Composite material, heat sink, semiconductor package, and method for manufacturing composite material

Abstract

A composite material containing copper and molybdenum, having a plate-like shape with a first surface as a main surface, a composition in which the copper content is 20% by mass or more and 70% by mass or less, a matrix containing molybdenum, and a structure including a first copper phase and a second copper phase dispersed in the matrix, wherein the cross-sectional area of ​​each of the second copper phases is 900 μm in a cross section perpendicular to the first surface. 2 or above, wherein the cross-sectional area of ​​each of the first copper phases is smaller than the cross-sectional area of ​​the second copper phase, the proportion of the total area of ​​the second copper phases to the area of ​​the cross section is 10% or more and 50% or less, the aspect ratio obtained by dividing the average major axis by the average minor axis of the second copper phase is 10 or more, and the second copper phase satisfies the relational expression Y≦−1.5551X+285, where X is the copper content in the composite material and Y is the Vickers hardness of the composite phase formed by the matrix and the first copper phase.

Description

[Technical Field]

[0001] The present disclosure relates to a composite material, a heat sink, a semiconductor package, and a method for manufacturing the composite material. This application claims priority based on Japanese Patent Application No. 2024-084648 filed on May 24, 2024, the entire contents of which are incorporated herein by reference. [Background technology]

[0002] Patent Document 1 discloses a composite material used in heat dissipation components of semiconductor devices. This composite material is a Mo-Cu composite material combining molybdenum (Mo) and Cu (copper). The Cu content in the composite material is 30% by mass or more and 70% by mass or less. This composite material contains a copper pool phase and a Mo-Cu composite phase, with the copper pool phase content being 10% by mass or more and 50% by mass or less. This composite material is manufactured, for example, as follows: A powder mixture of a copper-based material and Mo powder is pressed to form a compact. This compact is then pre-sintered. The resulting sintered compact is infiltrated with copper, and the resulting infiltrated compact is then subjected to plastic processing such as rolling. The copper-based material is a powder with an average minor axis of 50 μm or more and 200 μm or less. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] International Publication No. 2004 / 038049 Summary of the Invention

[0004] The composite material of the present disclosure is a composite material containing copper and molybdenum. The composite material has a plate-like shape with a first surface as a main surface. The composite material has a composition in which the copper content is 20 mass % or more and 70 mass % or less. The composite material has a structure including a matrix containing the molybdenum, and a first copper phase and a second copper phase dispersed in the matrix. In a cross section perpendicular to the first surface, the cross-sectional area of ​​each of the second copper phases is 900 μm2 or above, wherein the cross-sectional area of ​​each of the first copper phases is smaller than the cross-sectional area of ​​the second copper phase, the proportion of the total area of ​​the second copper phases to the area of ​​the cross section is 10% or more and 50% or less, and the aspect ratio of the average major axis of the second copper phases divided by the average minor axis is 10 or more. The composite material satisfies the relational expression Y≦−1.5551X+285, where X is the copper content in the composite material, and Y is the Vickers hardness of the composite phase formed by the matrix and the first copper phase. [Brief explanation of the drawings]

[0005] [Figure 1] FIG. 1 is a schematic perspective view of a composite material according to an embodiment. [Figure 2] FIG. 2 is a partially enlarged view of the cross section II-II of FIG. [Figure 3] FIG. 3 is a process diagram showing a method for producing a composite material according to an embodiment. [Figure 4] FIG. 4 is a schematic cross-sectional view of a heat sink according to the embodiment. [Figure 5] FIG. 5 is a schematic cross-sectional view of a heat sink according to another embodiment. [Figure 6] FIG. 6 is an exploded perspective view of the semiconductor package according to the embodiment. [Figure 7] FIG. 7 is a graph showing the relationship between the linear expansion coefficient and the thermal conductivity of the sample of Test Example 1. [Figure 8] FIG. 8 is a graph showing the relationship between the Cu content and the thermal conductivity of the sample of Test Example 1. [Figure 9] FIG. 9 is a graph showing the relationship between the Cu content of the sample of Test Example 1 and the Vickers hardness of the composite phase.

[0006] [Problem to be solved by this disclosure] The composite material used for the heat sink is required to have both high thermal conductivity and a low coefficient of linear expansion.

[0007] An object of the present disclosure is to provide a composite material that has a low coefficient of linear expansion and high thermal conductivity.

[0008] [Effects of this disclosure] The composite material of the present disclosure has a low coefficient of linear expansion and high thermal conductivity.

[0009] [Description of the embodiments of the present disclosure] The inventors discovered that the thermal conductivity of a composite material having a structure in which a fine first copper phase and a coarse second copper phase are dispersed in a matrix formed of molybdenum varies depending on the heating rate in the sintering process during production. They then found that increasing the heating rate makes the composite phase formed by the matrix and the first copper phase less likely to harden, thereby improving the thermal conductivity of the composite material.

[0010] The present disclosure is based on the above findings. First, embodiments of the present disclosure will be listed and described.

[0011] (1) The composite material of the present disclosure is a composite material containing copper and molybdenum. The composite material has a plate-like shape with a first surface as a main surface. The composite material has a composition in which the copper content is 20% by mass or more and 70% by mass or less. The composite material has a structure including a matrix containing the molybdenum, and a first copper phase and a second copper phase dispersed in the matrix. In a cross section perpendicular to the first surface, the cross-sectional area of ​​each of the second copper phases is 900 μm 2 or above, wherein the cross-sectional area of ​​each of the first copper phases is smaller than the cross-sectional area of ​​the second copper phase, the proportion of the total area of ​​the second copper phases to the area of ​​the cross section is 10% or more and 50% or less, and the aspect ratio of the average major axis of the second copper phases divided by the average minor axis is 10 or more. The composite material satisfies the relational expression Y≦−1.5551X+285, where X is the copper content in the composite material, and Y is the Vickers hardness of the composite phase formed by the matrix and the first copper phase.

[0012] The composite material (1) above has a low coefficient of linear expansion and a high thermal conductivity. One reason for this is that it has a structure in which a first copper phase and a second copper phase are dispersed in the matrix. Another reason is that it contains a specific proportion of the second copper phase, which has a large aspect ratio. Furthermore, the composite material (1) above has a high thermal conductivity while maintaining a similar coefficient of linear expansion compared to conventional composite materials with the same copper content, because the copper content ratio X and the Vickers hardness Y of the composite phase satisfy the above-mentioned relationship.

[0013] (2) In the composite material of (1) above, the average minor axis may be 5 μm or more and 200 μm or less.

[0014] The composite material (2) above is likely to have both high thermal conductivity and a low coefficient of linear expansion.

[0015] (3) The composite material described in (1) or (2) above has a linear expansion coefficient of 8.4 × 10 from room temperature to 800°C. -6 / K or less.

[0016] The composite material (3) above has a linear expansion coefficient close to that of semiconductor elements or ceramic parts. The composite material is used, for example, as a material for heat sinks to dissipate heat generated by heat sources such as semiconductor elements. If the linear expansion coefficient of the composite material is close to that of semiconductor elements, the difference in linear expansion coefficients is small. Therefore, thermal stress generated at the interface between the composite material and semiconductor elements is small. Damage to semiconductor elements due to thermal stress is less likely to occur.

[0017] (4) The composite material of any one of (1) to (3) above may have a thermal conductivity of 180 W / m·K or more.

[0018] The composite material (4) above has high thermal conductivity.

[0019] (5) The heat sink of the present disclosure has a first surface and a second surface opposite to the first surface. The heat sink includes at least one first layer and multiple second layers, and has a structure in which the first layers and the second layers are alternately stacked. The first layer is formed from any of the composite materials described in (1) to (4) above. The second layer is formed from copper or a copper alloy.

[0020] The heat sink (5) above has a first layer formed from the composite material, and therefore has both high thermal conductivity and a low coefficient of linear expansion.

[0021] (6) In the heat sink of (5) above, the layer forming the first surface and the second surface may be the second layer.

[0022] The first and second surfaces are surfaces that transfer heat by coming into contact with other members. The second layer, which is made of copper or a copper alloy, has a higher thermal conductivity than the first layer. The heat sink of (5) above, in which the first and second surfaces are made of the second layer, facilitates heat transfer between the heat sink and other members.

[0023] (7) In the heat sink of (5) or (6) above, at least one of the first surface and the second surface may be provided with a plating layer.

[0024] The heat sink of (7) above has a plating layer on at least one of the first surface and the second surface, which makes it easy to bond other members with a bonding material such as a brazing material.

[0025] (8) A semiconductor package according to the present disclosure includes the heat sink according to any one of (5) to (7) above, and a semiconductor element, the semiconductor element being disposed on the first surface.

[0026] The semiconductor package (8) is suitable for heat dissipation of a semiconductor element because it includes a heat sink that has both high thermal conductivity and a low coefficient of linear expansion. The semiconductor package (8) is less likely to damage the semiconductor element because thermal stress generated at the interface between the semiconductor element and the heat sink tends to be small.

[0027] (9) A method for producing a composite material according to the present disclosure includes the steps of: mixing copper powder and molybdenum powder to obtain a mixed powder; press-molding the mixed powder to obtain a compact; sintering the compact to obtain a sintered body; infiltrating the sintered body with a copper material to obtain an infiltrated body; and compressing the infiltrated body. The copper powder has an average particle size of 50 μm or more and 200 μm or less. In the step of obtaining the sintered body, the temperature rise rate during sintering the compact is 0.1° C. / second or more.

[0028] The composite material manufacturing method (9) above can produce a composite material with high thermal conductivity while having a low coefficient of linear expansion. The manufactured composite material has a structure in which a first copper phase and a second copper phase are dispersed in a matrix formed by a sintered body of molybdenum powder. By sintering the compact, a sintered body in which the molybdenum particles are bonded to each other is obtained. By setting the heating rate during sintering to 0.1°C / second or higher, the molybdenum particles are less likely to coarsen. Therefore, the composite phase formed by the matrix and the first copper phase is less likely to harden, and the thermal conductivity of the composite material is increased.

[0029] (10) In the method for producing a composite material of (9) above, in the step of obtaining the sintered body, the compact may be sintered at a temperature of 1100°C or higher and 1400°C or lower.

[0030] When sintering the compact, the sintering temperature is 1100°C or higher, which facilitates bonding between molybdenum particles. When the sintering temperature is 1400°C or lower, the molybdenum particles are less likely to become coarse.

[0031] (11) In the method for producing a composite material according to (9) or (10), the copper material may be heated to a temperature of 1200° C. or higher in the step of obtaining the infiltrant.

[0032] According to the above (11), the copper material is easily infiltrated into the sintered body.

[0033] (12) In any one of the methods for producing a composite material according to (9) to (11), the step of compressing the infiltrant may compress the infiltrant at a compression rate of 30% or more.

[0034] According to the manufacturing method (12) above, a dense composite material is easily obtained.

[0035] (13) In any of the methods for producing a composite material described in (9) to (12), the infiltrant may be rolled at a temperature of 60°C or higher and 350°C or lower in the step of compressing the infiltrant.

[0036] According to the manufacturing method (13) above, the infiltrant can be easily compressed.

[0037] [Details of the embodiments of the present disclosure] Specific examples of the composite material, heat sink, and semiconductor package of the present disclosure are described below. The same reference numerals in the figures indicate the same objects. The sizes of components shown in each drawing are expressed for the purpose of clarifying the description and do not necessarily represent the actual dimensional relationships. It should be noted that the present invention is not limited to these examples, but is defined by the scope of the claims, and is intended to include all modifications within the meaning and scope equivalent to the claims.

[0038] 《Composite materials》 A composite material 1 according to an embodiment will be described with reference to FIGS. 1 and 2. The composite material 1 is a composite material containing Cu and Mo. The composite material 1 has a plate-like shape. As shown in FIG. 2, the composite material 1 has a structure including a matrix 2 and a first copper phase 3f and a second copper phase 3c dispersed in the matrix 2. The structure of the composite material 1 is a structure in which the second copper phase 3c is dispersed in a composite phase 4 formed by the matrix 2 and the first copper phase 3f. One of the characteristics of the composite material 1 is that it satisfies the relation Y≦−1.5551X+285. X is the Cu content in the composite material 1. The Cu content refers to the mass percentage of Cu when the mass of the entire composite material 1 is taken as 100% by mass. Y is the Vickers hardness of the composite phase 4.

[0039] The composite material 1 is a plate-like body having a first surface 1a and a second surface 1b as its principal surfaces. A principal surface is a surface of the plate that has a larger area than the other surfaces. The second surface 1b is the surface opposite the first surface 1a. The second surface 1b is substantially parallel to the first surface 1a. In the following description, the X direction refers to the direction along the length of the composite material 1. The Y direction refers to the direction along the width of the composite material 1. The Z direction refers to the direction along the thickness of the composite material 1. The X and Y directions are parallel to the first surface 1a. The X and Y directions are perpendicular to each other. The Z direction is perpendicular to the first surface 1a. The Z direction is perpendicular to the X and Y directions. Figure 2 shows a cross section perpendicular to the first surface 1a of the composite material 1. The cross section of the composite material 1 shown in Figure 2 is an XZ cross section. The XZ cross section is perpendicular to the Y direction. In Figure 2, the X direction is the horizontal direction, and the Z direction is the vertical direction.

[0040] The length and width of the composite material 1 are determined from a rectangle with the smallest area that circumscribes the outline of the composite material 1 in a plan view. The length of the composite material 1 is the length of the long side of the rectangle. The width of the composite material 1 is the length of the short side of the rectangle. When the length and width of the composite material 1 are equal, that is, when the rectangle is a square, one side of the square is the length of the composite material 1, and the length of the second side perpendicular to the first side is the width of the composite material 1.

[0041] <Cu content> The Cu content in the composite material 1 is 20% by mass or more and 70% by mass or less. The Cu content is the sum of the Cu content in the first copper phase 3f and the Cu content in the second copper phase 3c. The higher the Cu content, the higher the thermal conductivity of the composite material 1. On the other hand, the lower the Cu content, the higher the Mo content, and therefore the lower the linear expansion coefficient of the composite material 1. By setting the Cu content to 20% by mass or more and 70% by mass or less, the composite material 1 can have both high thermal conductivity and a low linear expansion coefficient.

[0042] The Cu content may be 23% by mass or more and 60% by mass or less, and further 30% by mass or more and 60% by mass or less. When the Cu content is more than 40% by mass, and further 45% by mass or more, the thermal conductivity of the composite material 1 becomes higher. When the Cu content is 50% by mass or less, particularly 40% by mass or less, the linear expansion coefficient of the composite material 1 becomes lower.

[0043] The Mo content in composite material 1 is, for example, 30% by mass or more and 80% by mass or less. The Mo content refers to the mass percentage of Mo when the total mass of composite material 1 is 100% by mass. Composite material 1 in this example has a composition containing Cu and Mo, with the remainder consisting of unavoidable impurities. The composition of composite material 1 can be determined, for example, by inductively coupled plasma optical emission spectroscopy (ICP-OES).

[0044] (Matrix) The matrix 2 is formed from a material containing Mo. The content of Mo in the matrix 2 is 50% by mass or more. The content of Mo in the matrix 2 means the mass percentage of Mo when the mass of the matrix 2 is 100% by mass. The content of Mo in the matrix 2 may be 70% by mass or more, or even 90% by mass or more. In this example, the matrix 2 is formed from molybdenum with a purity of 99% or more. That is, the content of Mo in the matrix 2 is 99% by mass or more. The matrix 2 may be formed from a molybdenum alloy.

[0045] The matrix 2 has a skeleton structure formed by bonding together Mo particles formed of molybdenum or a molybdenum alloy.

[0046] (First copper phase) The first copper phase 3f is formed of a material containing Cu. The Cu content in the first copper phase 3f is 50 mass% or more. The Cu content in the first copper phase 3f means the mass percentage of Cu when the mass of the first copper phase 3f is 100 mass%. The Cu content in the first copper phase 3f may be 70 mass% or more, or even 90 mass% or more. In this example, the first copper phase 3f is formed of copper with a purity of 99% or more. That is, the Cu content in the first copper phase 3f is 99 mass% or more. The first copper phase 3f may be formed of a copper alloy.

[0047] <Cross-sectional area> The size of the first copper phase 3f is smaller than the size of the second copper phase 3c. In the cross section of the composite material 1, the cross-sectional area of ​​the first copper phase 3f is smaller than the cross-sectional area of ​​the second copper phase 3c described below. The first copper phase 3f is a copper phase other than the second copper phase 3c. The cross section of the composite material 1 is a cross section perpendicular to the first surface 1a. In this example, the cross section of the composite material 1 is an XZ cross section.

[0048] (second copper phase) The second copper phase 3c is formed of a material containing Cu. The Cu content in the second copper phase 3c is 50% by mass or more. The Cu content in the second copper phase 3c means the mass percentage of Cu when the mass of the second copper phase 3c is 100% by mass. The Cu content in the second copper phase 3c may be 70% by mass or more, or even 90% by mass or more. In this example, the second copper phase 3c is formed of copper with a purity of 99% or more. That is, the Cu content in the second copper phase 3c is 99% by mass or more. The second copper phase 3c may be formed of a copper alloy.

[0049] The second copper phase 3c has a flat shape extending in a direction parallel to the first surface 1a. In other words, the second copper phase 3c extends in a direction perpendicular to the Z direction. The flat shape means a shape in which the length in the Z direction is shorter than the length in at least one of the X and Y directions. In this example, the shape of the second copper phase 3c seen three-dimensionally is a thin plate extending in the X and Y directions. The multiple second copper phases 3c are oriented in the X direction.

[0050] <Cross-sectional area> The size of the second copper phase 3c is larger than the size of the first copper phase 3f. The cross-sectional area of ​​the second copper phase 3c is 900 μm 2 The cross-sectional area of ​​the second copper phase 3c means the area of ​​each of the second copper phases 3c in the cross section of the composite material 1.

[0051] The cross-sectional areas of the first copper phase 3f and the second copper phase 3c can be determined as follows. The cross section of the composite material 1 is observed using a microscope. The cross section of the composite material 1 is a cross section perpendicular to the first surface 1a. The cross section of the composite material 1 is, for example, an XZ cross section. The microscope is an optical microscope or a scanning electron microscope (SEM). The image of the cross section observed using the microscope is processed to extract the copper phase formed by copper or a copper alloy. The individual areas of all copper phases within the observation range are determined using image analysis. Here, the area of ​​each copper phase is determined as follows. For copper phases that are individually independent in the matrix 2 and whose areas can be measured, the individual areas are determined. When adjacent copper phases are connected by locally thin connections and the maximum thickness of the connections is less than 5 μm, the adjacent copper phases are considered not to be connected, and the area of ​​each copper phase is determined. The maximum thickness of the connections is the maximum thickness of the connections in the longitudinal direction. 900 μm 2 The copper phase having an area of ​​900 μm or more is the second copper phase 3c. 2A copper phase having an area of ​​less than 1 μm is the first copper phase 3f. A copper phase whose area is too small to measure and a thin copper phase whose maximum thickness is less than 5 μm are also considered to be the first copper phase 3f. The magnification of the microscope is, for example, 50 times or more and 500 times or less. The observation range is, for example, a horizontal length of 220 μm or more and 2500 μm or less and a vertical length of 150 μm or more and 2000 μm or less. The horizontal length is the length in the X direction, and the vertical length is the length in the Z direction.

[0052] The average cross-sectional area of ​​the first copper phase 3f is, for example, 100 μm 2 Below that, another 10 μm 2 The average cross-sectional area of ​​the first copper phase 3f is the average value of the cross-sectional areas of all the first copper phases 3f whose areas were measurable within the observation range. The number of the first copper phases 3f whose cross-sectional areas are measured is, for example, 20 or more, and further, 30 or more. The average cross-sectional area of ​​the second copper phase 3c is, for example, 1000 μm 2 Above that, another 2000μm 2 The average cross-sectional area of ​​the second copper phase 3c is the average value of the cross-sectional areas of all the second copper phases 3c within the observation range. The number of second copper phases 3c whose cross-sectional areas are measured is, for example, 10 or more, and further 20 or more. The average cross-sectional area of ​​the second copper phase 3c is, for example, 10 times or more the average cross-sectional area of ​​the first copper phase 3f. The upper limit of the average cross-sectional area of ​​the second copper phase 3c is, for example, 32,000 μm 2 is.

[0053] <Area ratio> The area ratio of the second copper phase 3c in the cross section of the composite material 1 is 10% or more and 50% or less. The area ratio of the second copper phase 3c means the ratio of the total area A3c of the second copper phase 3c to the area A1 of the cross section of the composite material 1. The total area A3c is the total cross-sectional area of ​​all the second copper phases 3c in the cross section of the composite material 1. The area ratio of the second copper phase 3c is expressed as a percentage [A3c / A1]. When the area ratio of the second copper phase 3c is 10% or more, the thermal conductivity of the composite material 1 is likely to be high. When the area ratio of the second copper phase 3c is 50% or less, the linear expansion coefficient of the composite material 1 is likely to be low. The area ratio of the second copper phase 3c may be 15% or more and 45% or less, or even 20% or more and 40% or less.

[0054] The area ratio of the second copper phase 3c can be determined as follows. As described above, the image of the cross section observed with a microscope is image-processed to extract all of the second copper phases 3c within the observation range. The total area of ​​the second copper phases 3c is calculated. The total area of ​​the second copper phases 3c divided by the area of ​​the observation range is the area ratio of the second copper phases 3c.

[0055] The area ratio of the first copper phase 3f in the cross section of the composite material 1 is, for example, 10% or more and 60% or less. The area ratio of the first copper phase 3f can be determined in the same manner as the area ratio of the second copper phase 3c. The total area of ​​the first copper phase 3f is the sum of the cross-sectional areas of all the first copper phases 3f whose areas were measurable within the observation range.

[0056] Aspect ratio The aspect ratio of the second copper phase 3c is 10 or more. The aspect ratio of the second copper phase 3c is the value obtained by dividing the average major axis Lxa of the second copper phase 3c in the cross section of the composite material 1 by the average minor axis Lza. The aspect ratio of the second copper phase 3c is expressed as [Lxa / Lza]. When the aspect ratio of the second copper phase 3c is 10 or more, the linear expansion coefficient of the composite material 1 tends to be low. The aspect ratio of the second copper phase 3c may be 15 or more, or even 20 or more. The upper limit of the aspect ratio of the second copper phase 3c is, for example, 200. When the aspect ratio of the second copper phase 3c is 200 or less, the thermal conductivity of the composite material 1 is less likely to decrease. The aspect ratio of the second copper phase 3c may be, for example, 10 to 200, or even 15 to 180.

[0057] The aspect ratio of the second copper phase 3c can be determined as follows. As described above, an image of the cross section observed with a microscope is subjected to image processing to extract all of the second copper phases 3c within the observation range. A plurality of second copper phases 3c are selected. The number of second copper phases 3c selected is, for example, 5 or more, or even 10 or more. All of the second copper phases 3c within the observation range may be selected. The major axis and minor axis of each selected second copper phase 3c are determined. The major axis and minor axis of the second copper phase 3c can be determined as follows. The outline of the second copper phase 3c is extracted by image analysis, and a rectangle with the smallest area circumscribing the outline is determined. The length of the long side of this rectangle is the major axis of the second copper phase 3c, and the length of the short side of the rectangle is the minor axis of the second copper phase 3c. The average value of the major axes of the selected second copper phases 3c is the average major axis, and the average value of the minor axes of the selected second copper phases 3c is the average minor axis. The aspect ratio of the second copper phase 3c is the average major axis divided by the average minor axis of the second copper phase 3c.

[0058] The major axis of the second copper phase 3c is, for example, the maximum length of the second copper phase 3c in the X direction. The minor axis of the second copper phase 3c is, for example, the maximum length of the second copper phase 3c in the Z direction.

[0059] <Average minor axis> The average minor axis of the second copper phase 3c is, for example, 5 μm or more and 200 μm or less. When the average minor axis of the second copper phase 3c is 5 μm or more, the thermal conductivity of the composite material 1 is likely to be high. When the average minor axis of the second copper phase 3c is 200 μm or less, the linear expansion coefficient of the composite material 1 is likely to be low. The average minor axis of the second copper phase 3c may be 5 μm or more and 100 μm or less, or even 5 μm or more and 50 μm or less.

[0060] The average major axis of the second copper phase 3c is, for example, 50 μm or more and 2000 μm or less, and further, 100 μm or more and 1000 μm or less.

[0061] <Relationship between Cu content and Vickers hardness of composite phase> The composite phase 4 is a structure formed by the matrix 2 and the first copper phase 3f. In the composite material 1, the content ratio X [mass%] of Cu and the Vickers hardness Y [Hv] of the composite phase 4 satisfy the relationship of the following formula (1). (1) Y≦-1.5551X+285

[0062] The Vickers hardness of the composite phase 4 can be determined as follows: A cross section of the composite material 1 is obtained. The cross section of the composite material 1 is a cross section perpendicular to the first surface 1a. The cross section of the composite material 1 is, for example, an XZ cross section. The Vickers hardness of the composite phase 4 at the cross section of the composite material 1 is measured. The Vickers hardness is measured using a micro Vickers hardness tester. The Vickers hardness of the composite phase 4 is determined from the size of the indentation made when an indenter is pressed into the cross section of the composite phase 4 with a constant load. When measuring the Vickers hardness of the composite phase 4, the indenter is made to contact the composite phase 4 while avoiding the second copper phase 3c. The Vickers hardness is measured at different positions on the composite phase 4. The number of Vickers hardness measurement points is, for example, 5 or more, or even 10 or more. The Vickers hardness of the composite phase 4 is the average value of the measured Vickers hardnesses. The test load is, for example, 10 gf or more and 300 gf or less. 1 gf (gram force) is approximately 9.8 mN (millinewtons). The test temperature is room temperature. Room temperature means 5°C or higher and 35°C or lower.

[0063] <Coefficient of linear expansion> The linear expansion coefficient of the composite material 1 is, for example, 8.4 × 10 -6 / K or less. The linear expansion coefficient of the composite material 1 means the linear expansion coefficient in a direction parallel to the first surface 1a in the temperature range from room temperature to 800°C. In other words, the linear expansion coefficient of the composite material 1 is the linear expansion coefficient in a direction perpendicular to the Z direction. In this example, the linear expansion coefficient of the composite material 1 is the linear expansion coefficient in the X direction. The linear expansion coefficient of the composite material 1 is 8.1 × 10 -6 / K or less, 7.8×10 -6 / K or less, and 7.0 × 10 -6 / K or less is also acceptable.

[0064] It is desirable that the linear expansion coefficient of composite material 1 is close to that of semiconductor elements or ceramic parts. Composite material 1 is used, for example, as a material for heat sinks to dissipate heat generated by heat sources such as semiconductor elements. If the difference between the linear expansion coefficient of composite material 1 and that of semiconductor elements is large, thermal stress will occur at the interface between them, which could result in damage to the semiconductor elements. If the linear expansion coefficient of composite material 1 is close to that of semiconductor elements, the difference in linear expansion coefficients will be small, and the thermal stress occurring at the interface will be small.

[0065] The linear expansion coefficient can be measured using a thermal dilatometer. For example, a TD5000SA manufactured by Bruker AXS can be used as the thermal dilatometer. The linear expansion coefficient of the composite material 1 is calculated by measuring the change in length in a direction parallel to the first surface 1a when the temperature changes in a temperature range from room temperature to 800°C. A sample used for measuring the linear expansion coefficient is cut out from the composite material 1. The shape of the sample is a rectangular plate. The planar size of the sample is 5 mm × 15 mm. The linear expansion coefficient of the composite material 1 is the average value of the linear expansion coefficients of three samples.

[0066] <Thermal conductivity> The thermal conductivity of the composite material 1 is, for example, 180 W / m·K or more. The thermal conductivity of the composite material 1 means the thermal conductivity in the direction perpendicular to the first surface 1a at room temperature. In other words, the thermal conductivity of the composite material 1 is the thermal conductivity in the Z direction. The thermal conductivity of the composite material 1 may be 200 W / m·K or more, 220 W / m·K or more, or even 230 W / m·K or more.

[0067] The thermal conductivity of composite material 1 can be measured using the laser flash method. It is calculated based on the thermal diffusivity of composite material 1 and the volume ratio and specific heat of the materials that make up composite material 1. A NETZSCH LFA457 MicroFlash can be used to measure the thermal diffusivity. The sample used to measure the thermal diffusivity is cut from composite material 1. The sample is a 10 mm diameter disk. Prior to measuring the thermal conductivity of composite material 1, the thermal conductivity of a pure copper sample of the same shape is measured under the same conditions. The thermal conductivity of the pure copper sample is used as a reference to correct the measured thermal conductivity of composite material 1. The specific heats of the materials that make up composite material 1 are determined based on the "Metal Data Book, 4th Edition" (2004, Maruzen Publishing), edited by the Japan Institute of Metals. The specific heat of copper is 386 J / (kg·K). The specific heat of molybdenum is 251 J / (kg·K). The thermal conductivity of Composite Material 1 is the average of the thermal conductivities of the three samples.

[0068] <<Method for manufacturing composite materials>> The composite material 1 can be produced by a composite material production method according to an embodiment. As shown in Fig. 3, the composite material production method includes a mixing step S1, a molding step S2, a sintering step S3, an infiltration step S4, and a compression step S5. For an explanation of the configuration of the composite material 1, see Fig. 2.

[0069] (Mixing process) The mixing step S1 is a step of mixing copper powder and molybdenum powder to obtain a mixed powder. The copper powder contains Cu particles formed of copper or a copper alloy. The molybdenum powder contains Mo particles formed of molybdenum or a molybdenum alloy. The Cu particles and Mo particles may have any shape. The particle shape may be, for example, spherical, fibrous, or plate-like.

[0070] The copper powder has an average particle size of 50 μm or more and 200 μm or less. The average particle size refers to the average particle size measured by the Fischer Subsieve Sizer (FSSS) method. When the copper powder has an average particle size of 50 μm or more, a composite material 1 containing a second copper phase 3c, as shown in FIG. 2, can be produced. The larger the average particle size of the copper powder, the larger the size of the second copper phase 3c. When the average particle size of the copper powder is 200 μm or less, the copper powder is easily dispersed uniformly in the mixed powder. As a result, the second copper phase 3c is easily dispersed uniformly in the matrix 2. The copper powder may have an average particle size of 100 μm or more and 200 μm or less.

[0071] The average particle size of the molybdenum powder is, for example, 1 μm or more and 10 μm or less. When the average particle size of the molybdenum powder is 1 μm or more and 10 μm or less, a sintered body having many fine voids is easily obtained in the sintering step S3. When the molten copper penetrates these fine voids, it is easy to produce a composite material 1 in which the first copper phase 3f is uniformly dispersed in the matrix 2. Furthermore, when the average particle size of the molybdenum powder is 1 μm or more, the molybdenum powder is easy to handle. The average particle size of the molybdenum powder may be 3 μm or more and 7 μm or less.

[0072] A known method can be used to mix the copper powder and the molybdenum powder. A mixer such as a V-type mixer can be used for mixing. The copper powder content in the mixed powder is, for example, 10% by mass or more and 70% by mass or less. The copper powder content refers to the mass percentage of the copper powder when the mass of the entire mixed powder is 100% by mass. The higher the copper powder content, the more the second copper phase 3c increases. Therefore, the area percentage of the second copper phase 3c in the cross section of the composite material 1 increases.

[0073] (molding process) The compacting step S2 is a step of press-molding the mixed powder to obtain a compact. In the compacting step S2, for example, the mixed powder is pressed at 490 MPa or less, i.e., 5 ton / cm. 2The press molding is performed at a pressure of 490 MPa or less. When the press molding pressure is 490 MPa or less, voids are easily formed inside the molded body. As a result, a porous sintered body is obtained in the sintering step S3. The press molding pressure is set so that voids remain inside the molded body while the shape of the molded body is maintained. The higher the press molding pressure, the smaller the voids in the molded body. The press molding pressure is, for example, 49 MPa or more and 392 MPa or less. When the press molding pressure is 49 MPa or more, the shape of the molded body is easily maintained. When the press molding pressure is 392 MPa or less, a molded body with a large number of fine voids is easily obtained. The press molding pressure may be 78 MPa or more and 294 MPa or less.

[0074] (Sintering process) The sintering step S3 is a step in which the green body is sintered to obtain a sintered body. The green body shrinks during sintering. Depending on the sintering temperature, the volume of the green body shrinks by approximately 5% to 20% after sintering. In other words, the size of the sintered body is smaller than that of the green body. In the sintering step S3, for example, the green body is sintered at a temperature of 1100°C or higher and 1400°C or lower. A sintering temperature of 1100°C or higher makes it easier to obtain a sintered body in which Mo particles are bonded together. This sintered body becomes the matrix 2 of the composite material 1. The matrix of the sintered body has a skeletal structure formed by the bonding of Mo particles, and therefore has higher mechanical strength than the matrix of the green body. Therefore, the mechanical strength of the composite material 1 is increased. If the sintering temperature is too high, the bonding between Mo particles progresses too much, causing the Mo particles to coarsen. This coarsening of the Mo particles tends to harden the matrix 2. As a result, the composite phase 4 hardens, which tends to reduce the thermal conductivity of the composite material 1. By setting the sintering temperature to 1400°C or lower, the Mo particles are less likely to coarsen. Therefore, the composite phase 4 is less likely to harden, and the thermal conductivity of the composite material 1 is likely to be high. The sintering temperature may be 1150°C or higher and 1350°C or lower, or even 1200°C or higher and 1300°C or lower. If the sintering temperature is higher than the melting point of the copper powder, the copper powder melts and becomes liquid when the compact is sintered. The sintering time held at the sintering temperature is, for example, 1 hour or higher and 4 hours or lower. The sintering time may also be 1.5 hours or higher and 3 hours or lower. The compact may be sintered in a non-oxidizing atmosphere such as a vacuum or hydrogen atmosphere. By sintering the compact in a non-oxidizing atmosphere, the copper powder or molybdenum powder is less likely to be oxidized.

[0075] The heating rate during sintering of the compact is 0.1°C / second (°C / s) or higher. A heating rate of 0.1°C / s or higher prevents the Mo particles from coarsening. This prevents the composite phase 4 from hardening, resulting in a high thermal conductivity of the composite material 1. The heating rate is calculated from the temperature rise ΔT from the starting temperature Ti before sintering to the set sintering temperature Ts and the time t required to reach the sintering temperature. The temperature rise ΔT is the temperature difference between the sintering temperature Ts and the starting temperature Ti, expressed as [Ts-Ti]. The time t is the time required to reach the sintering temperature after starting the heating. The heating rate is expressed as [ΔT / t]. The heating rate may be 0.12°C / s or higher, and may even be 0.13°C / s or higher. The faster the heating rate, the higher the thermal conductivity of the composite material 1 tends to be. The heating rate may be 0.1°C / s or higher and 0.5°C / s or lower, or even 0.12°C / s or higher and 0.4°C / s or lower.

[0076] (Infiltration process) The infiltration step S4 is a step in which a copper material is infiltrated into the sintered body to obtain an infiltrated body. The copper material is made of copper or a copper alloy. To infiltrate the copper material, the copper material is placed on the sintered body and heated to a temperature above the melting point of the copper material. The molten copper material penetrates the sintered body and fills the voids remaining inside the sintered body. By infiltrating the copper material, a composite material 1 can be produced in which a first copper phase 3f and a second copper phase 3c are dispersed in a matrix 2.

[0077] In the infiltration step S4, for example, the copper material is heated to a temperature of 1200°C or higher. By setting the infiltration temperature at 1200°C or higher, the copper material is sufficiently melted and easily infiltrated into the sintered body. The infiltration temperature may be 1250°C or higher, or even 1300°C or higher. The upper limit of the infiltration temperature is, for example, 1400°C. The infiltration temperature may be 1200°C or higher and 1400°C or lower, or even 1300°C or higher and 1400°C or lower. The infiltration time held at the infiltration temperature is, for example, 1 hour or higher and 4 hours or lower. The infiltration time may be 1.5 hours or higher and 3 hours or lower. The infiltration of the copper material may be performed in a non-oxidizing atmosphere such as a hydrogen atmosphere. Infiltrating the copper material in a non-oxidizing atmosphere makes it difficult for the molten copper to oxidize. If the infiltration temperature is higher than the sintering temperature described above, the sintered body may shrink slightly during infiltration. In other words, the size of the infiltrated body will be slightly smaller than the size of the sintered body.

[0078] (Compression process) The compression step S5 is a step of compressing the infiltrant. The compression process thins the infiltrant. Compression refers to a process of reducing the thickness of the infiltrant by applying pressure. In other words, compression is a process of compressing the infiltrant in the Z direction. Examples of compression processes include rolling, forging, and pressing. Compression may be performed warm. Heating the infiltrant to compress it facilitates deformation and makes it easier to compress. The processing temperature is, for example, 60°C to 350°C, and preferably 70°C to 350°C.

[0079] The infiltrant is compressed to obtain the composite material 1. The compressed composite material 1 is densified and tends to have high thermal conductivity. Furthermore, the compression process causes the second copper phase 3c to elongate in a direction parallel to the first surface 1a and deform into a flat shape, shortening the length of the second copper phase 3c in the Z direction.

[0080] In the compression step S5, the infiltrant is compressed at a processing rate of 30% or more, for example. The processing rate refers to the ratio of the processing amount ΔC divided by the thickness C0 before compression. The processing amount ΔC is the difference between the thickness C0 before compression and the thickness C1 after compression, and is expressed as [C0-C1]. The processing rate is expressed as a percentage of [ΔC / C0] = [(C0-C1) / C0]. The higher the processing rate, the more likely the composite material 1 is to be densified. A processing rate of 30% or more makes it easier to obtain a dense composite material. Furthermore, if the processing rate is 30% or more, the second copper phase 3c elongates in the X and Y directions, and the length of the second copper phase 3c in the Z direction is likely to be shorter. The processing rate may be 50% or more, or even 60% or more. If the processing rate is too high, the infiltrant may crack. The upper limit of the processing rate is, for example, 95% or less. The processing rate may be 30% or more and 95% or less, and further may be 50% or more and 75% or less.

[0081] In the compression step S5, the infiltrant is rolled at a temperature of, for example, 60°C or higher and 350°C or lower. Warm rolling of the infiltrant facilitates compression of the infiltrant. When the infiltrant is rolled, the second copper phase 3c is elongated in the rolling direction (RD). The rolling direction (RD) is the X direction. In the rolled composite material 1, the length of the second copper phase 3c in the X direction tends to increase.

[0082] 《Heat sink》 A heat sink 10 according to an embodiment will be described with reference to FIG. 4. The heat sink 10 has a plate-like shape. The heat sink 10 has a first surface 10a and a second surface 10b. The second surface 10b is the opposite surface to the first surface 10a. FIG. 4 shows a cross section perpendicular to the first surface 10a of the heat sink 10. In other words, FIG. 4 shows a cross section cut in the direction along the thickness of the heat sink 10, i.e., in the Z direction. The cross section of the heat sink 10 shown in FIG. 4 is an XZ cross section.

[0083] The heat sink 10 is a laminate in which first layers 11 and second layers 12 are alternately stacked in the Z direction. The heat sink 10 includes at least one first layer 11 and multiple second layers 12. The first layer 11 is formed from the composite material 1 described above. The second layer 12 is formed from copper or a copper alloy. The second layer 12, which is formed from copper or a copper alloy, has a higher thermal conductivity than the first layer 11, which is formed from the composite material 1.

[0084] The heat sink 10 has a structure in which first layers 11 and second layers 12 are alternately laminated. Adjacent first layers 11 and second layers 12 are in contact with each other. The heat sink 10 has the first layers 11 formed from the composite material 1, and thus can have both high thermal conductivity and a low linear expansion coefficient. The heat sink 10 is used, for example, as a heat spreader.

[0085] (Number of layers) The number of first layers 11 is 1 or more. The number of second layers 12 is 2 or more. The total number of stacked first layers 11 and second layers 12 is an odd number of 3 or more. The number of first layers 11, the number of second layers 12, and the number of stacked layers are not particularly limited and can be selected appropriately. In the example shown in FIG. 4, the number of first layers 11 is 1, the number of second layers 12 is 2, and the total number of stacked layers is 3.

[0086] The layer forming the first surface 10a and the second surface 10b is the first layer 11 or the second layer 12. In other words, the layer forming the first surface 10a and the layer forming the second surface 10b are layers of the same material. In the example shown in FIG. 4, the first layer 11 and the second layer 12 are alternately stacked so that the second layer 12 is located at the first surface 10a and the second surface 10b. Unlike the example shown in FIG. 4, the first layer 11 and the second layer 12 may also be alternately stacked so that the first layer 11 is located at the first surface 10a and the second surface 10b.

[0087] The first surface 10a and the second surface 10b may be formed by the second layer 12. The first surface 10a and the second surface 10b are surfaces that transfer heat by coming into contact with other members. When the first surface 10a and the second surface 10b are formed by the second layer 12, heat is easily transferred between the heat sink 10 and other members. When the first surface 10a is a contact surface with a heat source such as a semiconductor element, heat from the heat source can be transferred efficiently.

[0088] The heat sink 10 may have a symmetrical structure with respect to a layer located at the center of the stacking direction. A heat sink 10 having such a symmetrical structure in the stacking direction is less likely to warp or deform due to temperature changes. In the example shown in Figure 4, the layer located at the center of the stacking direction is the first layer 11.

[0089] 4 has a three-layer structure in which, from the first surface 10a, the second layer 12 / the first layer 11 / the second layer 12 are stacked. The number of stacked first layers 11 and second layers 12 may be five or more. For example, a five-layer heat sink 10 has a structure in which, from the first surface 10a, the second layer 12 / the first layer 11 / the second layer 12 / the first layer 11 / the second layer 12 are stacked.

[0090] The thickness T1 of each first layer 11 and the thickness T2 of each second layer 12 may be the same or different. When the heat sink 10 includes a plurality of first layers 11, all of the first layers 11 may have the same thickness, or some of the first layers 11 may have different thicknesses. When the heat sink 10 includes a plurality of second layers 12, all of the second layers 12 may have the same thickness, or some of the second layers 12 may have different thicknesses. If all of the first layers 11 have the same thickness and all of the second layers 12 have the same thickness, a heat sink 10 having a structure symmetrical in the stacking direction can be obtained.

[0091] The thickness T1 of the first layer 11 is, for example, 0.15 mm or more. The thickness T2 of the second layer 12 is, for example, 0.10 mm or more. When the thickness T1 of the first layer 11 is 0.15 mm or more, the linear expansion coefficient of the heat sink 10 tends to be low. The thickness T1 of the first layer 11 may be 0.18 mm or more, 0.20 mm or more, or even 0.22 mm or more. When the thickness T2 of the second layer 12 is 0.10 mm or more, the thermal conductivity of the heat sink 10 tends to be high. The thickness T2 of the second layer 12 may be 0.12 mm or more, or even 0.15 mm or more. The upper limit of the thickness T1 is, for example, 70% of the thickness T0 of the heat sink 10. The upper limit of the thickness T2 is, for example, 40% of the thickness T0 of the heat sink 10. The upper limit of the thickness T1 may be 50% of the thickness T0. The upper limit of the thickness T2 may be 30% of the thickness T0.

[0092] The thermal conductivity and linear expansion coefficient of heat sink 10 can be adjusted by, for example, changing the thickness of each of first layer 11 and second layer 12, or by changing the number of layers stacked between first layer 11 and second layer 12. Therefore, heat sink 10 can have a high thermal conductivity and a linear expansion coefficient close to that of semiconductor elements or ceramic parts.

[0093] The thickness T0 of the heat sink 10 can be determined as follows: A cross section of the heat sink 10 is obtained. The cross section of the heat sink 10 is a cross section perpendicular to the first surface 10a. The cross section of the heat sink 10 is, for example, an XZ cross section. The thickness of the heat sink 10 is measured at the cross section of the heat sink 10. The thickness of the heat sink 10 is the length of the heat sink 10 along the Z direction. The length of the heat sink 10 along the Z direction is the distance between the first surface 10a and the second surface 10b. The thickness of the heat sink 10 is measured at three or more locations. The thickness T0 of the heat sink 10 is the average value of the measured thicknesses of the heat sink 10.

[0094] The thickness T1 of the first layer 11 and the thickness T2 of the second layer 12 can be determined as follows. The thickness of the first layer 11 and the thickness of the second layer 12 are measured in a cross section of the composite material 1. The thickness of the first layer 11 is the length of the first layer 11 along the Z direction. The thickness of the second layer 12 is the length of the second layer 12 along the Z direction. The thickness of the first layer 11 and the thickness of the second layer 12 are each measured at three or more locations. The thickness T1 of the first layer 11 is the average value of the measured thicknesses of the first layer 11. The thickness T2 of the second layer 12 is the average value of the measured thicknesses of the second layer 12.

[0095] As shown in FIG. 5, the heat sink 10 may have a plating layer 15 on at least one of the first surface 10a and the second surface 10b. The plating layer 15 may be present on either the first surface 10a or the second surface 10b, or on both surfaces. The plating layer 15 may be present only on a portion of at least one of the first surface 10a and the second surface 10b. The heat sink 10 shown in FIG. 6 has the plating layer 15 on both the first surface 10a and the second surface 10b. In the example shown in FIG. 5, the plating layer 15 is present on the entire surface of each of the first surface 10a and the second surface 10b. The plating layer 15 may be provided to cover the entire surface of the heat sink 10. The entire surface of the heat sink 10 refers to not only the first surface 10a and the second surface 10b, but also all surfaces, including the side surfaces. The plating layer 15 facilitates bonding of the heat sink 10 to another component using a bonding material. The other component may be, for example, a heat source such as a semiconductor device. The bonding material is, for example, solder, brazing material, or nano-silver paste. The plating layer 15 is formed of a metal that has good affinity with the bonding material. The material of the plating layer 15 is, for example, at least one metal selected from the group consisting of nickel, gold, silver, copper, and lead, or an alloy thereof.

[0096] Semiconductor Package A semiconductor package 100 according to the embodiment will be described with reference to Fig. 6. The semiconductor package 100 includes the above-described heat sink 10, a semiconductor element 30, a case member 40, a lid 41, and terminals 50a and 50b.

[0097] The heat sink 10 used in this example is a heat spreader for the semiconductor package 100. The semiconductor element 30 is a heat source during operation. The semiconductor element 30 is disposed on a first surface 10a of the heat sink 10.

[0098] The case member 40 is made of, for example, a ceramic material. The ceramic material is, for example, alumina (Al2O3). The case member 40 in this example has a rectangular frame shape. The case member 40 is disposed on the first surface 10a so as to surround the semiconductor element 30. The lid 41 is made of, for example, a ceramic material or a metal material. The lid 41 closes the opening of the case member 40.

[0099] The terminals 50a and 50b are inserted into the case member 40. First ends of the terminals 50a and 50b are located within a space defined by the first surface 10a, the case member 40, and the lid 41. Second ends of the terminals 50a and 50b are located outside the space. The terminals 50a and 50b are formed of, for example, a metal material. The metal material is, for example, Kovar. Although not shown, the first ends of the terminals 50a and 50b are electrically connected to the semiconductor element 30. The semiconductor package 100 is electrically connected to a device or circuit separate from the semiconductor package 100 via the second ends of the terminals 50a and 50b.

[0100] The semiconductor element 30 and the heat sink 10 are bonded together by a first bonding material (not shown). The case member 40 and the heat sink 10 are bonded together by a second bonding material (not shown). The first bonding material and the second bonding material are, for example, solder, brazing material, or nano-silver paste. The first bonding material and the second bonding material may be the same material or different materials. When the case member 40 and the heat sink 10 are bonded together, the heat sink 10 is exposed to a high temperature of, for example, approximately 800°C. The difference between the linear expansion coefficient of the heat sink 10 including the first layer 11 formed from the composite material 1 and the linear expansion coefficient of the case member 40 formed from ceramics is small. Therefore, thermal stress generated at the interface between the heat sink 10 and the case member 40 is small. Damage to the semiconductor element 30 and other components is unlikely to occur.

[0101] A heat sink 60 is attached to the second surface 10b of the heat sink 10. The heat sink 60 is, for example, a metal plate having a flow path formed therein through which a refrigerant flows. The heat sink 60 is not limited to this. The heat sink 60 may be, for example, a cooling fin. The material of the heat sink 60 is, for example, copper, a copper alloy, aluminum, or an aluminum alloy. The heat sink 60 and the heat sink 10 may be joined by a joining material (not shown). Unlike the example shown in FIG. 6, the heat sink 10 may also function as a heat sink. In this case, the heat sink 10 may be thicker or have a larger area.

[0102] [Test Example 1] Composite materials were fabricated and their properties were evaluated.

[0103] (Sample No. 1A-Sample No. 15A) Samples No. 1A to No. 15A are collectively referred to as Sample 1. Sample 1 was manufactured as follows: Copper powder and molybdenum powder were prepared. The copper powder was electrolytic copper powder. The copper powder contained 99.9% by mass or more of Cu, with the remainder consisting of unavoidable impurities. The average particle size of the copper powder was 106 μm or more, specifically 110 μm. The molybdenum powder contained 99.9% by mass or more of Mo, with the remainder consisting of unavoidable impurities. The average particle size of the molybdenum powder was 4.0 μm. The average particle size was determined by the FSSS method. The average particle sizes of the copper powder and molybdenum powder used are shown in Table 1. In Table 1, "Cu powder" refers to copper powder, and "Mo powder" refers to molybdenum powder.

[0104] Copper powder and molybdenum powder were mixed in a predetermined ratio to obtain a mixed powder. The mixing ratio of copper powder and molybdenum powder varied for each sample. The copper powder and molybdenum powder were mixed using a V-type mixer.

[0105] The mixed powder was pressed in a press to obtain a compact with a length of 30 mm, a width of 30 mm, and a thickness of 6 mm. The pressure for pressing was 135 MPa. The pressing was performed in a cold state.

[0106] The green body was sintered to obtain a sintered body. After sintering, the sintered body was cooled to room temperature. To sinter the green body, the green body was placed in a sintering furnace and sintered in a hydrogen atmosphere at approximately 1200°C for 2 hours. The heating rate when sintering the green body was 0.13°C / s. The heating rate was calculated from the temperature rise from the starting temperature before sintering to the sintering temperature and the time it took to reach that temperature. The starting temperature was room temperature. The heating rate and sintering temperature are shown in Table 1.

[0107] The sintered body was infiltrated with copper material to obtain an infiltrated body. After infiltration, the infiltrated body was cooled to room temperature. The copper material was a plate made of copper with a purity of 99.9% or more. To infiltrate the copper material, the sintered body was placed in a heating furnace with the copper material placed on top, and held in a hydrogen atmosphere at approximately 1350°C for 2 hours. The infiltration temperature is shown in Table 1.

[0108] The infiltrant was warm-rolled to obtain a composite material. The reduction ratio varied for each sample. The reduction ratio was calculated based on the thickness before and after rolling. The thickness before rolling was approximately 5 mm. Warm-rolling was performed by holding the infiltrant at a temperature between 70°C and 350°C. Warm-rolling was performed by heating the infiltrant to 350°C and then rolling it until it cooled to 70°C or below. Once it cooled to 70°C, it was heated again and rolled, and this process was repeated. In this case, the processing temperature ranged from 70°C to 350°C. The processing temperatures and reduction ratios are shown in Table 1.

[0109] (Sample No. 1B-Sample No. 4B) Samples No. 1B to No. 4B are collectively referred to as Sample No. 2. Sample No. 2 was manufactured using the same procedure as Sample No. 1, except that the heating rate during sintering of the compact was set to 0.05°C / s.

[0110] (Sample No. 101-Sample No. 105) Samples No. 101 to No. 105 are collectively referred to as Sample No. 3. Sample No. 3 was made using copper powder with an average particle size of 30 μm. Sample No. 3 was manufactured using the same procedure as described above, with the heating rate during sintering of the compact set at 0.08°C / s.

[0111] (evaluation) For each of the produced samples, the Cu content in the composite material, the aspect ratio and area ratio of the second copper phase, and the Vickers hardness, thermal conductivity, and linear expansion coefficient of the composite phase were measured.

[0112] <Cu content> The Cu content was measured by ICP-OES and is shown in Table 2.

[0113] <Aspect ratio and area ratio of the second copper phase> A cross section of the composite material was obtained from each sample. The cross section of the composite material was observed using an optical microscope. The microscope magnification was 100x. The observation area was 1150 μm horizontal x 850 μm vertical. The average minor axis and average major axis of the second copper phase were determined using the measurement method described above. The aspect ratio of the second copper phase was calculated from the average minor axis and average major axis of the second copper phase. The area proportion of the second copper phase in the cross section of the composite material was also determined using the measurement method described above. The average minor axis and average major axis of the second copper phase, the aspect ratio, and the area proportion are shown in Table 2. In Table 2, "minor axis" refers to the average minor axis, and "major axis" refers to the average major axis. The values ​​of the average minor axis, average major axis, and aspect ratio shown in Table 2 are rounded to one decimal place and expressed as integers.

[0114] No second copper phase was observed in Samples No. 101 to 105. In other words, the structure of Samples No. 101 to 105 was a structure in which only fine first copper phases were dispersed in the matrix.

[0115] <Vickers hardness of composite phase> A cross section of the composite material was obtained from each sample. The Vickers hardness of the composite phase in the cross section of the composite material was determined using the measurement method described above. The Vickers hardness of the composite phase is the average value of the Vickers hardness measured at five points. The Vickers hardness was measured in accordance with JIS Z 2244. The micro Vickers hardness tester used was an MXT30 manufactured by Matsuzawa Co., Ltd. The test load was 100 gf. 100 gf is approximately 0.98 N. The Vickers hardness of the composite phase was measured at room temperature. The Vickers hardness of the composite phase is shown in Table 2. Note that for Samples No. 101 to No. 105, the Vickers hardness of the cross section of the composite material was measured.

[0116] <Thermal conductivity> The thermal conductivity of each sample was measured based on the above-mentioned measurement method and is shown in Table 2.

[0117] <Coefficient of linear expansion> The linear expansion coefficient of each sample was measured using the measurement method described above. The linear expansion coefficient was determined in both the rolling direction (RD) and the transverse direction. RD corresponds to the X direction. TD corresponds to the Y direction. The linear expansion coefficients are shown in Table 2.

[0118] [Table 1]

[0119] [Table 2]

[0120] <Relationship between linear expansion coefficient and thermal conductivity> The relationship between the linear expansion coefficient and thermal conductivity of each sample was investigated. Figure 7 is a graph plotting the data for the linear expansion coefficient and thermal conductivity of each sample. The linear expansion coefficient is the linear expansion coefficient of RD. The horizontal axis of Figure 7 is the linear expansion coefficient [10 -6 / K], and the vertical axis represents thermal conductivity [W / m·K]. In Figure 7, black circles (●) represent the data for the first samples, Sample No. 1A to Sample No. 15A. Crosses (×) represent the data for the second samples, Sample No. 1B to Sample No. 4B. Open circles (○) represent the data for the third samples, Sample No. 101 to Sample No. 105. The dotted line in Figure 7 is the approximate line L1a, which shows the relationship between the linear expansion coefficient and thermal conductivity of the first sample. This approximate line L1a was obtained by linear approximation from the data for Sample No. 1A to Sample No. 15A. The dashed line in Figure 7 is the approximate line L2a, which shows the relationship between the linear expansion coefficient and thermal conductivity of the second sample. This approximate line L2a was obtained by linear approximation from the data for Sample No. 1B to Sample No. 4B. 7, the dashed two-dot line is an approximate line L3a that shows the relationship between the linear expansion coefficient and thermal conductivity of Sample 3. Approximate line L3a was obtained by linear approximation from the data of Samples No. 101 to 105.

[0121] The graph in Figure 7 reveals the following. Among the first, second, and third samples, samples with approximately the same linear expansion coefficients are compared. This comparison reveals that the first sample has a higher thermal conductivity than the second and third samples, even though they have the same linear expansion coefficient. In other words, the first sample has a lower linear expansion coefficient than the second and third samples, even though they have the same thermal conductivity. Therefore, the first sample has a higher thermal conductivity while having a lower linear expansion coefficient than the second and third samples.

[0122] <Relationship between Cu content and thermal conductivity> We investigated the relationship between the Cu content and thermal conductivity of each sample. Figure 8 is a graph plotting data on the Cu content and thermal conductivity of each sample. The horizontal axis of Figure 8 represents the Cu content (mass%), and the vertical axis represents thermal conductivity (W / m·K). In Figure 8, black circles (●) represent the data for the first samples, Samples No. 1A to No. 15A. Crosses (×) represent the data for the second samples, Samples No. 1B to No. 4B. Open circles (○) represent the data for the third samples, Samples No. 101 to No. 105. The dotted line in Figure 8 is the approximate line L1b, which shows the relationship between the Cu content and thermal conductivity of the first sample. Approximate line L1b was calculated by linear approximation using the data for Samples No. 1A to No. 15A. In Figure 8, the dashed line represents the approximate line L2b, which shows the relationship between the Cu content and thermal conductivity of the second sample. The approximate line L2b was obtained by linear approximation from the data for each of Samples No. 1B to No. 4B. In Figure 8, the dashed line represents the approximate line L3b, which shows the relationship between the linear expansion coefficient and thermal conductivity of the third sample. The approximate line L3b was obtained by linear approximation from the data for each of Samples No. 101 to No. 105.

[0123] The graph in Figure 8 reveals the following: It can be seen that thermal conductivity increases in proportion to the Cu content. Among the first, second, and third samples, samples with approximately the same Cu content are compared. From this comparison, it can be seen that the first sample has a higher thermal conductivity than the second and third samples, even though they have the same Cu content. Therefore, it can be said that the first sample has a higher thermal conductivity than the second and third samples.

[0124] <Relationship between Cu content and Vickers hardness of composite phase> The relationship between the Cu content of the first and second samples and the Vickers hardness of the composite phase was investigated. Figure 9 is a graph plotting data on the Cu content of each sample and the Vickers hardness of the composite phase. The horizontal axis of Figure 9 represents the Cu content [mass %], and the vertical axis represents the Vickers hardness [Hv]. In Figure 9, black circles (●) represent the data for each of the first samples, Samples No. 1A to No. 15A. Crosses (×) represent the data for each of the second samples, Samples No. 1B to No. 4B.

[0125] The graph in Figure 9 reveals the following. Comparing the first and second samples, which have roughly the same Cu content, it is clear that the Vickers hardness of the composite phase of the first sample is lower than that of the second sample. In other words, the composite phase of the second sample is harder than that of the first sample.

[0126] From the graph in FIG. 9, the approximate line Lx of the boundary between the first sample and the second sample was determined. In FIG. 9, the approximate line Lx is shown by a dashed line. The approximate line Lx was determined by linear approximation. The equation representing the approximate line Lx was Lx = -1.5551X + 285, where X is the Cu content [mass %]. As shown in FIG. 9, the Vickers hardness of the composite phase of the first sample is lower than the approximate line Lx. The Vickers hardness of the composite phase of the second sample is higher than the approximate line Lx. The Vickers hardness Y [Hv] of the composite phase of the first sample satisfies the above-mentioned formula (1).

[0127] As described above, the graphs in Figures 7 and 8 show that the first sample has a lower coefficient of linear expansion and a higher thermal conductivity than the second sample, even though the Cu content is the same. Furthermore, the graph in Figure 9 shows that when the Cu content X [mass%] and the Vickers hardness Y [Hv] of the composite phase satisfy the above-mentioned formula (1), the first sample has a lower coefficient of linear expansion and a higher thermal conductivity. [Explanation of symbols]

[0128] 1 Composite materials 1a 1st side, 1b 2nd side 2. Matrix 3f First copper phase, 3c Second copper phase 4 Composite phase 10 Heat sink 10a 1st surface, 10b 2nd surface 11 1st layer, 12 2nd layer 15 plating layer 30 semiconductor element, 40 case member, 41 lid, 50a, 50b terminal 60 Heatsink 100 Semiconductor Packages T0, T1, T2 thickness S1 mixing process, S2 forming process, S3 sintering process S4 infiltration process, S5 compression process

Claims

1. A composite material comprising copper and molybdenum, a plate-like shape having a first surface as a main surface; a composition having a copper content of 20% by mass or more and 70% by mass or less; a structure including a matrix containing the molybdenum and a first copper phase and a second copper phase dispersed in the matrix, In a cross section perpendicular to the first surface, The cross-sectional area of ​​each of the second copper phases is 900 μm ２ That's all, a cross-sectional area of ​​each of the first copper phases is smaller than a cross-sectional area of ​​the second copper phase; a ratio of the total area of ​​the second copper phase to the area of ​​the cross section is 10% or more and 50% or less, the second copper phase has an aspect ratio of 10 or more, the aspect ratio being calculated by dividing the average major axis by the average minor axis; The relationship Y≦−1.5551X+285 is satisfied, X is the content of copper in the composite material, Y is the Vickers hardness of the composite phase formed by the matrix and the first copper phase; Composite material.

2. The composite material according to claim 1 , wherein the average minor axis is 5 μm or more and 200 μm or less.

3. The linear expansion coefficient from room temperature to 800°C is 8.4 x 10 －６ 3. The composite material according to claim 1 or claim 2, wherein the viscosity is 1 / K or less.

4. 3. The composite material according to claim 1, having a thermal conductivity of 180 W / m·K or more.

5. A heat sink having a first surface and a second surface opposite to the first surface, at least one first layer and a plurality of second layers; The first layer and the second layer have a structure in which they are alternately stacked, the first layer is formed from the composite material according to claim 1 or 2; the second layer is formed of copper or a copper alloy; Heat sink.

6. The heat sink of claim 5 , wherein the layer forming the first surface and the second surface is the second layer.

7. The heat sink according to claim 5 , further comprising a plating layer on at least one of the first surface and the second surface.

8. A heat sink comprising: the heat sink according to claim 5; and a semiconductor element; the semiconductor element is disposed on the first surface; Semiconductor package.

9. mixing copper powder and molybdenum powder to obtain a mixed powder; a step of press-molding the mixed powder to obtain a molded body; sintering the compact to obtain a sintered body; a step of infiltrating the sintered body with a copper material to obtain an infiltrated body; and compressing the infiltrant, The copper powder has an average particle size of 50 μm or more and 200 μm or less, In the step of obtaining the sintered body, the temperature rising rate when sintering the molded body is 0.13°C / sec or more and 0.5°C / sec or less. Composite material manufacturing methods.

10. The method for producing a composite material according to claim 9 , wherein in the step of obtaining the sintered body, the compact is sintered at a temperature of 1100° C. or higher and 1400° C. or lower.

11. 11. The method for producing a composite material according to claim 9, wherein the step of obtaining the infiltrant comprises heating the copper material to a temperature of 1200°C or higher and 1400°C or lower.

12. 11. The method for producing a composite material according to claim 9, wherein in the step of compressing the infiltrant, the infiltrant is compressed at a compression rate of 30% to 95%.

13. 11. The method for producing a composite material according to claim 9, wherein in the step of compressing the infiltrant, the infiltrant is rolled at a temperature of 60°C or higher and 350°C or lower.

Images