin or a tin alloy.

[34] The metal paste according to

[31] or

[32] , wherein the solder particles are In-Sn, In-Sn-Ag, Sn-Bi, Sn-Bi-Ag, Sn-Ag-Cu, or Sn-Cu alloy. [Effects of the Invention]

[0013] According to one aspect of the present invention, it is possible to provide a conductor that has sufficient conductivity and whose resistance value does not easily increase even when subjected to temperature changes, a substrate having a through electrode that has sufficient conductivity and excellent connection reliability, a method for manufacturing the same, and a metal paste used for forming the through electrode.

[0014] Furthermore, according to the above substrate and method, in which the conductor further comprises a resin cured product, it is possible to provide a substrate having a through electrode that is also excellent in airtightness and non-permeability (the property of not allowing liquid to penetrate). [Brief explanation of the drawing]

[0015] [Figure 1] This is a schematic diagram showing a method for manufacturing a substrate having a through electrode according to the first embodiment. [Figure 2] This is a schematic diagram showing a method for manufacturing a substrate having a through electrode according to the first embodiment. [Figure 3] This is a schematic diagram showing a method for manufacturing a substrate having a through electrode according to the first embodiment. [Figure 4] This is a schematic diagram showing a method for manufacturing a substrate having a through electrode according to the first embodiment. [Figure 5] This is a schematic diagram showing a method for manufacturing a substrate having a through electrode according to the first embodiment, and a substrate having a through electrode according to the first embodiment. [Figure 6] This is a schematic diagram showing a semiconductor device according to the first embodiment. [Figure 7] This is a schematic diagram showing a method for manufacturing a substrate having a through electrode according to the second embodiment. [Figure 8] This is a schematic diagram showing a method for manufacturing a substrate having a through electrode according to the second embodiment, and a substrate having a through electrode according to the second embodiment. [Figure 9] This is a schematic diagram showing a semiconductor device according to the second embodiment. [Figure 10] This is a schematic diagram showing a test specimen. [Figure 11] This is a cross-sectional photograph of a portion of the solder (SnBi58) containing voids in the sample prepared in Example 1. [Figure 12] This is a cross-sectional photograph of the portion of the sample prepared in Example 1 where there is a void around the outer edge of the solder (SnBi58). [Figure 13] This is a cross-sectional photograph of a portion of the solder (Sn96.5Ag3Cu0.5) where there is a void inside, in the sample prepared in Example 46. [Figure 14] This is a cross-sectional photograph of a portion of the copper layer containing a crack in the sample prepared in Comparative Example 1. [Modes for carrying out the invention]

[0016] The embodiments for carrying out the present invention will be described in detail below with reference to the drawings. The present invention is not limited to the following embodiments. In the drawings, the same or corresponding parts are denoted by the same reference numerals, and redundant descriptions are omitted.

[0017] (Method for manufacturing a substrate having a through electrode according to the first embodiment) Figures 1 to 5 are schematic diagrams illustrating a method for manufacturing a substrate having a through electrode according to the first embodiment.

[0018] A method for manufacturing a substrate having through electrodes according to the first embodiment includes a preparation step of preparing a substrate having through holes in an insulating substrate, wherein the through holes are connected to both main surfaces; a conductor forming step of forming a conductor in the through holes, wherein the conductor forming step includes a metal body forming step of forming a metal body having voids, containing a copper sintered body having a porous structure and solder, so as to fill at least the through holes; a resin impregnation step of impregnating the metal body with a curable resin composition; and a resin curing step of curing the curable resin composition impregnated into the metal body.

[0019] The method according to the first embodiment will be described using as an example a case in which a silicon wafer is prepared as the substrate and a substrate having through electrodes is manufactured (hereinafter sometimes referred to as a silicon through electrode substrate), but the following description may be interpreted by substituting the silicon wafer with other insulating substrates. Other insulating substrates include glass substrates, ceramic substrates, printed circuit boards, semiconductor package substrates, etc.

[0020] <Substrate preparation process> In this process, as shown in Figure 1(a), a silicon wafer 1 having through-holes 30 and a silicon substrate 40 having a metal coating 2 provided on the walls of the through-holes and the surface of the silicon wafer 1 can be prepared. The through-holes 30 are connected to both main surfaces of the silicon substrate 40.

[0021] The thickness of the silicon wafer 1 may be 100 μm or more, 200 μm or more, or 300 μm or more from the viewpoint of suppressing warping of the substrate after sintering, and may be 800 μm or less, 300 μm or less, 200 μm or less, or 100 μm or less from the viewpoint of reducing the weight and density of the substrate.

[0022] The upper limit of the diameter of the through-hole 30 may be 200 μm or less, 100 μm or less, or 60 μm or less, from the viewpoint of increasing the density of the resulting semiconductor device, and the lower limit of the diameter of the through-hole 30 is not particularly limited, but may be 20 μm or more, or 50 μm or more.

[0023] The number of through holes 30 provided in the silicon substrate 40 is determined from the viewpoint of increasing the density of the resulting semiconductor device, based on the number of holes per 1 cm of the main surface of the substrate. 2 There may be 100 or more, or 300 or more, per unit.

[0024] The metal coating 2 may be provided on both main surfaces of the silicon wafer 1 and on the walls of the through-holes 30, on at least one main surface of the silicon wafer 1 and on the walls of the through-holes 30, on the walls of the through-holes 30 only, or not provided at all. In the embodiment shown in Figure 1(a), the silicon substrate 40 has the metal coating 2 on both main surfaces of the silicon wafer 1 and on the walls of the through-holes 30.

[0025] Examples of metal coating 2 include titanium, nickel, chromium, copper, aluminum, palladium, platinum, and gold. From the viewpoint of adhesion, it is preferable that the metal coating 2 is a coating in which titanium, nickel, and copper are layered in that order. Adhesion is improved by oxidizing the surface of the silicon wafer 1 to silicon oxide and forming a titanium layer on top of the silicon oxide. Furthermore, by providing a nickel layer on top of the titanium layer and then a copper layer on top of that, the diffusion of copper into the silicon wafer 1 can be suppressed compared to the case where the copper layer is directly provided on top of the titanium layer. In addition, by providing a copper layer on the surface, the adhesion between the metal body formed in the metal body formation process described later and the silicon substrate 40 is improved.

[0026] <Metal body formation process> In this process, a porous metal body containing a copper sintered body and solder is formed to fill at least the through-holes. In this embodiment, the metal body may be formed to cover at least a portion of the main surface of the silicon substrate 40. In this case, a conductor is formed to fill the through-holes of the silicon substrate 40, and a conductor can also be provided on the main surface of the silicon substrate 40. The conductor provided on the main surface of the silicon substrate 40 can form wiring and silicon through-electrodes.

[0027] The metal body formation process may include a metal paste filling step of filling through-holes in a silicon substrate with a metal paste containing copper particles and solder particles, and a metal paste firing step of firing the metal paste to form the metal body. When forming the metal body on the main surface of the silicon substrate, layers of metal paste can be provided on both main surfaces of the silicon substrate during or after the metal paste filling step.

[0028] As part of the metal body formation process described above, for example, as shown in Figure 1(b), a metal paste 3 containing copper particles and solder particles can be applied to a silicon substrate 40, the metal paste 3 can be filled into the through holes 30, and layers of metal paste 3 can also be formed on both main surfaces of the silicon substrate 40. Details of the metal paste 3 will be described later.

[0029] Methods for applying the metal paste 3 to the silicon substrate 40 include, for example, screen printing, transfer printing, offset printing, jet printing, dispensers, jet dispensers, needle dispensers, comma coaters, slit coaters, die coaters, gravure coaters, slit coats, letterpress printing, intaglio printing, gravure printing, stencil printing, soft lithography, bar coating, applicators, particle deposition methods, spray coaters, spin coaters, dip coaters, and the like.

[0030] When the metal paste is also applied to the main surface of the silicon substrate, the thickness of the metal paste layer may be 1 μm or more, 2 μm or more, 3 μm or more, 5 μm or more, 10 μm or more, 15 μm or more, or 20 μm or more, and may be 300 μm or less, 250 μm or less, 200 μm or less, 150 μm or less, 120 μm or less, 100 μm or less, 80 μm or less, or 50 μm or less.

[0031] The metal paste 3 may be dried as appropriate, from the viewpoint of suppressing the flow of copper particles and solder particles during the sintering of the metal paste 3, and the generation of voids in the copper sintered body contained in the metal body. When drying the metal paste 3, the drying atmosphere may be an oxygen-free atmosphere such as nitrogen and noble gases, or a reducing atmosphere such as hydrogen and formic acid.

[0032] The drying method may be drying at room temperature, heating, or reduced pressure. For heating or reduced pressure drying, for example, a hot plate, hot air dryer, hot air heating furnace, nitrogen dryer, infrared dryer, infrared heating furnace, far infrared heating furnace, microwave heating device, laser heating device, electromagnetic heating device, heater heating device, steam heating furnace, hot plate press device, etc. may be used. The drying temperature and time may be adjusted as appropriate according to the type and amount of dispersion medium used. The drying temperature may be, for example, 50°C or higher and 180°C or lower. The drying time may be, for example, 1 minute or more and 120 minutes or less.

[0033] After the metal paste filling process, the metal paste 3 is fired to sinter the copper particles contained in the metal paste 3. In this way, as shown in Figure 2(c1), a metal-filled silicon substrate 50 is obtained in which a metal body 5 having voids (porosity) 4 fills the through holes 30, and which contains a copper sintered body having a porous structure and solder. In this embodiment, a metal-filled silicon substrate 50 is obtained in which metal bodies 5 are also provided on both main surfaces of the silicon substrate 40. Figure 2(c2) is a schematic diagram showing the structure of the metal body. The metal body includes a copper sintered body 12 having a porous structure, solder 14 scattered within the copper sintered body 12, and voids 4. The metal body can include voids 4 such as the porosity of the copper sintered body 12, voids inside the solder 14, and voids between the copper sintered body 12 and the solder 14.

[0034] Firing can be carried out by heat treatment. For heat treatment, heating means such as a hot plate, hot air dryer, hot air heating furnace, nitrogen dryer, infrared dryer, infrared heating furnace, far infrared heating furnace, microwave heating device, laser heating device, electromagnetic heating device, heater heating device, steam heating furnace, etc. can be used.

[0035] The atmosphere during firing may be an oxygen-free atmosphere from the viewpoint of suppressing oxidation of the copper sintered body, or a reducing atmosphere from the viewpoint of removing surface oxides from the copper particles in the metal paste 3. Examples of an oxygen-free atmosphere include the introduction of oxygen-free gases such as nitrogen or noble gases, or under vacuum. Examples of a reducing atmosphere include pure hydrogen gas, a mixed gas of hydrogen and nitrogen represented by a forming gas, nitrogen containing formic acid gas, a mixed gas of hydrogen and a noble gas, or a noble gas containing formic acid gas. When sintering the metal paste 3 by heating without pressurization, a pure hydrogen gas or a mixed gas of hydrogen and nitrogen represented by a forming gas is preferred, and a pure hydrogen gas is preferable. Heating in a pure hydrogen gas makes it possible to lower the sintering temperature of the copper particles. When using pure hydrogen gas, even if the substrate is thick at 600 μm and the diameter of the through-hole 30 is small at 10 μm, the gas can reach the center of the through-hole 30, making it easy to obtain a metal body 5 containing copper sintered body.

[0036] The maximum temperature reached during the heat treatment may be 150°C or higher, and may be 350°C or lower, 300°C or lower, or 260°C or lower, from the viewpoint of reducing thermal damage to each component and improving yield. If the maximum temperature reached is 150°C or higher, sintering tends to proceed sufficiently when the maximum temperature is held for 60 minutes or less. The maximum temperature is held for 1 minute or more, and may be 60 minutes or less, 40 minutes or less, or 30 minutes or less, from the viewpoint of completely volatilizing the dispersion medium and improving yield.

[0037] The firing of the metal paste may be carried out under pressure. In this case, under an atmosphere containing pure hydrogen gas, the pressure may be 0.05 MPa or higher, 0.1 MPa or higher, or 0.3 MPa or higher, and 20 MPa or lower, 15 MPa or lower, or 10 MPa or lower. Under an atmosphere containing nitrogen gas, the pressure may be 1 MPa or higher, or 3 MPa or higher, and 20 MPa or lower, 15 MPa or lower, or 10 MPa or lower.

[0038] By setting the pressure to 0.05 MPa or higher when using pure hydrogen gas, or 1 MPa or higher when using nitrogen gas, it becomes easier to suppress the generation of voids in the metal body 5 formed in the center of the through hole 30, and it becomes easier to obtain a metal body with good conductivity. Furthermore, by setting the pressure to above the above lower limit, if the silicon substrate 40 has a metal coating 2, it becomes easier to improve the bonding strength between the metal coating 2 and the metal body 5. Moreover, as shown in Figure 1(b), when pressurizing the silicon substrate 40 with a metal paste layer by sandwiching it from above and below with a pressurizing jig A, setting the pressure applied to the pressurizing jig A to above the above lower limit makes it easier to smooth the surface of the metal body formed on the main surface of the silicon substrate 40. A smooth surface of the metal body has the advantage of making it easier to form fine wiring when forming wiring by etching or the like in a later process. The pressurizing jig A is not particularly limited, but can be a commercially available one, or it can be made using a metal member with a flat part. For example, a pressurizing jig having two or more of the above-mentioned metal members can pressurize a silicon substrate by sandwiching it between metal members arranged so that their flat portions face each other. Pressurizing jig A may have a mechanism for adjusting the pressure applied to the silicon substrate. A spring or the like can be used as a means for adjusting the pressure.

[0039] If the pressure is 20 MPa or less, it becomes easier to suppress the warping of the silicon substrate 40. The inventors speculate that the reason for this effect is as follows. First, when the pressure is increased, the sintering density of the copper particles contained in the metal paste (particularly the density on the side in contact with the pressurizing jig A) increases, and the thermal expansion coefficient of the formed copper sintered body is thought to approach the typical thermal expansion coefficient of copper at 25°C, which is 16.5 μm / (m·K). On the other hand, the thermal expansion coefficient of silicon at 25°C is 2.6 μm / (m·K). Therefore, as the density of the copper sintered body increases, the difference in thermal expansion coefficients between the copper sintered body and silicon becomes larger, making warping more likely. In this embodiment, by setting the pressure to 20 MPa or less, the increase in the density of the copper sintered body is moderately suppressed, resulting in a smaller difference in thermal expansion coefficients between the copper sintered body and silicon, and thus suppressing warping.

[0040] Furthermore, if the pressure applied during firing is within the above range, a special pressurizing device is not required, thus reducing voids, improving joint strength, and connection reliability without compromising yield. Methods for applying pressure to a silicon substrate coated with metal paste include, for example, placing weights on it, using a pressurizing device, or using a fixing jig for pressurizing.

[0041] From the viewpoint of reducing the volume resistivity of the metal body, the porosity (percentage of voids contained in the metal body) of the metal body formed on the main surface of the silicon substrate may be 15 volume% or less, 14 volume% or less, 12 volume% or less, or 9 volume% or less, based on the total volume of the metal body (including voids). Furthermore, from the viewpoint of suppressing cracking and warping of the silicon substrate 40, the porosity of the metal body 5 may be 1 volume% or more, 3 volume% or more, or 5 volume% or more.

[0042] The presence of the porous structure described above in the metal body formed on the main surface of the silicon substrate makes it possible to reduce the coefficient of thermal expansion, thereby reducing the difference in coefficient of thermal expansion between the silicon wafer and the metal body, and suppressing cracking and warping of the silicon substrate.

[0043] From the viewpoint of reducing the volume resistivity of the metal body, the porosity of the metal body filling the through holes (the percentage of voids contained in the metal body) may be 15 volume% or less, 14 volume% or less, 12 volume% or less, or 9 volume% or less, based on the total volume of the metal body (including voids). Furthermore, from the viewpoint of relieving the stress applied to the copper sintered body and suppressing cracking and warping of the silicon substrate, the porosity of the metal body 5 may be 1 volume% or more, 3 volume% or more, or 5 volume% or more.

[0044] The metal body filling the through-hole has the porous structure described above, which relieves stress on the metal body and suppresses cracking and warping of the silicon substrate.

[0045] The porosity of a metal body is calculated using the following procedure. (i) A focused ion beam is used to expose the cross-section of the metal body in the metal-filled silicon substrate (the cross-section in the thickness direction of the substrate). (ii) The exposed cross section is photographed using a scanning electron microscope to capture cross-sectional images (a range of 10 μm in the thickness direction of the substrate and a range of 10 μm in the direction perpendicular to the thickness direction of the substrate). (iii) The obtained cross-sectional image is binarized so that the metal part and the porous part are separated. (iv) From the binarized cross-sectional image, the ratio of the area of ​​the porous portion to the total area of ​​the cross-sectional area of ​​the metal body is defined as the porosity of the metal body. When calculating the porosity of a metal body filled in a through-hole, in (i) above, the cross-section of the central part of the metal body filled in the through-hole is exposed. When calculating the porosity of the central part of the metal body filled in a through-hole, the area is observed from the center of the metal body filled in the through-hole, within a range of ±5 μm in the thickness direction of the substrate and ±5 μm in the direction perpendicular to the thickness direction of the substrate. When calculating the porosity of a metal body formed on the main surface of a metal-filled silicon substrate, in (i) above, the cross-section of the metal body on the main surface is exposed. When calculating the porosity of a metal body formed on the main surface of a metal-filled silicon substrate, the area from the surface of the metal body formed on the main surface up to 5 μm is observed. When calculating the porosity of the metal body used to calculate the filling rate of the resin cured material in the conductor described later, the observation points of the metal body can be appropriately set to be the same as the observation points of the conductor.

[0046] Furthermore, if the pressure applied during firing is within the above range, a special pressurizing device is not required, thus reducing voids, improving joint strength, and connection reliability without compromising yield. Methods for applying pressure to a silicon substrate coated with metal paste include, for example, placing weights on it, using a pressurizing device, or using a fixing jig for pressurizing.

[0047] The copper sintered body contained in the metal body may have a copper element ratio of 95% by mass or more, 97% by mass or more, 98% by mass or more, or 100% by mass among the constituent elements excluding light elements. If the above ratio of copper elements in the copper sintered body is within the above range, the formation of intermetallic compounds or the precipitation of dissimilar elements at the grain boundaries of the metallic copper crystal can be suppressed, the properties of the metallic copper constituting the copper sintered body tend to become stronger, and even better connection reliability can be obtained.

[0048] The amount of solder contained in the metal body may be 3 parts by mass or more, 10 parts by mass or more, or 10 parts by mass or more per 100 parts by mass of copper sintered body contained in the metal body, from the viewpoint of suppressing cracks in the metal body, or 20 parts by mass or less, 15 parts by mass or less, or 10 parts by mass or less, from the viewpoint of lowering the volume resistivity of the metal body.

[0049] In the metal body formation process, the metal paste may be heated and fired without pressurization. In this case, the porosity of the metal body formed on the main surface of the silicon substrate tends to increase, and the thermal expansion coefficient of the metal body decreases, making it less likely for the silicon substrate to crack or warp.

[0050] <Resin impregnation process> In this process, for example, the metal body 5 can be impregnated with the curable resin composition by applying the curable resin composition to the metal body-filled silicon substrate 50 obtained through the metal body formation process. In this embodiment, the curable resin composition is impregnated into the metal body 5 formed on both the main surface of the metal body 5 filling the through hole 30 and the silicon substrate 40. It is preferable that the voids 4 of the metal body 5 are sufficiently filled with the impregnated curable resin composition.

[0051] (Curable resin composition) Examples of components constituting a curable resin composition include thermosetting compounds. Examples of thermosetting compounds include oxetane compounds, epoxy compounds, episulfide compounds, (meth)acrylic compounds, phenol compounds, amino compounds, unsaturated polyester compounds, polyurethane compounds, silicone compounds, and polyimide compounds. Among these, epoxy compounds are particularly suitable because they further improve the curability and viscosity of the curable resin composition and enhance its properties and insulation reliability when stored at high temperatures.

[0052] The curable resin composition may further contain a thermosetting agent. Examples of thermosetting agents include imidazole curing agents, amine curing agents, phenol curing agents, polythiol curing agents, acid anhydrides, thermal cationic initiators, and thermal radical generators. These may be used individually or in combination of two or more. Of these, imidazole curing agents, polythiol curing agents, or amine curing agents are preferred because they can be cured rapidly at low temperatures. Furthermore, latent curing agents are preferred from the viewpoint of improving storage stability when the thermosetting compound and thermosetting agent are mixed. Latent curing agents are preferably latent imidazole curing agents, latent polythiol curing agents, or latent amine curing agents. The above thermosetting agents may be coated with a polymeric substance such as polyurethane resin or polyester resin.

[0053] The above-mentioned imidazole curing agents are not particularly limited and include 2-methylimidazole, 2-ethyl-4-methylimidazole, 1-cyanoethyl-2-phenylimidazole, 1-cyanoethyl-2-phenylimidazolium trimellitate, 2,4-diamino-6-[2'-methylimidazolyl-(1')]-ethyl-s-triazine, and 2,4-diamino-6-[2'-methylimidazolyl-(1')]-ethyl-s-triazine isocyanuric acid adduct.

[0054] The above-mentioned polythiol curing agent is not particularly limited and includes trimethylolpropane tris-3-mercaptopropionate, pentaerythritol tetrakis-3-mercaptopropionate, and dipentaerythritol hexa-3-mercaptopropionate. The solubility parameter of the polythiol curing agent is preferably 9.5 or higher, and preferably 12 or lower. The above solubility parameter is calculated by the Fedors method. For example, the solubility parameter of trimethylolpropane tris-3-mercaptopropionate is 9.6, and the solubility parameter of dipentaerythritol hexa-3-mercaptopropionate is 11.4.

[0055] The above-mentioned amine curing agents are not particularly limited and include hexamethylenediamine, octamethylenediamine, decamethylenediamine, 3,9-bis(3-aminopropyl)-2,4,8,10-tetraspiro[5.5]undecane, bis(4-aminocyclohexyl)methane, metaphenylenediamine, and diaminodiphenylsulfone.

[0056] Examples of the above-mentioned thermal cationic curing agents include iodonium-based cationic curing agents, oxonium-based cationic curing agents, and sulfonium-based cationic curing agents. Examples of the above-mentioned iodonium-based cationic curing agents include bis(4-tert-butylphenyl)iodonium hexafluorophosphate. Examples of the above-mentioned oxonium-based cationic curing agents include trimethyloxonium tetrafluoroborate. Examples of the above-mentioned sulfonium-based cationic curing agents include tri-p-tolylsulfonium hexafluorophosphate.

[0057] The above-mentioned thermal radical generator is not particularly limited and includes azo compounds and organic peroxides. Examples of azo compounds include azobisisobutyronitrile (AIBN). Examples of organic peroxides include di-tert-butyl peroxide and methyl ethyl ketone peroxide.

[0058] Methods for applying curable resin compositions include screen printing, transfer printing, offset printing, jet printing, dispensers, jet dispensers, needle dispensers, comma coaters, slit coaters, die coaters, gravure coaters, slit coats, letterpress printing, intaglio printing, gravure printing, stencil printing, soft lithography, bar coating, applicators, particle deposition methods, spray coaters, spin coaters, dip coaters, and the like.

[0059] The curable resin composition may be applied to one main surface of the metal-filled silicon substrate 50, or to a portion of the main surface. When applying the resin composition to both sides of the metal-filled silicon substrate 50, the resin composition may be applied to one main surface of the metal-filled silicon substrate 50, allowing the resin composition to penetrate to the main surface of the metal-filled silicon substrate 50 that was not coated, and then the resin composition may be applied to the main surface that was not coated. This allows the resin composition to spread throughout the pores 4.

[0060] By leaving the metal-filled silicon substrate 50 coated with a curable resin composition under a reduced pressure environment, the impregnation of the curable resin composition into the pores 4 of the metal body 5 can be improved.

[0061] In the resin impregnation process, it is preferable to impregnate the metal body with the cured resin composition such that the filling rate of the cured resin in the conductor formed after the resin curing process falls within the preferred range described later.

[0062] <Resin curing process> In this process, as shown in Figure 3(d1), a conductor 35 is formed by curing a curable resin composition (a curable resin composition filled in the voids 4) impregnated into a metal body 5, thereby forming a conductor 35 containing a metal body 5 with resin cured material 6 filled in the voids 4, and a substrate 51 having a silicon through-electrode is obtained, with silicon through-electrodes provided in through-holes 30. In this embodiment, a conductor 35 containing a metal body 5 with resin cured material 6 filled in the pores 4 is also provided on both main surfaces of the silicon substrate 40. Figure 3(d2) is a schematic diagram showing the structure of the conductor. The conductor includes a copper sintered body 12, solder 14, and resin cured material 6 filled in the voids. The solder 14 may be scattered within the conductor 35. The resin cured material 6 may be present in the pores of the copper sintered body 12, in the voids inside the solder 14, and in the voids between the copper sintered body 12 and the solder 14.

[0063] Curing of the curable resin composition can be performed by heat treatment. Heat treatment can be performed using heating means such as a hot plate, hot air dryer, hot air heating furnace, nitrogen dryer, infrared dryer, infrared heating furnace, far infrared heating furnace, microwave heating device, laser heating device, electromagnetic heating device, heater heating device, or steam heating furnace.

[0064] The atmosphere in the resin curing process may be an oxygen-free atmosphere from the viewpoint of suppressing oxidation of the metal body 5 (especially the copper sintered body), or a reducing atmosphere from the viewpoint of removing surface oxides from the metal body 5 (especially the copper sintered body). Examples of an oxygen-free atmosphere include the introduction of oxygen-free gases such as nitrogen or noble gases, or under vacuum. Examples of a reducing atmosphere include pure hydrogen gas, a mixed gas of hydrogen and nitrogen represented by foaming gas, nitrogen containing formic acid gas, a mixed gas of hydrogen and noble gases, and a noble gas containing formic acid gas.

[0065] The maximum temperature reached during the heat treatment in the resin curing process may be 150°C or higher, and may be 350°C or lower, 300°C or lower, or 260°C or lower, from the viewpoint of reducing thermal damage to each component and improving yield. If the maximum temperature reached is 150°C or higher, the resin composition tends to cure sufficiently when the maximum temperature is held for 60 minutes or less.

[0066] The conductor 35 formed in the resin curing process (conductor before the conductor removal process) may have a resin curing product content (filling rate) that satisfies the following conditions.

[0067] (Conductive material in through-holes) (a) In the region from point S1 to a depth of 10 μm (see (d1) in Figure 3) where a line L1 extending in the thickness direction of the substrate, passing through the central part C of the through hole 30 (the center of the hole length and the center of the hole diameter therein), intersects with the surface of the conductor 35, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores (vacancies) of the metal body. (b) In the region from point S1 to a depth of 10 to 20 μm, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body. (c) In the region from point S1 to a depth of 20 to 30 μm, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body. (d) In the range of ±5 μm in the thickness direction of the substrate and ±5 μm in the direction perpendicular to the thickness direction of the substrate from the central part C of the through hole 30, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body.

[0068] (Conductors on the main surface of the substrate) (e) In the region from the surface S2 of the conductor 35 formed on the main surface of the substrate to a depth of 5 μm (see (d1) in Figure 3), the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body. (f) In the region from the surface S2 of the conductor 35 formed on the main surface of the substrate to a depth of 10 μm, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body. (g) In the region from the surface S2 to a depth of 10 to 20 μm, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body. (h) In the region with a depth of 20 to 30 μm from the surface S2, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body.

[0069] The filling rate of the resin cured product 6 in the conductor 35 is calculated by the following procedure. (i) A focused ion beam is used to expose the cross-section of the conductor in the conductive silicon substrate (the cross-section in the thickness direction of the substrate). (ii) The exposed cross section is photographed using a scanning electron microscope to capture cross-sectional images (a range of 10 μm in the thickness direction of the substrate and a range of 10 μm in the direction perpendicular to the thickness direction of the substrate). (iii) The obtained cross-sectional image is binarized so that the metal portion and the resin cured portion are separated from the porous portion not filled by the resin cured material. (iv) From the binarized cross-sectional image, the ratio of the area of ​​the porous portion not filled by the cured resin to the total area of ​​the conductor cross-section is determined, and this is defined as the porosity of the conductor. (v) The porosity of the metal body before impregnation with the curable resin composition and the porosity of the conductor are substituted into the following formula (1) to calculate the filling rate of the cured resin in the conductor. Filling rate (%) of resin cured material in a conductor = [(BA) / B] × 100 ... Equation (1) [In formula (1), A represents the porosity (%) of the conductor, and B represents the porosity (%) of the metal.] When calculating the porosity of a conductor filled in a through-hole, in (i) above, the cross-section of the central part of the conductor in the through-hole is exposed. When calculating the porosity of a conductor formed on the main surface of a silicon substrate filled with conductor, in (i) above, the cross-section of the conductor on the main surface is exposed.

[0070] <Conductor removal process> In this process, at least a portion of the conductor 35 formed on the main surface of the silicon substrate 40 can be removed. Means for removing the conductor include chemical polishing, mechanical polishing, chemical-mechanical polishing, fly-cutting, and plasma treatment. Fly-cutting refers to planar cutting using a surface plane.

[0071] In this embodiment, from the viewpoint of being easily applicable using a general method, it is preferable that the removal means be one or more selected from the group consisting of etching, mechanical polishing, and chemical mechanical polishing.

[0072] The manufacturing method for a silicon substrate having silicon through-electrodes according to this embodiment includes a conductor removal step, which, for example, makes the surface of the conductor 35 formed on the main surface of the silicon substrate 40 flat, thus facilitating the formation of wiring.

[0073] In this embodiment, the filling rate of the resin cured product 6 in the conductor 35 after the conductor removal process may satisfy the following conditions. The filling rate can be calculated in the same manner as described above.

[0074] (Conductive material in through-holes) (a) In the region from point S3 to a depth of 10 μm (see Figure 4(e)) where a line L1 extending in the thickness direction of the substrate, passing through the central part C of the through hole 30 (the center of the hole length and the center of the hole diameter D therein), intersects with the surface of the conductor 35, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores (vacancies) of the metal body. (b) In the region from point S3 to a depth of 10 to 20 μm, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body. (c) In the region from point S3 to a depth of 20 to 30 μm, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body. (d) In the range of ±5 μm in the thickness direction of the substrate and ±5 μm in the direction perpendicular to the thickness direction of the substrate from the central part C of the through hole 30, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body.

[0075] (Conductors on the main surface of the substrate) (e) In the region from the surface S4 of the conductor 35 formed on the main surface of the substrate to a depth of 5 μm (see Figure 4(e)), the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores (vacancies) of the metal body. (f) In the region from the surface S4 of the conductor 35 formed on the main surface of the substrate to a depth of 10 μm, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body. (g) In the region from the surface S4 to a depth of 10 to 20 μm, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body. (h) In the region with a depth of 20 to 30 μm from the surface S4, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body.

[0076] The ratio L / D of the length L to the pore diameter D of the silicon through-electrode may be 1 or more, 5 or more, or 10 or more, or 15 or less, or 10 or less, or 5 or less, from the viewpoint of increasing the density of the resulting semiconductor device. The length L of the silicon through-electrode may be the thickness of the substrate having the silicon through-electrode. In this case, the ratio T / D of the thickness T of the substrate having the silicon through-electrode to the pore diameter D of the silicon through-electrode may be within the above range.

[0077] In the manufacturing method of the substrate having through electrodes of this embodiment, a resin curing material is filled into the conductive body, but the resin impregnation step and resin curing step described above may be omitted, and the through electrodes may be formed from a metal body.

[0078] The method for manufacturing a substrate having through electrodes according to this embodiment may further include a wiring formation step. The wiring formation step may include a resist formation step, an etching step, and a resist removal step, which will be described below.

[0079] <Resist Formation Process> In the resist formation process, as shown in Figure 4(f), an etching resist 8 is formed on a conductor 35 formed on the main surface of the silicon substrate 40.

[0080] Methods for forming the etching resist 8 include, for example, a method of silkscreen printing the resist ink, or a method of laminating a negative-type photosensitive dry film for etching resist onto copper foil, placing a photomask that transmits light in the shape of the wiring on top of it, exposing it with ultraviolet light, and removing the unexposed areas with a developer.

[0081] <Etching process> In the etching process, as shown in Figure 5(g), the conductor 35 in the portion not covered by the etching resist 8 is removed by etching. In this embodiment, a portion of the metal coating 2 provided on both main surfaces of the silicon wafer 1 is removed by etching.

[0082] Etching methods include, for example, using chemical etching solutions commonly used for printed circuit boards, such as a solution of cupric chloride and hydrochloric acid, a ferric chloride solution, a solution of sulfuric acid and hydrogen peroxide, or an ammonium persulfate solution.

[0083] <Resist Removal Process> In the resist removal process, the etching resist 8 formed on the conductive material 35 is removed.

[0084] The method for manufacturing a substrate having through electrodes according to this embodiment further includes a wiring formation step having the above-described step, thereby enabling the formation of wiring 9 including a conductor 35 on the main surface of the silicon substrate 40.

[0085] (Substrate having through electrodes according to the first embodiment) Figure 5(h) is a cross-sectional view showing one embodiment of a substrate having a silicon through-electrode that can be manufactured by the method according to the first embodiment described above. The substrate 52 having a silicon through-electrode shown in Figure 5(h) includes a silicon wafer 1 having through-holes 30, and comprises a silicon substrate with the through-holes 30 extending through both main surfaces, and a conductor that fills the through-holes 30. The conductor 35 includes a copper sintered body having a porous structure, solder scattered on the copper sintered body, and a metal body 5 having voids 4, the metal body 5 including voids 4, which are voids inside the solder and / or voids between the solder and the copper sintered body. The conductor 35 further comprises a resin cured product 6 that fills the voids 4.

[0086] In the substrate 52 having silicon through-electrodes shown in Figure 5(h), a metal coating 2 is provided on both main surfaces of the silicon wafer 1 and on the walls of the through-holes. However, the metal coating 2 does not have to be provided on the main surfaces, may be provided on only one main surface, or may not be provided on the walls of the through-holes. Furthermore, in the substrate 52 having silicon through-electrodes, wiring 9 comprising the metal coating 2 and conductor 35 is provided on both main surfaces of the silicon substrate 40. However, the wiring 9 may be provided on only one main surface of the silicon substrate 40.

[0087] The substrate 52 having silicon through-electrodes may also satisfy the following conditions regarding the filling rate of the resin cured material 6 in the conductor. The filling rate can be calculated in the same manner as described above.

[0088] (Conductive material in through-holes) (a) In the region from point S5 to a depth of 10 μm (see Figure 5(h)) where a line L1 extending in the thickness direction of the substrate, passing through the central part C of the through hole 30 (the center of the hole length and the center of the hole diameter therein), intersects with the surface of the conductor, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores (vacancies) of the metal body. (b) In the region from point S5 to a depth of 10 to 20 μm, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body. (c) In the region from point S5 to a depth of 20 to 30 μm, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body. (d) In the range of ±5 μm in the thickness direction of the substrate and ±5 μm in the direction perpendicular to the thickness direction of the substrate from the central part C of the through hole 30, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body.

[0089] (Conductors on the main surface of the substrate) (e) In the region from the surface S6 of the conductor formed on the main surface of the substrate to a depth of 5 μm (see Figure 5(h)), the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body. (f) In the region from the surface S6 of the conductor 35 formed on the main surface of the substrate to a depth of 10 μm, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body. (g) In the region from the surface S6 to a depth of 10 to 20 μm, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body. (h) In the region with a depth of 20 to 30 μm from the surface S6, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body.

[0090] (Semiconductor device according to the first embodiment) A semiconductor device manufactured using a substrate having through electrodes (a substrate having silicon through electrodes) according to the first embodiment will be specifically described with reference to Figure 6. Figure 6 is a schematic cross-sectional view showing one embodiment of the semiconductor device of the present invention. In the semiconductor device 100 shown in Figure 6(a), the interposer substrate 25 and the substrate having silicon through electrodes are connected via a flip-chip connection by directly connecting the wiring 27 on the interposer substrate 25 to the conductor 35 of the substrate 52 having silicon through electrodes. The gap between the interposer substrate 25 and the substrate 52 having silicon through electrodes is filled without gaps with cured adhesive 20 and sealed. On the main surface of the substrate 52 having silicon through electrodes opposite to the interposer substrate 25, substrates 52 having silicon through electrodes are repeatedly stacked. The substrates 52 having silicon through electrodes are connected to each other by the conductor 35. The gaps between the substrates 52 having silicon through electrodes are filled without gaps with cured adhesive 20 and sealed.

[0091] The semiconductor device 100 may be obtained, for example, by the following method. That is, a laminate is obtained by laminating substrates 52 having silicon through electrodes via an adhesive. The adhesive may be cured during lamination. The obtained laminate and the interposer substrate 25 are electrically connected by pressing them together, forming a connection between the laminate and the interposer substrate 25. A dicing tape is attached to the side of the formed connection opposite to the side on which the interposer substrate 25 is provided, and dicing is performed along the dicing line to obtain the semiconductor device 100.

[0092] In the semiconductor device 200 shown in Figure 6(b), the wiring 27 on the interposer substrate 25 and the conductor 35 of the substrate 52 having silicon through electrodes are connected via fine bumps 15, thereby creating a flip-chip connection between the interposer substrate 25 and the substrate 52 having silicon through electrodes. The gap between the interposer substrate 25 and the substrate 52 having silicon through electrodes is filled and sealed with cured adhesive 20. On the main surface of the substrate 52 having silicon through electrodes opposite to the interposer substrate 25, substrates 52 having silicon through electrodes are repeatedly stacked via fine bumps 15. The gaps between the substrates 52 having silicon through electrodes are filled and sealed with cured adhesive 20.

[0093] The semiconductor device 200 may be obtained, for example, by the following method. That is, a laminate is obtained by laminating a substrate 52 having silicon through-electrodes with fine bumps 15 provided on one main surface via an adhesive. The adhesive may be cured at the time of lamination. The obtained laminate and the interposer substrate 25 are electrically connected by pressing them together to form a connection between the laminate and the interposer substrate 25. A dicing tape is attached to the side of the formed connection opposite to the side on which the interposer substrate 25 is provided, and dicing is performed along the dicing line to obtain the semiconductor device 200.

[0094] (Method for manufacturing a substrate having a through electrode according to the second embodiment) Figures 7 and 8 are schematic diagrams illustrating a method for manufacturing a substrate having a through-electrode according to the second embodiment, specifically a method for manufacturing a substrate having a through-electrode, which involves manufacturing a silicon through-electrode.

[0095] A method for manufacturing a substrate having through electrodes according to the second embodiment includes a preparation step of preparing a substrate having non-through holes, wherein the non-through holes are open on one main surface, the substrate includes an insulating substrate having non-through holes, a conductor forming step of forming a conductor in the non-through holes, and a grinding step of providing through electrodes by grinding the surface of the substrate on which the conductor has been formed that is opposite to the surface where the non-through holes are open, wherein the conductor forming step includes a metal body forming step of forming a metal body having voids, containing a copper sintered body having a porous structure and solder so as to fill at least the non-through holes, a resin impregnation step of impregnating the metal body with a curable resin composition, and a resin curing step of curing the curable resin composition impregnated into the metal body.

[0096] The method according to the second embodiment will also be described using the same method as the first embodiment, with a silicon wafer as the substrate and a silicon through-electrode being manufactured as the substrate. However, the following description can be interpreted by substituting the silicon wafer with other insulating substrates.

[0097] <Substrate preparation process> In this process, as shown in Figure 7(a), a silicon wafer 1 having non-through holes 31 and a silicon substrate 41 having the walls and bottom of the non-through holes 31 and a metal coating 2 provided on the surface of the silicon wafer 1 can be prepared. The non-through holes 31 open to one main surface of the silicon wafer 1.

[0098] The thickness of the silicon wafer 1 may be 20 μm or more, 30 μm or more, or 50 μm or more, and may be 500 μm or less, 400 μm or less, or 300 μm or less, from the viewpoint of suppressing warping of the substrate after sintering.

[0099] The upper limit of the hole diameter of the non-through holes 31 may be 200 μm or less, 100 μm or less, or 60 μm or less, from the viewpoint of increasing the density of the resulting semiconductor device, and the lower limit of the hole diameter of the non-through holes 31 is not particularly limited, but may be 10 μm or more, or 30 μm or more.

[0100] The length (depth) of the non-through hole 31 can be appropriately set according to the length of the silicon through electrode to be formed.

[0101] The metal coating 2 may be provided on the surface of the silicon wafer 1 where the non-through holes 31 are open, as well as on the walls and bottom surfaces of the non-through holes 31, or it may not be provided on the walls and bottom surfaces of the non-through holes 31. In the embodiment shown in Figure 7(a), the silicon substrate 41 has the metal coating 2 on the surface of the silicon wafer 1 where the non-through holes 31 are open, as well as on the walls and bottom surfaces of the non-through holes 31.

[0102] As the metal coating 2, the same material as in the first embodiment can be used.

[0103] <Metal body formation process> In this step, a porous metal body containing a copper sintered body and solder is formed to fill at least the non-through holes. In this embodiment, the metal body may be formed to cover at least a portion of the surface of the silicon substrate 41 where the non-through holes 31 are open. In this case, a conductor is formed to fill the non-through holes 31 of the silicon substrate 41, and a conductor can also be provided on the surface of the silicon substrate 41 where the non-through holes 31 are open.

[0104] The metal body formation process may include a metal paste filling step in which a metal paste containing copper particles and solder particles is filled into non-through holes in a silicon substrate, and a metal paste firing step in which the metal paste is fired to form a metal body. When forming a metal body on a surface of the silicon substrate where non-through holes are open, a layer of metal paste can be provided on the surface of the silicon substrate where non-through holes are open during or after the metal paste filling step.

[0105] As part of the metal body formation process described above, for example, as shown in Figure 7(b), a metal paste 3 containing copper particles and solder particles is applied to a silicon substrate 41, the metal paste 3 is filled into the non-through holes 31, and a layer of metal paste 3 is also provided on the surface of the silicon substrate 41 where the non-through holes 31 are open. Details of the metal paste 3 will be described later.

[0106] A method for applying the metal paste 3 to the silicon substrate 41 is the same as in the first embodiment.

[0107] When metal paste is applied to the surface of the silicon substrate 41 where the non-through holes 31 are open, the thickness of the metal paste layer may be 30 μm or less, 20 μm or less, 15 μm or less, or 10 μm or less, from the viewpoint of suppressing warping of the silicon substrate 41 and reducing the burden in the conductive material removal process described later.

[0108] The metal paste 3 may be dried as appropriate, similar to the first embodiment. When drying the metal paste 3, the drying atmosphere, drying method, and drying temperature may be the same as in the first embodiment.

[0109] After the metal paste filling process, the metal paste 3 is fired to sinter the copper particles contained in the metal paste 3. In this way, as shown in Figure 8(c), a metal-filled silicon substrate 60 is obtained in which the non-through holes 31 are filled with metal bodies 5 containing porosity 4, i.e., having a porous structure. In this embodiment, a metal-filled silicon substrate 60 is obtained in which metal bodies 5 are also provided on the surface of the silicon substrate 41 where the non-through holes 31 are open.

[0110] The firing conditions may be the same as those in the first embodiment.

[0111] The porosity of the metal body formed on the surface of the silicon substrate where non-through holes are open may be the same as the porosity of the metal body formed on the main surface of the silicon substrate in the first embodiment. The porosity of the metal body can be calculated using the same procedure as in the first embodiment.

[0112] The porosity of the metal body filling the non-through holes may be the same as that of the metal body filling the through holes in the first embodiment.

[0113] The proportion of copper elements among the constituent elements, excluding light elements, in the copper sintered body contained in the metal body may be the same as in the first embodiment. Similarly, the solder content in the metal body may be the same as in the first embodiment.

[0114] <Resin impregnation process> In this process, for example, the metal body 5 can be impregnated with the curable resin composition by applying the curable resin composition to the metal body-filled silicon substrate 60 obtained through the metal body formation process. In this embodiment, the curable resin composition is impregnated into the metal body 5 that fills the non-through holes 31 and into the metal body 5 formed on the surface of the silicon substrate 41 where the non-through holes are open. It is preferable that the voids 4 of the metal body 5 are sufficiently filled with the impregnated curable resin composition.

[0115] As the curable resin composition, the same one as in the first embodiment can be used. The method for applying the curable resin composition is the same as in the first embodiment. The filling rate of the curable resin composition into the pores 4 of the metal body 5 can be appropriately changed according to the filling rate of the resin cured product 6 of the conductor 35 after the resin curing process.

[0116] <Resin curing process> In this process, a conductor 35 is formed by curing the curable resin composition (the curable resin composition filled in the pores 4) impregnated into the metal body 5, thereby curing the metal body 5 with the cured resin 6 filled in the pores 4. In this embodiment, a conductor 35, which includes the metal body 5 with the cured resin 6 filled in the pores 4, is also provided on the surface of the silicon substrate 41 where the non-through holes are open.

[0117] The curing conditions for the resin composition in the resin curing process may be the same as those in the first embodiment.

[0118] The conductor formed in the resin curing process (conductor before the conductor removal process and grinding process) may have its resin curing content adjusted so that the resin curing content in the silicon through-electrode being formed meets the conditions described below. For example, the conductor may have a resin curing content that satisfies the following conditions.

[0119] (Conductive material with non-through holes) (a) In the region from the point where a line extending in the thickness direction of the substrate, passing through the center of the non-through hole (the center of the hole length and the center of the hole diameter therein), intersects with the surface of the conductor, to a depth of 10 μm, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body. (b) In the region from the point where a line passing through the center of the non-through hole and extending in the thickness direction of the substrate intersects with the surface of the conductor, to a depth of 10 to 20 μm, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores (vacancies) of the metal body. (c) In a region 20 to 30 μm deep from the point where a line passing through the center of the non-through hole and extending in the thickness direction of the substrate intersects with the surface of the conductor, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body.

[0120] (Conductors on the main surface of the substrate) (e) In the region from the surface of the conductor formed on the main surface of the substrate to a depth of 5 μm, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body. (f) In the region from the surface of the conductor formed on the main surface of the substrate to a depth of 10 μm, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body. (g) In a region 10 to 20 μm deep from the surface of the conductor formed on the main surface of the substrate, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body. (h) In the region of the conductor formed on the main surface of the substrate at a depth of 20 to 30 μm from the surface, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body.

[0121] The filling rate of the resin cured product 6 in the conductor is calculated in the same manner as described in the procedure for manufacturing a substrate having a silicon through-electrode according to the first embodiment.

[0122] <Conductor removal process> The manufacturing method for a substrate having a silicon through-electrode according to this embodiment may further include a conductor removal step after the resin curing step. In this step, at least a portion of the conductor 35 formed on one main surface of the silicon substrate 41 can be removed. The means for removing the conductor may be the same as in the first embodiment.

[0123] <Grinding process> In this process, as shown in Figure 8(d), a substrate 61 having a silicon through-electrode with a silicon electrode is obtained by grinding the side of the silicon substrate opposite to the side where the non-through-hole 31 is open, on which the conductor 35 is formed. In other words, in this process, grinding exposes the conductor 35 on the side of the silicon substrate opposite to the side where the non-through-hole 31 is open, thereby forming a silicon through-electrode. In Figure 8(d), the conductor 35 formed on the side of the silicon substrate where the non-through-hole 31 is open is removed by the conductor removal process.

[0124] Grinding methods include, for example, mechanical polishing and chemical mechanical polishing.

[0125] In this embodiment, the filling rate of the resin cured product 6 in the conductor 35 after the grinding process may satisfy the following conditions. The filling rate can be calculated in the same manner as described above.

[0126] (a) In the region from point S21 to a depth of 10 μm (see Figure 8(d)) where a line L2 extending in the thickness direction of the substrate, passing through the central part E of the silicon through electrode (the center in length L and the center of the hole diameter D therein), intersects with the surface of the conductor 35, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores (vacancies) of the metal body. (b) In the region from point S21 to a depth of 10 to 20 μm, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body. (c) In the region from point S21 to a depth of 20 to 30 μm, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body. (d) In a range of ±5 μm in the thickness direction of the substrate and ±5 μm in the direction perpendicular to the thickness direction of the substrate from the central part E of the silicon through electrode, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body.

[0127] The ratio L / D of the length L to the pore diameter D of the silicon through-electrode may be the same as in the first embodiment.

[0128] In the manufacturing method of the substrate having through electrodes of this embodiment, a resin curing material is filled into the conductive body, but the resin impregnation step and resin curing step described above may be omitted, and the through electrodes may be formed from a metal body.

[0129] Furthermore, the method for manufacturing a substrate having through electrodes according to the second embodiment may further include a wiring formation step similar to that of the method according to the first embodiment described above.

[0130] (Substrate having through electrodes according to the second embodiment) Figure 8(d) is a cross-sectional view showing one embodiment of a substrate having a silicon through-electrode that can be manufactured by the method according to the second embodiment described above. The substrate 61 having a silicon through-electrode shown in Figure 8(d) includes a silicon wafer 1 having through-holes, and comprises a silicon substrate with through-holes extending through both main surfaces, and a conductor 35 that fills the through-holes. The conductor 35 includes a copper sintered body having a porous structure, solder scattered on the copper sintered body, and a metal body 5 having voids 4, the metal body 5 including voids 4 which are voids inside the solder and / or voids between the solder and the copper sintered body. The conductor 35 further comprises a resin cured product 6 that fills the voids 4.

[0131] In the substrate 61 having silicon through-electrodes shown in Figure 8(d), the conductor 35 is not formed on both main surfaces, but the conductor 35 may be formed on one of the main surfaces. In the substrate 61 having silicon through-electrodes shown in Figure 8(d), a metal coating 2 is provided on the wall surface of the through-hole, but the metal coating 2 does not have to be provided on the wall surface of the through-hole.

[0132] The filling rate of the resin cured product 6 in the conductor 35 contained in the substrate 61 having a silicon through-electrode may be the same as the filling rate of the resin cured product 6 in the conductor 35 after the grinding process.

[0133] (Semiconductor device according to the second embodiment) A semiconductor device manufactured using a substrate having through electrodes (a substrate having silicon through electrodes) according to the second embodiment will be specifically described with reference to Figure 9. Figure 9 is a schematic cross-sectional view showing one embodiment of the semiconductor device of the present invention. In the semiconductor device 300 shown in Figure 9, substrates 61 having silicon through electrodes are repeatedly stacked. The substrates 61 having silicon through electrodes are electrically connected to each other.

[0134] The semiconductor device 300 may have a filling rate of the resin cured product 6 in the conductor 35 that satisfies the following conditions. The filling rate can be calculated in the same manner as described above.

[0135] (Conductor for through-silicon electrodes) (a) In the region from point S22 to a depth of 10 μm (see Figure 9) where a line L2 extending in the thickness direction of the substrate, passing through the central part E of the silicon through electrode (the center in length L and the center of the hole diameter D therein), intersects with the surface of the conductor 35, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores (vacancies) of the metal body. (b) In the region from point S22 to a depth of 10 to 20 μm, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body. (c) In the region from point S22 to a depth of 20 to 30 μm, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body. (d) In a range of ±5 μm in the thickness direction of the substrate and ±5 μm in the direction perpendicular to the thickness direction of the substrate from the central part E of the silicon through electrode, the filling rate of the cured resin may be 80 volume% or more, 90 volume% or more, or 95 volume% or more, based on the total volume of the internal space of the pores of the metal body.

[0136] (Metal paste) A metal paste containing copper particles and solder particles, used in the manufacturing method of a substrate having through electrodes according to the first and second embodiments, will be described.

[0137] The metal paste may contain, for example, first copper particles having a particle size (maximum diameter) of 0.8 μm or more.

[0138] The particle size (maximum diameter) of the first copper particle may be 1.2 μm or larger. The particle size (maximum diameter) of the first copper particle may be 10 μm or smaller, or 8.0 μm or smaller.

[0139] The average particle size (average maximum diameter) of the first copper particles contained in the metal paste may be 0.8 μm or more, 1.2 μm or more, 10 μm or less, or 8 μm or less, from the viewpoint of improving the sintering density in the through-holes and suppressing the formation of voids in the through-holes.

[0140] The particle size (maximum diameter) and average particle size (average maximum diameter) of the first copper particles can be determined, for example, from the SEM image of the particles. An example of how to calculate the particle size (maximum diameter) of the first copper particles from an SEM image is given below. The powder of the first copper particles is placed on a carbon tape for SEM with a spatula to create an SEM sample. This SEM sample is observed at 5000x magnification using an SEM device. A rectangle circumscribing the first copper particle in the SEM image is drawn using image processing software, and the longer side of the rectangle is taken as the particle size (maximum diameter). This measurement is performed on 50 or more first copper particles using multiple SEM images, and the average value of the particle size (average maximum diameter) is calculated.

[0141] The shape of the first copper particles may be, for example, spherical, lumpy, needle-shaped, flattened (flake-shaped), or nearly spherical. The first copper particles may also be aggregates of copper particles having these shapes.

[0142] The first copper particles are preferably flattened (flake-shaped). In this case, the orientation of the first copper particles is substantially parallel to the coating surface of the metal paste, which suppresses volume shrinkage when the copper particles in the metal paste are sintered, making it easier to suppress voids that occur in through holes. Furthermore, by suppressing volume shrinkage when the copper particles in the metal paste are sintered, cracks in the metal body formed on at least one main surface of the substrate (e.g., a silicon substrate) can be suppressed.

[0143] The aspect ratio of the first copper particles may be 4 or greater, or 6 or greater. If the aspect ratio is within the above range, the first copper particles in the metal paste are more likely to be oriented parallel to the coating surface of the metal paste, and volume shrinkage when the copper particles in the metal paste are sintered can be suppressed. This further suppresses wire breakage due to thermal stress when wiring is formed from a conductor provided on the main surface of a substrate (e.g., a silicon substrate). In addition, the adhesion between the metal body and the metal film formed on the substrate (e.g., a silicon substrate) can be improved. The aspect ratio (major axis / thickness) of the copper particles in the metal paste can be determined, for example, by observing an SEM image of the particles and measuring the major axis and thickness.

[0144] The metal paste preferably contains first copper particles having a particle size (maximum diameter) of 0.8 μm or more and 10 μm or less, and an aspect ratio of 4 or more. By including such first copper particles in the metal paste, the volume shrinkage when the copper particles in the metal paste are sintered can be sufficiently reduced, making it easy to form a metal body with a porous structure and a sufficiently formed conductive network within through-holes or non-through-holes. As a result, the generation of voids in through-holes or non-through-holes can be suppressed, a metal body that is less prone to cracking on the main surface of the substrate (e.g., a silicon substrate) can be formed, and when wiring is formed from a conductor containing this metal body, disconnection due to thermal stress of the wiring can be further suppressed.

[0145] The metal paste may contain copper particles having a particle size (maximum diameter) of 0.8 μm or more and 10 μm or less, and an aspect ratio of less than 2. However, the content of copper particles having a particle size (maximum diameter) of 0.8 μm or more and 10 μm or less, and an aspect ratio of less than 2, is preferably 50 parts by mass or less, and more preferably 30 parts by mass or less, per 100 parts by mass of first copper particles having a particle size (maximum diameter) of 0.8 μm or more and 10 μm or less, and an aspect ratio of 4 or more. By limiting the content of copper particles having a particle size (maximum diameter) of 0.8 μm or more and 10 μm or less, and an aspect ratio of less than 2, it is possible to form a metal body with a porous structure while having a sufficiently formed conductive network in the through-hole or non-through-hole, while suppressing the generation of voids in the through-hole by the first copper particles in the metal paste. Furthermore, on the main surface of the substrate (e.g., a silicon substrate), the first copper particles tend to orient themselves approximately parallel to the surface on which the metal paste is applied, thereby more effectively suppressing volume shrinkage and forming a metal body that is less prone to cracking. When wiring is formed from a conductor containing this metal body, disconnection due to thermal stress on the wiring can be further suppressed.

[0146] To make such effects even easier to obtain, the content of copper particles having a particle size (maximum diameter) of 0.8 μm or more and 10 μm or less and an aspect ratio of less than 2 may be 20 parts by mass or less, 10 parts by mass or less, or 0 parts by mass, per 100 parts by mass of the first copper particles having a particle size (maximum diameter) of 0.8 μm or more and 10 μm or less and an aspect ratio of 4 or more.

[0147] The content of the first copper particles in the metal paste may be 15% by mass or more, 20% by mass or more, or 50% by mass or more, and 85% by mass or less, 70% by mass or less, or 50% by mass or less, based on the total mass of metal particles contained in the metal paste. If the content of the first copper particles is within the above range, the effects described above will be more easily obtained.

[0148] The first copper particles may be treated with a surface treatment agent from the viewpoints of dispersion stability and oxidation resistance. The surface treatment agent may be removed during wiring formation (when the copper particles are sintered). Examples of such surface treatment agents include aliphatic carboxylic acids such as palmitic acid, stearic acid, arachidic acid, and oleic acid; aromatic carboxylic acids such as terephthalic acid, pyromellitic acid, and o-phenoxybenzoic acid; aliphatic alcohols such as cetyl alcohol, stearyl alcohol, isobornyl cyclohexanol, and tetraethylene glycol; aromatic alcohols such as p-phenylphenol; alkylamines such as octylamine, dodecylamine, and stearylamine; aliphatic nitriles such as stearonitrile and decanenitrile; silane coupling agents such as alkylalkoxysilane; and polymer treatment agents such as polyethylene glycol, polyvinyl alcohol, polyvinylpyrrolidone, and silicone oligomer. The surface treatment agent may be used alone or in combination of two or more kinds.

[0149] The treatment amount of the surface treatment agent may be an amount of one molecular layer or more on the particle surface. Such a treatment amount of the surface treatment agent varies depending on the specific surface area of the first copper particles, the molecular weight of the surface treatment agent, and the minimum coating area of the surface treatment agent. The treatment amount of the surface treatment agent is usually 0.001 mass% or more.

[0150] The treatment amount of the surface treatment agent is related to the number of molecular layers (n) adhered to the surface of the first copper particles, the specific surface area (A p )(unit: m 2 / g) of the first copper particles, the molecular weight (M s )(unit: g / mol) of the surface treatment agent, the minimum coating area (S S )(unit: m 2 / particle) of the surface treatment agent, and Avogadro's number (N A )(6.02×10 23 particles) and can be calculated therefrom. Specifically, the treatment amount of the surface treatment agent (mass%) = {(n·A p ·M s ) / (S S ·N A + n·A p ·M sIt is calculated according to the formula )} × 100%.

[0151] The specific surface area of ​​the first copper particle can be calculated by measuring the specific surface area of ​​the dried copper particle using the BET specific surface area measurement method. The minimum coating area of ​​the surface treatment agent is 2.05 × 10⁻⁶ when the surface treatment agent is a linear saturated fatty acid. -19 m 2 It is / 1 molecule. For other surface treatment agents, it can be measured, for example, by calculation from a molecular model, or by the method described in "Chemistry and Education" (Katsuhiro Ueeda, Sumio Inafuku, Iwao Mori, 40(2), 1992, pp. 114-117). An example of a quantitative method for surface treatment agents is shown. Surface treatment agents can be identified by thermal desorption gas / gas chromatography-mass spectrometry of the dried powder obtained by removing the dispersion medium from the metal paste, thereby determining the carbon number and molecular weight of the surface treatment agent. The carbon content of the surface treatment agent can be analyzed by carbon content analysis. An example of a carbon content analysis method is high-frequency induction heating furnace combustion / infrared absorption spectroscopy. From the identified carbon number, molecular weight and carbon content of the surface treatment agent, the amount of surface treatment agent can be calculated using the above formula.

[0152] Commercially available copper particles can be used as the first copper particles. Examples of commercially available first copper particles include MA-C025 (manufactured by Mitsui Mining & Smelting Co., Ltd., average particle size 4.1 μm), 3L3 (manufactured by Fukuda Metal Foil & Powder Industry Co., Ltd., average particle size 7.3 μm), 1110F (manufactured by Mitsui Mining & Smelting Co., Ltd., average particle size 5.8 μm), and 2L3 (manufactured by Fukuda Metal Foil & Powder Industry Co., Ltd., average particle size 9 μm).

[0153] When manufacturing metal paste, it is possible to use a paste that contains first copper particles having a particle size (maximum diameter) of 0.8 μm or more and 10 μm or less and an aspect ratio of 4 or more, and in which case the content of copper particles having a particle size (maximum diameter) of 0.8 μm or more and 10 μm or less and an aspect ratio of less than 2 is 50 parts by mass or less, preferably 30 parts by mass or less, per 100 parts by mass of the first copper particles having a particle size (maximum diameter) of 0.8 μm or more and 10 μm or less and an aspect ratio of 4 or more. Commercially available products consisting of such copper particles may be selected and used.

[0154] The ratio of the pore diameter of the through-hole to the particle size (maximum diameter) of the first copper particle (pore diameter (μm) / particle size (μm)) may be 4 or more, 8 or more, or 10 or more, and may be 150 or less, 100 or less, or 50 or less, from the viewpoint of suppressing volume shrinkage and forming a metal body that is less prone to cracking. For non-through-holes, the ratio of the pore diameter of the non-through-hole to the particle size (maximum diameter) of the first copper particle (pore diameter (μm) / particle size (μm)) can be set so that the ratio of the pore diameter of the non-through-hole to the particle size (maximum diameter) of the first copper particle (pore diameter (μm) / particle size (μm)) falls within the above range.

[0155] In one embodiment, the metal paste may contain the first copper particles described above and second copper particles having a particle size (maximum diameter) of 0.5 μm or less. In this case, when the copper particles are sintered, the interposition of the second copper particles between the first copper particles tends to improve the conductivity of the resulting wiring. That is, it is preferable to use both the first and second copper particles. When preparing a metal paste containing only the second copper particles as the copper particles, the volume shrinkage and sintering shrinkage associated with the drying of the dispersion medium are large, making it easy for the sintered body to peel off from the metal film provided on the insulating substrate (e.g., a silicon wafer) when the copper particles are sintered, making it difficult to obtain sufficient airtightness and connection reliability. However, by using both the first and second copper particles, the volume shrinkage when the metal paste is sintered is suppressed, and the adhesion between the metal body formed in the through-hole and the metal film formed on the wall surface of the through-hole or non-through-hole can be improved. This makes fracture due to thermal stress in the copper sintered body within the through-hole less likely, further improving airtightness and connection reliability against thermal stress.

[0156] The second copper particles can act as copper particles that suitably bond the first copper particles together. Furthermore, the second copper particles have superior sinterability compared to the first copper particles and can have the function of promoting the sintering of the copper particles. For example, compared to using the first copper particles alone, it becomes possible to sinter the copper particles at a lower temperature. Also, when preparing a metal paste containing only the second copper particles, the volume shrinkage and sintering shrinkage associated with the drying of the dispersion medium are large, so the metal body formed inside the through-holes or non-through-holes tends to shrink in volume, easily creating voids inside the through-holes or non-through-holes. In particular, by using flattened first copper particles and second copper particles in combination, the flattened first copper particles act as copper particles that are suitably bonded by the second copper particles, thereby making it easier to form a metal body with voids (porous structure) while suppressing the generation of voids inside the through-holes.

[0157] The average particle size (average maximum diameter) of the second copper particles contained in the metal paste may be 0.01 μm or more, 0.03 μm or more, 0.05 μm or more, or 0.08 μm or more, and may be 0.5 μm or less, 0.4 μm or less, 0.3 μm or less, or 0.2 μm or less.

[0158] If the average particle size (average maximum diameter) of the second copper particles is 0.01 μm or larger, it becomes easier to obtain effects such as reduced synthesis costs for the second copper particles, good dispersibility, and reduced use of surface treatment agents. If the average particle size (average maximum diameter) of the second copper particles is 0.5 μm or smaller, it becomes easier to obtain the effect of excellent sinterability of the second copper particles. From the viewpoint of achieving the above effects even more effectively, the average particle size (average maximum diameter) of the second copper particles may be 0.05 μm or larger, 0.1 μm or larger, or 0.2 μm or larger, and may be 0.5 μm or smaller, 0.4 μm or smaller, or 0.3 μm or smaller.

[0159] The second copper particles may contain 20% by mass or more of copper particles with a particle size (maximum diameter) of 0.01 μm or more and 0.5 μm or less. From the viewpoint of the sinterability of the metal paste, the second copper particles may contain 30% by mass or more, 50% by mass or more, or 85% by mass or less of copper particles with a particle size of 0.01 μm or more and 0.5 μm or less. When the content of copper particles with a particle size (maximum diameter) of 0.01 μm or more in the second copper particles is 20% by mass or more, the dispersibility of the copper particles is further improved, and the increase in viscosity and decrease in paste concentration can be further suppressed.

[0160] The content of the second copper particles in the metal paste may be 20% by mass or more, 30% by mass or more, 35% by mass or more, or 40% by mass or more, and may be 85% by mass or less, 80% by mass or less, or 75% by mass or less, based on the total mass of metal particles contained in the metal paste. If the content of the second copper particles is within the above range, it becomes easier to form a metal body with excellent adhesion to a metal coating provided on a substrate (e.g., a silicon substrate) while suppressing the occurrence of voids in through holes or non-through holes, and it is possible to form a copper sintered body that is less prone to cracking on the main surface of the substrate (e.g., a silicon substrate), and when wiring is formed from a conductor containing this copper sintered body, it is possible to further suppress wire breakage due to thermal stress of the wiring.

[0161] The content of the second copper particles in the metal paste may be 20% by mass or more and 85% by mass or less, based on the sum of the masses of the first copper particles and the second copper particles. If the content of the second copper particles is 20% by mass or more, the spaces between the first copper particles can be sufficiently filled, forming a metal body that is less prone to cracking, and wiring formed from a conductor containing this metal body will be less prone to disconnection due to thermal stress. If the content of the second copper particles is 85% by mass or less, the volume shrinkage when the copper particles are sintered can be sufficiently suppressed, thereby suppressing the generation of voids in through holes or non-through holes, and forming a metal body that is less prone to cracking, and wiring formed from a conductor containing this metal body will be less prone to disconnection due to thermal stress.

[0162] From the viewpoint of making the above effects easier to obtain, the content of the second copper particles may be 30% by mass or more, 35% by mass or more, or 40% by mass or more, or 85% by mass or less, or 80% by mass or less, based on the sum of the mass of the first copper particles and the mass of the second copper particles.

[0163] The shape of the second copper particles may be, for example, spherical, lumpy, needle-shaped, flattened (flake-shaped), or approximately spherical. The second copper particles may also be aggregates of copper particles having these shapes. From the viewpoint of dispersibility and packing, the shape of the second copper particles may be spherical, approximately spherical, or flattened (flake-shaped), and from the viewpoint of flammability and miscibility with the first copper particles, they may be spherical or approximately spherical.

[0164] The aspect ratio of the second copper particles may be 5 or less, or 3 or less, from the viewpoint of dispersibility, packing, and miscibility with the first copper particles.

[0165] The second copper particles may be treated with a specific surface treatment agent. Examples of specific surface treatment agents include organic acids having 8 to 16 carbon atoms. Examples of organic acids having 8 to 16 carbon atoms include caprylic acid, methylheptanoic acid, ethylhexanoic acid, propylpentanoic acid, pelargonic acid, methyloctanoic acid, ethylheptanoic acid, propylhexanoic acid, capric acid, methylnonanoic acid, ethyloctanoic acid, propylheptanoic acid, butylhexanoic acid, undecanoic acid, methyldecanoic acid, ethylnonanoic acid, propyloctanoic acid, butylheptanoic acid, lauric acid, methylundecanoic acid, ethyldecanoic acid, propylnonanoic acid, butyloctanoic acid, Saturated fatty acids such as pentylheptanoic acid, tridecanoic acid, methyldodecanoic acid, ethylundecanoic acid, propyldecanoic acid, butylnonanoic acid, pentyloctanoic acid, myristic acid, methyltridecanoic acid, ethyldodecanoic acid, propylundecanoic acid, butyldecanoic acid, pentylnonanoic acid, hexyloctanoic acid, pentadecanoic acid, methyltetradecanoic acid, ethyltridecanoic acid, propyldodecanoic acid, butylundecanoic acid, pentyldecanoic acid, hexylnonanoic acid, palmitic acid, methylpentadecanoic acid, ethyltetradecanoic acid, propyltridecanoic acid, butyldodecanoic acid, pentylundecanoic acid, hexyldecanoic acid, heptylnonanoic acid, methylcyclohexanecarboxylic acid, ethylcyclohexanecarboxylic acid, propylcyclohexanecarboxylic acid, butylcyclohexanecarboxylic acid, pentylcyclohexanecarboxylic acid, hexylcyclohexanecarboxylic acid, heptylcyclohexanecarboxylic acid, octylcyclohexanecarboxylic acid, nonylcyclohexanecarboxylic acid, etc. Examples include acids, unsaturated fatty acids such as nonenic acid, methylnonenic acid, 10-hydroxy-2-decenoic acid, undecenoic acid, dodecenoic acid, tridecenoic acid, tetradecenoic acid, myristoleic acid, pentadecenoic acid, hexadecenoic acid, palmitoleic acid, and sabienoic acid; and aromatic carboxylic acids such as terephthalic acid, pyromellitic acid, o-phenoxybenzoic acid, methylbenzoic acid, ethylbenzoic acid, propylbenzoic acid, butylbenzoic acid, pentylbenzoic acid, hexylbenzoic acid, heptylbenzoic acid, octylbenzoic acid, and nonylbenzoic acid. Organic acids may be used individually or in combination of two or more.By combining such organic acids with the second copper particles described above, it tends to be possible to achieve both good dispersibility of the second copper particles and good desorption of the organic acid during sintering.

[0166] The amount of surface treatment agent applied may be an amount that adheres to the surface of the second copper particles in a layer of one to three molecules. The amount of surface treatment agent applied may be 0.07% by mass or more, 0.10% by mass or more, or 0.2% by mass or more, and may be 2.1% by mass or less, 1.6% by mass or less, or 1.1% by mass or less. The amount of surface treatment applied to the second copper particles can be calculated for the first copper particles using the method described above. The same applies to the specific surface area, the molecular weight of the surface treatment agent, and the minimum coating area of ​​the surface treatment agent.

[0167] As the second copper particle, either a synthesized one or a commercially available one can be used.

[0168] The total content of first copper particles and second copper particles in the metal paste may be 70% by mass or more, 80% by mass or more, 90% by mass or more, or 93% by mass or more, based on the total mass of metal particles contained in the metal paste. If the total content of first copper particles and second copper particles is within the above range, it becomes easier to suppress the generation of voids in through holes or non-through holes. From the viewpoint of making it easier to obtain such effects, the total content of first copper particles and second copper particles may be 80% by mass or more, 90% by mass or more, or 95% by mass or more, based on the total mass of metal particles.

[0169] Solder particles can be those containing tin or tin alloys. Examples of tin alloys include In-Sn, In-Sn-Ag, Sn-Bi, Sn-Bi-Ag, Sn-Ag-Cu, and Sn-Cu alloys, as shown below. • In-Sn (In 52% by mass, Bi 48% by mass, melting point: 118℃) ·In-Sn-Ag (In20% by mass, Sn77.2% by mass, Ag2.8% by mass, melting point: 175℃) • Sn-Bi (Sn 43% by mass, Bi 57% by mass, melting point: 138℃) • Sn-Bi-Ag (Sn 42% by mass, Bi 57% by mass, Ag 1% by mass, melting point: 139℃) ·Sn-Ag-Cu (96.5% by mass of Sn, 3% by mass of Ag, 0.5% by mass of Cu, melting point: 217℃) • Sn-Cu (Sn 99.3% by mass, Cu 0.7% by mass, melting point: 227°C)

[0170] The metal paste may contain solder particles, for example, solder particles with a particle size (maximum diameter) of 1.0 μm or more.

[0171] The particle size (maximum diameter) of the solder particles may be 2 μm or larger, 15 μm or smaller, or 8.0 μm or smaller.

[0172] The average particle size (average maximum diameter) of the solder particles contained in the metal paste may be 1.0 μm or larger, 2.0 μm or larger, 15 μm or smaller, or 8 μm or smaller, from the viewpoint of suppressing volume shrinkage due to sintering in through holes and suppressing void formation in through holes. Having such a particle size for the solder particles allows for sufficient reduction of volume shrinkage when copper particles in the metal paste are sintered, making it easy to form a metal body with a porous structure and a sufficiently formed conductive network in through holes or non-through holes. As a result, the formation of voids in through holes or non-through holes can be suppressed, a metal body that is less prone to cracking on the main surface of the substrate (e.g., a silicon substrate) can be formed, and when wiring is formed from a conductor containing this metal body, disconnection due to thermal stress of the wiring can be further suppressed.

[0173] It is preferable that the solder particles have a uniform particle size. In this case, they tend to be highly dispersed and scattered within the metal body, and voids are generated within the scattered solder or around the outer edge of the solder (between the solder and the copper sintered body), thereby suppressing the generation of continuous voids and cracks longer than 5 μm, and preventing disconnections within vias and disconnections of wiring formed on the main surface of the substrate (e.g., a silicon substrate).

[0174] The particle size (maximum diameter) and average particle size (average maximum diameter) of solder particles can be calculated from SEM images using, for example, the following procedure: Place the solder particle powder onto a carbon tape for SEM using a spatula to create an SEM sample. Observe this SEM sample at 5000x magnification using an SEM device. Create a rectangle circumscribing the solder particles in the SEM image using image processing software, and define the particle size (maximum diameter) as the longer side of the rectangle. Perform this measurement on 50 or more solder particles using multiple SEM images to calculate the average particle size (average maximum diameter).

[0175] The shape of the solder particles may be, for example, spherical, lumpy, needle-shaped, flattened (flake-shaped), or nearly spherical. The solder particles may also be aggregates of solder particles having these shapes.

[0176] The solder particles are preferably spherical. In this case, they are uniformly dispersed inside the metal body, and voids are generated inside the uniformly dispersed solder or around the outer periphery of the solder (between the solder and the copper sintered body). This suppresses volume shrinkage when the copper particles in the metal paste are sintered, making it easier to suppress wire breakage that occurs in through holes. Furthermore, by suppressing volume shrinkage when the copper particles in the metal paste are sintered, cracks in the metal body formed on at least one main surface of the substrate (e.g., a silicon substrate) can be suppressed.

[0177] The solder particle content in the metal paste may be 1 part by mass or more, 2 parts by mass or more, or 3 parts by mass or more, and may be 25 parts by mass or less, 20 parts by mass or less, or 15 parts by mass or less, per 100 parts by mass of copper particles contained in the metal paste. If the solder particle content is within the above range, the above-mentioned effects will be more easily obtained. In other words, it becomes easier to form a metal body with a sufficiently formed conductive network while having a porous structure in through holes or non-through holes, while suppressing the generation of voids in through holes. Furthermore, when the metal paste contains the above-mentioned first copper particles, the first copper particles tend to be oriented substantially parallel to the coating surface of the metal paste on the main surface of the substrate (e.g., a silicon substrate), and by more effectively suppressing volume shrinkage, it is possible to form a metal body that is less prone to cracking. When wiring is formed from a conductor containing this metal body, disconnection due to thermal stress of the wiring can be further suppressed. If the solder particle content is 1 part by mass or more per 100 parts by mass of copper particles, it becomes easier to suppress the formation of voids inside the metal body, making it easier to prevent cracks longer than 5 μm from forming and causing wire breaks inside vias. On the other hand, if the solder particle content is 25 parts by mass or less per 100 parts by mass of copper particles, the resistivity of the wiring is less likely to increase.

[0178] The total content of copper particles (preferably, the content of first copper particles and second copper particles) and solder particles in the metal paste may be 90% by mass or more, based on the total mass of metal particles contained in the metal paste. In this case, it becomes easier to suppress the generation of voids in through holes or non-through holes. From the viewpoint of obtaining such effects even more easily, the total content of copper particles (preferably, the content of first copper particles and second copper particles) and solder particles may be 95% by mass or more, or even 100% by mass, based on the total mass of metal particles.

[0179] The metal paste may further contain other metal particles besides copper particles and solder particles. Examples of other metal particles include nickel, silver, gold, palladium, and platinum particles. The average particle size (maximum diameter) of the other metal particles may be 0.01 μm or more, 0.05 μm or more, 5 μm or less, 3.0 μm or less, or 2.0 μm or less. If other metal particles are included, their content may be less than 20% by mass and 10% by mass or less, based on the total mass of metal particles contained in the metal paste, from the viewpoint of obtaining sufficient bonding properties. Other metal particles may not be included. The shape of the other metal particles is not particularly limited.

[0180] The metal paste may contain a dispersion medium. The dispersion medium is not particularly limited and may be volatile, for example. Examples of volatile dispersion media include monohydric and polyhydric alcohols such as pentanol, hexanol, heptanol, octanol, decanol, ethylene glycol, diethylene glycol, propylene glycol, butylene glycol, α-terpineol, and isobornylcyclohexanol (MTPH); ethylene glycol butyl ether, ethylene glycol phenyl ether, diethylene glycol methyl ether, diethylene glycol ethyl ether, diethylene glycol butyl ether, diethylene glycol isobutyl ether, diethylene glycol hexyl ether, triethylene glycol methyl ether, diethylene glycol dimethyl ether, diethylene glycol diethyl ether, diethylene glycol dibutyl ether, diethylene glycol butyl methyl ether, diethylene glycol isopropyl methyl ether, triethylene glycol dimethyl ether, triethylene glycol butyl methyl ether, propylene glycol propyl ether, and dipropylene glycol Examples include ethers such as methyl ether, dipropylene glycol ethyl ether, dipropylene glycol propyl ether, dipropylene glycol butyl ether, dipropylene glycol dimethyl ether, tripropylene glycol methyl ether, and tripropylene glycol dimethyl ether; esters such as ethylene glycol ethyl ether acetate, ethylene glycol butyl ether acetate, diethylene glycol ethyl ether acetate, diethylene glycol butyl ether acetate, dipropylene glycol methyl ether acetate (DPMA), ethyl lactate, butyl lactate, γ-butyrolactone, and propylene carbonate; acid amides such as N-methyl-2-pyrrolidone, N,N-dimethylacetamide, and N,N-dimethylformamide; aliphatic hydrocarbons such as cyclohexane, octane, nonane, decane, and undecane; aromatic hydrocarbons such as benzene, toluene, and xylene; mercaptans having alkyl groups with 1 to 18 carbon atoms; and mercaptans having cycloalkyl groups with 5 to 7 carbon atoms.Examples of mercaptans having an alkyl group with 1 to 18 carbon atoms include ethyl mercaptan, n-propyl mercaptan, i-propyl mercaptan, n-butyl mercaptan, i-butyl mercaptan, t-butyl mercaptan, pentyl mercaptan, hexyl mercaptan, and dodecyl mercaptan. Examples of mercaptans having a cycloalkyl group with 5 to 7 carbon atoms include cyclopentyl mercaptan, cyclohexyl mercaptan, and cycloheptyl mercaptan.

[0181] The content of the dispersion medium may be 3 parts by mass or more, 5 parts by mass or more, 20 parts by mass or less, or 12 parts by mass or less, based on the total mass of metal particles contained in the metal paste being 100 parts by mass. If the content of the dispersion medium is within these ranges, the viscosity of the metal paste can be adjusted to a more appropriate level, and the formation of voids in through-holes can be more easily suppressed.

[0182] The metal paste may, as needed, contain wetting agents such as nonionic surfactants and fluorinated surfactants; defoaming agents such as silicone oil; and ion trapping agents such as inorganic ion exchangers.

[0183] The metal paste described above can be prepared by mixing copper particles, solder particles, and any other components (additives, other metal particles, etc.) in a dispersion medium. After mixing each component, stirring may be performed. The maximum diameter of the dispersion may be adjusted by a classification operation.

[0184] If the metal paste contains the first copper particles and the second particles described above, the second copper particles, a surface treatment agent, and a dispersion medium may be pre-mixed and dispersed to prepare a dispersion of the second copper particles. This dispersion may then be further mixed with the first copper particles, solder particles, other metal particles as needed, and any additives. This procedure improves the dispersibility of the second copper particles and improves their mixability with the first copper particles, thereby further improving the performance of the metal paste. Aggregates may be removed by subjecting the dispersion of the second copper particles to a classification operation. [Examples]

[0185] The present invention will be described in more detail below with reference to examples and comparative examples, but the present invention is not limited to the following examples.

[0186] (Synthesis of the second copper particle) [Synthesis of copper nonanoate] 91.5 g (0.94 mol) of copper hydroxide (Kanto Chemical Co., Ltd., special grade) was mixed with 150 mL of 1-propanol (Kanto Chemical Co., Ltd., special grade) and stirred. Then, 370.9 g (2.34 mol) of nonanoic acid (Kanto Chemical Co., Ltd., 90% or higher) was added. The resulting mixture was heated and stirred in a separable flask at 90°C for 30 minutes. The resulting solution was filtered while still heated to remove undissolved material. After cooling, the generated copper nonanoate was filtered by suction and washed with hexane until the washing solution was clear. The resulting powder was dried in an explosion-proof oven at 50°C for 3 hours to obtain copper(II) nonanoate. The yield was 340 g (yield 96% by mass).

[0187] [Synthesis of the second copper particle] 15.01 g (0.040 mol) of copper(II) nonanoate obtained above and 7.21 g (0.040 mol) of anhydrous copper(II) acetate (Kanto Chemical Co., Ltd., special grade) were placed in a separable flask, and 22 mL of 1-propanol and 32.1 g (0.32 mol) of hexylamine (Tokyo Chemical Industries, Ltd., 99% purity) were added. The mixture was heated and stirred in an oil bath at 80°C to dissolve the substances. The solution was then transferred to an ice bath and cooled until the internal temperature reached 5°C. After that, 7.72 mL (0.16 mol) of hydrazine monohydrate (Kanto Chemical Co., Ltd., special grade) was added and stirred in the ice bath. The molar ratio of copper to hexylamine was 1:4. Next, the mixture was heated and stirred in an oil bath at 90°C. During this time, a reduction reaction accompanied by foaming proceeded, and the reaction was completed within 30 minutes. The inner wall of the separable flask exhibited a coppery luster, and the solution changed to a dark red color. A solid was obtained by centrifuging at 9000 rpm (revolutions / minute) for 1 minute. The solid was then washed with 15 mL of hexane three times to remove the acid residue, yielding a copper particle powder (second copper particle) with a copper luster.

[0188] The copper particles synthesized as described above were observed using a transmission electron microscope (JEOL Ltd., product name: JEM-2100F). The average length of the long axis of 200 randomly selected copper particles was 104 nm. The shape of the second particle was spherical.

[0189] (Preparation of metal paste) <Examples 1-55> The raw materials listed below were mixed in the proportions shown in Tables 1-6 to prepare the metal paste.

[0190] [First copper particle] Flattened 1.4μm: 1100YP (Manufactured by Mitsui Mining & Smelting Co., Ltd., average particle size 1.4μm (D50), product name)

[0191] [Second copper particle] Spherical 100nm: Copper particles synthesized as described above.

[0192] [Solder particles] SnBi58: SnBi58 solder STC-3 (manufactured by Mitsui Mining & Smelting Co., Ltd., product name, average particle size 4.1 μm (D50), spherical) SnAgCu:Sn96.5Ag3Cu0.5 solder STC-3 (manufactured by Mitsui Mining & Smelting Co., Ltd., product name, average particle size 4.1 μm (D50), spherical)

[0193] [others] Diethylene glycol: Manufactured by Fujifilm Wako Pure Chemical Corporation

[0194] <Comparative Example 1> A metal paste was prepared by mixing 70 parts by mass of 1100YP (manufactured by Mitsui Mining & Smelting Co., Ltd., average particle size 1.4 μm (D50), trade name) as the first copper particles, 30 parts by mass of the copper particles synthesized above as the second copper particles, 5 parts by mass of diethylene glycol (manufactured by Fujifilm Wako Pure Chemical Industries, Ltd.), and 5 parts by mass of a resin component. As the resin component, a mixture of acrylic resin as an organic binder and a mixture of carbitol and terpineol as organic solvents (with a mass ratio of carbitol:terpineol = 1:1 in the mixture) was used in a mass ratio of 1:2.

[0195] (Preparation process for silicon substrate) <Examples 1-55 and Comparative Example 1> A silicon substrate was prepared with through-holes, and titanium, nickel, and copper layers were formed in that order on both main surfaces and on the walls of the through-holes. The silicon substrate had a diameter of 6 inches and a thickness of 500 μm. The hole diameters of the through-holes in the silicon substrate are shown in Tables 1 to 6. The titanium, nickel, and copper layers were formed sequentially by sputtering.

[0196] (Metal body formation process) <Examples 1-40, 46-55> The prepared metal paste was applied to both main surfaces of the silicon substrate using a metal spatula, filling the through-holes with the paste. After application, the metal paste was dried in air at 90°C for 10 minutes. After drying, a 30 μm thick layer of metal paste was formed on the silicon substrate.

[0197] A silicon substrate with a metal paste layer formed on it was placed in a tube furnace (manufactured by AVC Co., Ltd.), and the air inside the tube furnace was replaced with argon gas by flowing argon gas at a rate of 1 L / min. Then, the metal paste was sintered by heating for 10 minutes while flowing hydrogen gas at a rate of 300 mL / min, followed by sintering at 250°C for 60 minutes. After that, the argon gas flow rate was changed to 0.3 L / min, and the substrate was cooled and removed into the air at a temperature below 50°C to obtain a silicon substrate filled with metal. The thickness of the metal formed on both main surfaces of the silicon substrate after sintering was 25 μm.

[0198] <Examples 41-45> A metal-filled silicon substrate was obtained in the same manner as in Example 1, except that the silicon substrate was pressurized using the pressurization method described below. The thickness of the metal formed on both main surfaces of the silicon substrate after sintering was 30 μm. [Method of applying pressure] A silicon substrate with a metal paste layer formed on it was pressurized from both sides using a pressurizing jig. The pressure applied to the silicon substrate was adjusted to the pressures listed in Tables 1-6. The pressurizing jig consisted of a flat aluminum plate and a spring, allowing for adjustment of the pressurizing pressure. The silicon substrate pressurized by the pressurizing jig was placed in a tube furnace (manufactured by AVC Co., Ltd.), and argon gas was flowed through it at a rate of 1 L / min to replace the air inside the tube furnace with argon gas.

[0199] <Examples 26-30, 51-55> A metal-filled silicon substrate was obtained in the same manner as in Example 1, except that the heating time was increased for 10 minutes and the sintering treatment was performed at 225°C for 60 minutes. The thickness of the metal bodies formed on both main surfaces of the silicon substrate after sintering was 30 μm.

[0200] <Comparative Example 1> A metal-filled silicon substrate was obtained in the same manner as in Example 1, except that the heating time was increased for 10 minutes and the sintering treatment was performed at 300°C for 60 minutes. The thickness of the metal bodies formed on both main surfaces of the silicon substrate after sintering was 30 μm.

[0201] (Measurement of porosity in metal bodies) <Examples 1-73 and Comparative Example 1> A focused ion beam processing and observation system (Hitachi High-Technologies Corporation, product name: MI4050) was used to expose the cross-sections of the central part of the through-holes in the silicon substrate and the cross-sections of the metal bodies provided on the main surface of the silicon substrate using a focused ion beam, and these cross-sections were observed. When observing the cross-section of the central part of the through-holes, the observation was conducted within a range of ±5 μm in the thickness direction of the silicon substrate and ±5 μm in the direction perpendicular to the thickness direction of the silicon substrate, starting from the center of the metal body filling the through-hole. When observing the cross-section of the metal bodies provided on the main surface of the silicon substrate, the observation was conducted within a range of 10 μm in the thickness direction of the silicon substrate and 10 μm in the direction perpendicular to the thickness direction of the silicon substrate, in the region up to 5 μm from the surface of the metal body formed on the main surface of the silicon substrate. For observation, a scanning electron microscope (Hitachi High-Technologies Corporation, product name: S-3700N) was used, and a cross-sectional image (approximately 10 μm square) of the metal body was captured at a magnification of 10,000x. Five observation points were selected. The obtained cross-sectional images were binarized using image analysis software (Adobe Photoshop® Elements) to separate the metallic and porous portions. For each of the five observation points, the ratio of the porous portion area to the total area of ​​the metal body cross-section was defined as the porosity. The average of the porosities from the five observation points was defined as the porosity of the metal body. The results are shown in Tables 1-6.

[0202] (Resin impregnation process) <Examples 1-55> The curable resin composition shown below was applied to one side of a metal-filled silicon substrate using a roll coater. Next, the metal-filled silicon substrate was placed in a container, and the container was vacuumed to a gauge pressure of 100 kPa. The metal-filled silicon substrate was held in the vacuum for 10 minutes, and then removed from the container. It was confirmed that the curable resin composition had impregnated the metal in the through-holes and that the curable resin composition had reached the side of the metal in the through-holes opposite to the side to which the curable resin composition had been applied. The curable resin composition remaining on the coated surface of the metal-filled silicon substrate was removed with a rubber spatula. Next, the curable resin composition was applied to the side opposite to the coated side using a roll coater, and the curable resin composition remaining on the surface of the metal-filled silicon substrate was removed as much as possible with a rubber spatula.

[0203] [Curable resin composition] YDF-170 (manufactured by Toto Kasei Co., Ltd., product name for bisphenol F type epoxy resin, epoxy equivalent = 170): 95 parts by mass 2PZ-CN (manufactured by Shikoku Chemicals Co., Ltd., trade name for an imidazole compound): 5 parts by mass

[0204] <Comparative Example 1> The resin impregnation process was not performed.

[0205] (Resin curing process) <Examples 1-55> A silicon substrate having through-electrodes was obtained by impregnating a metal body with a curable resin composition and holding it in a nitrogen atmosphere at 180°C for 1 hour.

[0206] <Comparative Example 1> The resin curing process was not performed.

[0207] (Conductor removal process) <Examples 1-55 and Comparative Example 1> Mechanical polishing was performed on both sides of a substrate with a silicon through-electrode until the thickness of the metal body on both sides of the substrate was 20 μm. A ceramic jig (manufactured by Kemet Japan Co., Ltd.) was used as the sample holder for attaching the substrate with the silicon through-electrode, and Alcowax (manufactured by Nichika Seikou Co., Ltd.) was used as the material for attaching the substrate to the sample holder. In addition, DP-suspension P-3 μm, 1 μm, and 1 / 4 μm (manufactured by Struas) were used in order as polishing agents.

[0208] [Filling density of cured resin in conductive materials] <Examples 1-55 and Comparative Example 1> A substrate having a mechanically polished silicon through-electrode was cut in the thickness direction, and the cross-section of the center of the through-hole in the silicon substrate and the cross-section of the conductor provided on the main surface of the silicon substrate were exposed by a focused ion beam and observed. When observing the cross-section of the center of the through-hole in the silicon substrate, the range observed was ±5 μm in the thickness direction of the silicon substrate and ±5 μm in the direction perpendicular to the thickness direction of the silicon substrate from the center of the through-hole. When observing the cross-section of the conductor provided on the main surface of the silicon substrate, the range observed was 10 μm in the thickness direction of the silicon substrate and 10 μm in the direction perpendicular to the thickness direction of the silicon substrate, within the region up to 5 μm from the surface of the conductor provided on the main surface of the silicon substrate. A focused ion beam processing observation device (Hitachi High-Technologies Corporation, product name: MI4050) was used. For observation, a scanning electron microscope (Hitachi High-Technologies Corporation, product name: S-3700N) was used, with a magnification of 10,000x, and a cross-sectional image (approximately 10 μm square) of the conductor was captured. Five observation points were selected. The obtained cross-sectional images were binarized using image analysis software (Adobe Photoshop® Elements) to separate the metal portion, the resin cured portion, and the space not filled by the resin cured material in the porous portion. For each of the five observation points, the ratio of the area of ​​the space not filled by the resin cured material in the porous portion to the total area of ​​the conductor cross-section was calculated and defined as the porosity. The average of the porosities from the five observation points was defined as the porosity of the conductor. The filling rate of the resin cured material in the conductor was calculated by substituting the porosity of the metal and the porosity of the conductor into the following equation (1). Filling rate (%) of resin cured material in a conductor = [(BA) / B] × 100 ... Equation (1) [In formula (1), A represents the porosity (%) of the conductor, and B represents the porosity (%) of the copper sintered body.]

[0209] (Wiring formation process (resist formation, etching, and resist removal)) <Examples 1-55 and Comparative Example 1> A dry film H-W425 (manufactured by Hitachi Chemical Co., Ltd., product name) for UV-curable etching resists was laminated to the metal surfaces on both sides of a substrate having a silicon through-electrode that had undergone mechanical polishing. Subsequently, a photomask was placed over the substrate to expose the wiring pattern, and after resist development, etching of the copper sintered body, and removal of the resist, wiring was formed to obtain a substrate (test piece 55) having a silicon through-electrode as shown in Figure 10. In the obtained substrate (test piece 55) having a silicon through-electrode, the conductors filling the through-holes are electrically connected by conductors (wirings) provided on the substrate surface.

[0210] (Initial resistance value) <Examples 1-55 and Comparative Example 1> The linked connection resistance was measured as the initial resistance value of a substrate (test piece 55) having silicon through-holes. When the diameter of the through-holes in the silicon substrate was 20 μm, the resistance value of 20 connected through-holes was measured. When the diameter of the through-holes was 30 μm, the resistance value of 30 connected through-holes was measured. When the diameter of the through-holes was 50 μm, the resistance value of 30 connected through-holes was measured. When the diameter of the through-holes was 100 μm, the resistance value of 100 connected through-holes was measured. When the diameter of the through-holes was 200 μm, the resistance value of 200 connected through-holes was measured. The measured linked connection resistance was evaluated according to the following criteria. A rating of B or higher was considered good. The results are shown in Tables 1-6. A: Resistance is less than 10mΩ B: Resistance value of 10mΩ or more, less than 30mΩ C: Resistance value between 30mΩ and 100mΩ D: Resistance value of 100mΩ or more, less than 500mΩ E: Resistance value of 500mΩ or more

[0211] (Temperature cycle connectivity test) <Examples 1-55 and Comparative Example 1> A substrate (test piece 55) with silicon through-hole electrodes was placed in a temperature cycle tester (TSA-72SE-W, manufactured by ESPEC Corporation), and a temperature cycle connection reliability test was performed under the following conditions: low temperature side: -40°C, 15 minutes; room temperature: 2 minutes; high temperature side: 125°C, 15 minutes; defrost cycle: automatic; cycle count: 50, 100, 300, 500 cycles. When the diameter of the through-holes in the silicon substrate was 20 μm, the resistance value of 20 connected through-holes was measured. When the diameter of the through-holes in the silicon substrate was 30 μm, the resistance value of 30 connected through-holes was measured. When the diameter of the through-holes in the silicon substrate was 50 μm, the resistance value of 30 connected through-holes was measured. When the diameter of the through-holes in the silicon substrate was 100 μm, the resistance value of 100 connected through-holes was measured. When the diameter of the through-holes in the silicon substrate was 200 μm, the resistance value of 200 connected through-holes was measured. The measured connection resistance values ​​were evaluated according to the following criteria. A rating of B or higher after 500 temperature cycles was considered good. The results are shown in Tables 1-6. A: Resistance change rate is less than 1% relative to the initial resistance value. B: Resistance change rate is 1% or more but less than 3% relative to the initial resistance value. C: Resistance change rate is 3% or more but less than 5% relative to the initial resistance value. D: Resistance change rate is 5% or more but less than 10% of the initial resistance value. E: Resistance change rate is 10% or more but less than 20% of the initial resistance value. F: Resistance change rate is 20% or more relative to the initial resistance value.

[0212] (Cracked circuit board) <Examples 1-55 and Comparative Example 1> A substrate (test piece 55) with a silicon through-electrode was visually inspected to check for cracks in the silicon substrate. A circle (○) indicated no cracks, while a cross (×) indicated any cracks, even partial ones. The results are shown in Tables 1-6.

[0213] (airtightness) <Examples 1-55 and Comparative Example 1> The airtightness of a substrate (test piece 55) having silicon through electrodes was evaluated. The evaluation was performed using a helium leak detector ("UL200" manufactured by LEYBOLD). Specifically, the substrate having silicon through electrodes was set in a jig, evacuated until the inlet pressure of the measuring machine reached 5 Pa, and when the inlet pressure reached 5 Pa, He pressurization (0.1 MPa) was performed for 30 seconds, and then the leak amount was measured and evaluated according to the following criteria. The results are shown in Tables 1 to 6. A: The leak amount is less than 1×10 -11 Pa·m 3 / sec B: The leak amount is 1×10 or more and less than 1×10 -11 -10 Pa·m 3 / sec C: The leak amount is 1×10 or more and less than 1×10 -10 -9 Pa·m 3 / sec D: The leak amount is 1×10 or more and less than 1×10 -9 -8 Pa·m 3 / sec E: The leak amount is 1×10 or more and less than 1×10 -8 -6 Pa·m 3 / sec F: The leak amount is 1×10 or more and 1×10 -6 Pa·m 3 / sec or more.

[0214] (Adhesion of wiring - Pull strength -) <Examples 1 to 55 and Comparative Example 1> For a substrate having silicon through electrodes obtained in the same manner except that a 2 mm×2 mm wiring pattern was formed in the wiring formation step, the tip area was 1 mm<00…33>​​​​Stud pins were joined vertically by soldering to form a test specimen. The specimen was fixed, the stud pins were gripped in the chuck of a tensile testing machine, and it was pulled vertically upward at a lifting speed of 50 mm / min. The fracture load at which the metal body on the main surface of the silicon substrate peeled off from the silicon substrate was measured. The adhesion strength was then calculated from the obtained fracture load measurement and the fracture area of ​​the metal body using the following formula. The measurement values ​​were averaged from 10 points and evaluated according to the following criteria. The results are shown in Tables 1 to 6.

[0215] Adhesion strength (MPa) = Breaking load (kgf) / Breaking area (mm²) × 9.8 (N / kgf). A: Adhesion strength (MPa) of 50 MPa or higher B: Adhesion strength (MPa) is between 40 MPa and less than 50 MPa C: Adhesion strength (MPa) is between 30 MPa and less than 40 MPa D: Adhesion strength (MPa) is between 20 MPa and less than 30 MPa E: Adhesion strength (MPa) is between 5 MPa and less than 20 MPa F: Adhesion strength (MPa) is less than 5 MPa

[0216] (Wiring formation properties - presence or absence of cracks) <Examples 1-55 and Comparative Example 1> A substrate with silicon through-electrodes, obtained in the same manner as the previous one except for the formation of five 2mm x 2mm wiring patterns during the wiring formation process, was examined using an optical microscope to check for the presence or absence of cracks (length 0.5mm or longer) in the wiring patterns. The magnification was set to 500x, and the following criteria were used for evaluation. The results are shown in Tables 1-6. A: No cracks were found. B: One or more cracks, but less than two. C: Two or more cracks, but less than five. D: 5 or more cracks, but less than 10 cracks E: 10 or more cracks, but less than 20. F: More than 20 cracks

[0217] (Volume resistivity) <Examples 1-55 and Comparative Example 1> The volume resistivity of a conductor formed on a silicon substrate was measured. Volume resistivity was calculated from the surface resistance value measured with a four-point needle surface resistance meter (Mitsubishi Analytec, product name: Loresta GP) and the film thickness obtained with a non-contact surface / layer cross-sectional shape measurement system (VertScan, Ryoka Systems Co., Ltd.). The results are shown in Tables 1-6.

[0218] (Observation results of gaps around solder) Figures 11 and 12 show cross-sectional images (approximately 20 μm square) of the metal body of the metal-filled silicon substrate fabricated in Example 1, taken using a digital microscope (Keyence Corporation, product name: VHX-6000). As shown in Figures 11 and 12, the metal body contains a copper sintered body, solder (SiBi58), and voids. The voids are located inside the copper sintered body and solder, and around the outer periphery of the solder (between the solder and the copper sintered body). The void 4a shown in Figure 11 is located inside the solder 14a, and the void 4b shown in Figure 12 is located around the outer periphery of the solder 14b.

[0219] Figure 13 shows a cross-sectional image (approximately 10 μm × 19 μm square) of the metal-filled silicon substrate fabricated in Example 46, taken using a digital microscope (Keyence Corporation, product name: VHX-6000). The void 4c shown in Figure 13 is located inside the solder 14c.

[0220] <Comparative Example 1> Figure 14 shows a cross-sectional image (approximately 10 μm × 12 μm square) of the metal-filled silicon substrate fabricated in Comparative Example 1, taken using a digital microscope (Keyence Corporation, product name: VHX-6000). As shown in Figure 14, cracks 18 were observed inside the copper sintered body 12d constituting the metal.

[0221] In the metal-filled silicon substrates fabricated in the examples, it is believed that the occurrence of the aforementioned cracks is suppressed by the scattering of solder originating from solder particles.

[0222] [Table 1]

[0223]

Table 2

[0224]

Table 3

[0225]

Table 4

[0226]

Table 5

[0227]

Table 6

Explanation of Symbols

[0228] 1…Silicon wafer, 2…Metal film, 3…Metal paste, 4…Vacancy (porous, void), 5…Metal body, 6…Cured resin, 8…Etching resist, 9…Wiring, 12…Copper sintered body, 14…Solder, 15…Fine bump, 20…Cured adhesive, 25…Interposer substrate, 27…Wiring, 30…Through hole, 31…Non-through hole, 35…Conductor, 40, 41…Silicon substrate, 50, 60…Silicon substrate filled with metal body, 51, 52, 61…Substrate with silicon through electrode, 55…Test piece, 100, 200, 300…Semiconductor device, A…Pressing jig

Claims

[Claim 1] A metal paste used to form through electrodes, Containing copper particles and solder particles, A metal paste comprising, as the copper particles, first copper particles having a particle size of 0.8 μm or more, and second copper particles having a particle size of 0.5 μm or less.

Images