Heat dissipation components and their manufacturing methods

Abstract

This invention provides a heat dissipation component and a method for manufacturing the same. The heat dissipation component is a plate-shaped heat dissipation component comprising a metal-silicon carbide composite containing aluminum or magnesium. At least one of the two main surfaces of the heat dissipation component is convexly curved outward from the heat dissipation component. Let the flatness of one of its main surfaces, as specified by JIS B 0621, be f1, and let the flatness of the other main surface, which is a different main surface from the first main surface, be f2, as specified by JIS B 0621, be f2, where f2 is at least 10 μm smaller than f1.

Description

[0001] This application is a divisional application of Chinese invention patent application No. 202080006024.1 (PCT application number PCT / JP2020 / 003089), filed on January 29, 2020, entitled "Heat Dissipation Component and Manufacturing Method Thereof". Technical Field

[0002] This invention relates to heat dissipation components and methods for manufacturing the same. More specifically, it relates to plate-shaped heat dissipation components comprising a metal-silicon carbide composite containing aluminum or magnesium, and methods for manufacturing the same. Background Technology

[0003] In recent years, heat dissipation components for power modules used in electric vehicles and electric railways have gradually replaced copper with heat dissipation components made of metal-silicon carbide composites.

[0004] As a metal-silicon carbide complex, aluminum and its alloys are frequently used.

[0005] Heat dissipation components are often used in conjunction with other components (such as heat sinks and heat dissipation units), and the characteristics of the joints are important.

[0006] For example, when joining heat dissipation components with heat sinks or heat dissipation units, holes are usually provided on the periphery of the heat dissipation component to thread it onto the heat sink, heat dissipation unit, etc. However, if the surface of the heat dissipation component that contacts the heat sink is concave, or if there are a large number of tiny bumps and depressions, gaps will be generated between the heat dissipation component and the heat sink, resulting in reduced heat transfer.

[0007] In view of the above problems, in order to minimize the gap between the heat dissipation component and the heat sink, several solutions have been proposed for heat dissipation components with convex bending surfaces that engage with the heat sink.

[0008] The reason is that, as mentioned above, heat dissipation components are usually fixed to heat sinks and the like with screws or other fixing components. In this case, by making the mating surface with the heat sink and the like convex, the mating surface becomes "moderately flat" when fixed by the fixing component, and the mating (tightness) with the heat sink and the like is improved.

[0009] For example, Patent Document 1 describes a silicon carbide composite material characterized by being a plate-shaped composite material obtained by impregnating a porous silicon carbide molded body with a metal mainly composed of aluminum. The plate-shaped composite material has four or more holes in its surface for threading the convex surface of the plate-shaped composite material toward other heat dissipation components. The warpage amount (Cx; μm) relative to the length of 10 cm in the direction between the holes (X direction) and the warpage amount (Cy; μm) relative to the length of 10 cm in the direction perpendicular to the direction between the holes (Y direction) are related as follows: 50 ≤ Cx ≤ 250 and -50 ≤ Cy ≤ 200 (excluding Cy = 0).

[0010] As another example, Patent Document 2 describes a silicon carbide composite, which is a plate-shaped composite obtained by impregnating a porous silicon carbide molded body with a metal mainly composed of aluminum, and has a warpage of less than 250 μm relative to the length of the main surface of the composite of 10 cm.

[0011] Existing technical documents

[0012] Patent Literature

[0013] Patent Document 1: Japanese Patent No. 3468358

[0014] Patent Document 2: International Publication No. 2015 / 115649 Summary of the Invention

[0015] Problems to be solved by the invention

[0016] As described above, it is known that (1) firstly, a bent heat dissipation component is manufactured, and (2) when it is joined with a heat sink or the like, the bending is flattened by the force of a screw, thereby improving the fit between the heat dissipation component and the heat sink or the like, and thus improving the heat dissipation performance.

[0017] However, typically, electrical components and other parts are connected to the side of the heat dissipation component opposite to the side that contacts the heat sink. Therefore, especially during mass production, it is sometimes difficult to align the connecting parts of the bent heat dissipation component, or the connection of the parts themselves is difficult.

[0018] That is, when manufacturing power modules and the like using one side of a bent heat dissipation component as a connecting part, there is room for improvement in manufacturing stability (yield, etc.).

[0019] The present invention was made in view of the above circumstances.

[0020] One of the objectives of this invention is to improve manufacturing stability (yield, etc.) when manufacturing power modules or the like by connecting components to one side of a bent heat dissipation member.

[0021] Methods for solving problems

[0022] The inventors of this application conducted in-depth research to solve the aforementioned problems. As a result, they discovered that the flatness of the two main surfaces of the plate-shaped heat dissipation component, as specified in JIS B 0621, may be closely related to solving the problems. Based on this understanding, the invention described below was completed.

[0023] According to the present invention, a heat dissipation component is provided, which is a plate-shaped heat dissipation component comprising a metal-silicon carbide composite containing aluminum or magnesium.

[0024] At least one of the two main surfaces of the heat dissipation component is convexly curved outwards.

[0025] Let the flatness of one of the main surfaces, as defined by JIS B 0621, be f1, and let the flatness of another main surface, which is different from the one main surface, be f2, as defined by JIS B 0621. In this case, f2 is more than 10 μm smaller than f1.

[0026] Furthermore, according to the present invention, a method for manufacturing the heat dissipation component is provided, comprising the following steps:

[0027] The process of preparing plate-shaped metal-silicon carbide composites containing aluminum or magnesium, and

[0028] The process of machining at least a portion of one side of the composite to form the main surface.

[0029] This invention provides the following technical solution:

[0030] (1) Heat dissipation component, which is a plate-shaped heat dissipation component containing a metal-silicon carbide composite material containing aluminum or magnesium.

[0031] At least one of the two main surfaces of the heat dissipation component is convexly curved outwards from the heat dissipation component.

[0032] Let the flatness of one of the main surfaces, as defined by JIS B 0621, be f1. Let the flatness of another main surface, which is different from the one main surface, be f2, as defined by JIS B 0621, be f2, where f2 is more than 10 μm smaller than f1.

[0033] (2) In the heat dissipation component described in (1) above, one main surface and / or the other main surface have a surface metal layer containing aluminum or magnesium.

[0034] (3) In the heat dissipation component described in (1) or (2) above, f1 is 100 μm or more and 700 μm or less.

[0035] (4) In any one of the heat dissipation components (1) to (3) above, f2 is 300 μm or less.

[0036] (5) In any one of (1) to (4) above, the heat dissipation component is substantially rectangular. Let the length of the long side of the rectangle be a, the length of the short side be b, the straight line connecting the midpoints of the two short sides of the main surface be l1, the straight line connecting the midpoints of the two long sides of the main surface be l2, the maximum distance between a point on the curve formed by the main surface when the heat dissipation component is viewed in section containing l1 and approximately perpendicular to the main surface and l1 is h1, and the maximum distance between a point on the curve formed by the main surface when the heat dissipation component is viewed in section containing l2 and approximately perpendicular to the main surface and l2 is h2. In this case, h1 / a≥h2 / b.

[0037] (6) In the heat dissipation component described in (5) above, the value of (h1 / a) / (h2 / b) is 1.00 or more and 1.9 or less.

[0038] (7) In any one of the heat dissipation components described in (1) to (6) above, the average length RS of the roughness curve element of the main surface is... m It is between 50μm and 250μm.

[0039] (8) In any one of the heat dissipation components described in (1) to (7) above, the average length RS of the roughness curve element of the other main surface m It is between 50μm and 200μm.

[0040] (9) A method for manufacturing a heat dissipation component according to any one of (1) to (8) above, comprising the following steps: preparing a plate-shaped metal-silicon carbide composite containing aluminum or magnesium, and machining a portion of at least one side of the composite to form the main surface.

[0041] Invention Effects

[0042] According to the present invention, manufacturing stability (yield, etc.) can be improved when manufacturing power modules, etc., by attaching components to one side of a bent heat dissipation member. Attached Figure Description

[0043] The above-mentioned objectives, as well as other objectives, features, and advantages, can be further illustrated by the preferred embodiments described below and the accompanying drawings.

[0044]

【 Figure 1 [Illustration] is a schematic diagram illustrating the heat dissipation component of this embodiment. Figure 1 (a) is a top view of the heat dissipation component of this embodiment. Figure 1 (b) is therefore Figure 1 (a) is a cross-sectional view of the heat dissipation component when section α is cut.

[0045]

【 Figure 2 The diagram is used to explain, in particular, at least one of the two main surfaces of the heat dissipation component of this embodiment. Figure 2 (a) is a diagram showing only one of the two main faces of the heat dissipation component. Figure 2 (b) is therefore Figure 2 (a) is a cross-sectional view of section β when the main face is cut. Figure 2 (c) is Figure 2 (a) is a cross-sectional view of the main face when section γ is cut off. Detailed Implementation

[0046] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[0047] In all the accompanying drawings, the same symbols are used for the same constituent elements, and descriptions are omitted where appropriate.

[0048] To avoid complexity, (i) if there are multiple identical constituent elements in the same drawing, sometimes only one of them is marked, and not all constituent elements are marked; (ii) especially Figure 2 After that, sometimes regarding... Figure 1 The same constituent elements will not be marked with symbols again.

[0049] All accompanying drawings are for illustrative purposes only. The shapes, dimensions, and proportions of the components in the drawings may not correspond to the actual objects. In particular, to make the explanations easier to understand, shapes and proportions are sometimes exaggerated. Specifically, the degree of "bend" in the drawings is exaggerated compared to the actual objects.

[0050] Unless otherwise specified, the word "approximately" in this specification means that it includes a range that takes into account manufacturing tolerances, assembly deviations, etc.

[0051] Unless otherwise specified, for all values ​​(especially measured values) in this instruction manual that may vary with temperature, the values ​​at room temperature (23°C) may be used.

[0052] <Heat dissipation components>

[0053] Figure 1 (a) is a top view of the heat dissipation member (heat dissipation member 1) of this embodiment.

[0054] Heat dissipation component 1 is plate-shaped.

[0055] The main material of heat dissipation component 1 is a metal-silicon carbide composite containing aluminum or magnesium (details of the material and the manufacturing method of heat dissipation component 1 will be explained later).

[0056] The heat dissipation component 1 can typically be substantially rectangular. That is, when viewed from above with one of its main surfaces as the top surface, the shape of the heat dissipation component 1 is substantially rectangular.

[0057] Here, "substantially rectangular" means that at least one of the four corners of the heat dissipation component 1 may not be a right angle, but may be processed into a shape with rounded corners (of course, the four corners may also be right angles).

[0058] In addition, when at least one of the four corners of the heat dissipation component 1 is machined into a rounded shape, the point where the straight sections of the short and long sides of the heat dissipation component 1 intersect when viewed from above can be defined as the "vertex" of the rectangle. Furthermore, the "short side length" and "long side length" of the heat dissipation component 1 can be defined with the aforementioned "vertex" as the starting or ending point.

[0059] As an example, the length of the heat dissipation component 1, both longitudinally and transversely, is approximately 40mm × 90mm to 250mm × 140mm.

[0060] Regarding the thickness of the heat dissipation component 1, as an example, it is 2 mm or more and 6 mm or less, preferably 3 mm or more and 5 mm or less. If the thickness of the heat dissipation component 1 is not uniform, it is preferable that at least the thickness of the center of gravity portion of the heat dissipation component 1 is within the above range.

[0061] Figure 1 (b) is to use heat dissipation component 1 as Figure 1 (a) is a cross-sectional view when the surface α is cut.

[0062] The plate-shaped heat dissipation component 1 has two main surfaces (one main surface is the main surface 2A, and the other main surface is the main surface 2B). Typically, the main surface 2A is the surface that is joined with the heat sink, etc., and the main surface 2B is the surface that is connected with the power components, etc.

[0063] In the heat dissipation component 1, in particular, at least the main surface 2A is convexly curved not towards the inside of the heat dissipation component 1, but towards the outside of the heat dissipation component 1. Generally, the main surface 2A is convexly curved outward as a whole, and does not have any locally concave parts.

[0064] Let the flatness of the main surface 2A as defined by JIS B 0621 be f1, and the flatness of the main surface 2B as defined by JIS B 0621 be f2, then f2 is more than 10 μm smaller than f1.

[0065] In existing heat dissipation components with bends, the degree of bend on the main surface 2A side is often basically the same as that on the main surface 2B side. Therefore, when connecting components such as power elements to the main surface 2B side, there are difficulties in alignment or the connection of the components themselves is difficult.

[0066] On the other hand, Figure 1 In the heat dissipation component 1, f1 (representing the curvature of the main surface 2A) is larger, and f2 (representing the curvature of the main surface 2B) is smaller. Therefore, on the main surface 2A side, sufficient bonding between the heat dissipation component 1 and the heat sink, etc., can be achieved, thereby obtaining adequate heat dissipation. Furthermore, compared to conventional heat dissipation components, on the main surface 2B side, the alignment and connection of components can be easily performed.

[0067] Flatness, as defined by JIS B 0621, is the minimum distance between two geometrically parallel planes that hold a planar object together.

[0068] For example, the VR-3000 device manufactured by KEYENCE can be used to measure flatness f1 and flatness f2.

[0069] In addition, the measurement of f1 is preferably performed on the entire main surface 2A. However, depending on the size of the heat dissipation member 1, sometimes the entire main surface 2A may not be included within the field of view of the measuring device. In this case, f1 is measured in a manner that the center (geometric centroid) of the main surface 2A when viewed from above coincides with the center of the field of view of the measuring device. In this case, it is preferable to make the long side of the essentially rectangular main surface 2A parallel to the long side of the field of view (the field of view of the flatness measuring device is usually rectangular). The measurement of f2 is performed in the same manner.

[0070] For heat dissipation component 1, a more detailed explanation will follow.

[0071] [Supplementary explanations for f1, f2, and main page 2B, etc.]

[0072] In heat dissipation component 1, f2 is at least 10 μm smaller than f1, preferably at least 50 μm smaller than f1, and more preferably at least 100 μm smaller than f1.

[0073] On the other hand, from the viewpoints of ease of manufacturing of the heat dissipation component 1 itself and appropriate amount of bending, it is preferable that the difference between f1 and f2 (f1-f2) is 600μm or less, more preferably 500μm or less, and even more preferably 400μm or less.

[0074] The value of f1 is preferably 100 μm or more and 700 μm, more preferably 200 μm or more and 600 μm or less.

[0075] The value of f2 is preferably 300 μm or less, more preferably 250 μm or less, even more preferably 200 μm or less, and particularly preferably 100 μm or less. As an example, the lower limit of f2 can be 0. Alternatively, as another example, the lower limit of f2 can be 50 μm or more.

[0076] It is believed that by properly designing the value of f1, a moderate (not excessive) installation force can be used to securely fix the heat dissipation component 1 to the heat sink with good connection. This is related to further improving heat dissipation and reducing cracks caused by excessive force.

[0077] It is believed that by properly designing the value of f2, the alignment and connection of components connected to the main surface 2B can be made easier.

[0078] The main surface 2B can be bent in a convex shape towards the outside of the heat dissipation member 1, or conversely, it can be bent in a concave shape towards the outside of the heat dissipation member 1. Taking into account the bending of the heat dissipation member 1 when manufacturing electrical equipment as the final product, it is preferable that the main surface 2B is bent in a concave shape towards the outside of the heat dissipation member 1.

[0079] More specifically, when the main surface 2B is concavely curved toward the outside of the heat dissipation member 1, and the degree of curvature is expressed in terms of flatness, as described above, it is preferably 300 μm or less, more preferably 250 μm or less, more preferably 200 μm or less, and particularly preferably 100 μm or less.

[0080] Of course, the main surface 2B can be bent in a convex shape towards the outside of the heat dissipation member 1. As long as the degree of bending represented by f2 is more than 10 μm smaller than f1, it is unlikely to cause adverse situations in the connection with power components, etc.

[0081] [Materials that constitute main surfaces 2A and 2B, etc.]

[0082] As an option, the main surface 2A and / or the main surface 2B (i.e. the surface of the heat dissipation component 1) can be a metal-silicon carbide composite containing aluminum or magnesium.

[0083] Alternatively, the main surface 2A and / or the main surface 2B (the surface of the heat dissipation member 1) can be a metal layer. For example, the main surface 2A and / or the main surface 2B preferably have a surface metal layer containing aluminum or magnesium. In this case, the portion of the heat dissipation member 1 other than the surface metal layer can be a metal-silicon carbide composite or the like. The metal contained in the metal surface layer is preferably the same as the metal contained in the metal-silicon carbide composite. This is due to reasons related to the manufacturing of the heat dissipation member 1 (the manufacturing method of the heat dissipation member 1 is described below).

[0084] Preferably, a plating layer is provided on the outer side of the aforementioned surface metal layer. Solder is commonly used in the connection of electrical components, and the plating layer can improve the wettability of the solder.

[0085] The plating layer can be, for example, a plating layer containing Ni.

[0086] When the heat dissipation component 1 has a metal surface layer, the average thickness of the surface metal layer is not particularly limited, but for example, it is 10 μm or more and 300 μm or less, preferably 30 μm or more and 150 μm or less.

[0087] When the heat dissipation component 1 has a plating layer, the average thickness of the plating layer is not particularly limited, but for example, it is 3 μm or more and 15 μm or less, preferably 4 μm or more and 10 μm or less.

[0088] Regarding the values ​​of f1 and f2, the flatness value of the outermost surface of the heat dissipation component 1 is adopted as a principle.

[0089] For example, when the heat dissipation component 1 is composed only of a plate-shaped metal-silicon carbide composite, the flatness of the two principal surfaces of the plate-shaped metal-silicon carbide composite can be measured as the values ​​of f1 and f2.

[0090] In addition, when the outermost surface of the heat dissipation component 1 is a surface metal layer or a plating layer, the flatness of the surface of the surface metal layer or plating layer of its outermost surface can be measured as the values ​​of f1 and f2.

[0091] [Regarding the surface roughness specifications of main surfaces 2A and 2B]

[0092] As another perspective, by appropriately designing the "surface roughness index" of main surfaces 2A and 2B, performance such as heat dissipation can be further improved.

[0093] For example, it is preferable to use the average length RS of the roughness curve feature of the main surface 2A. m The design should be 50μm or larger and 250μm or smaller, and more preferably 70μm or larger and 160μm or smaller.

[0094] By using RS of main face 2A m Setting it to an appropriate size further improves its fit with heat sinks and other components, resulting in better heat dissipation. The detailed mechanism is unclear, but it is believed that it may be achieved through RS... m By setting the appropriate size, the tiny gaps between the heat dissipation component 1 and the heat sink are reduced, and the fit is further improved.

[0095] In addition, it is preferable to set the average length RS of the roughness curve element of the main surface 2B. m The design should be 50μm or larger and 200μm or smaller, and more preferably 70μm or larger and 160μm or smaller.

[0096] It is believed that by passing RS of main face 2B m Setting it to an appropriate size can, for example, improve the connectivity between the heat dissipation component 1 and the power components.

[0097] [Regarding the curvature of main face 2A]

[0098] Figure 2 This is a diagram used to illustrate the shape of the main surface 2A of the heat dissipation component 1.

[0099] Figure 2 (a) is a diagram showing only the main surface 2A of the heat dissipation component 1. Figure 2 (b) is therefore Figure 2 (a) is a cross-sectional view of the main face 2A when section β is cut off. Figure 2 (c) is Figure 2 (a) is a cross-sectional view of the main plane 2A when section γ cuts through it. In these figures, auxiliary lines, etc., are shown for the purposes of the following explanation.

[0100] Here, let the length of the long side of the rectangular heat dissipation component 1 be a, the length of the short side be b, the straight line connecting the midpoints of the two short sides of the main surface 2A (represented by M1 and M3 in the figure) be l1, and the straight line connecting the midpoints of the two long sides of the main surface 2A (represented by M2 and M4 in the figure) be l2.

[0101] Furthermore, let h1 be the maximum distance between a point on the curve formed by the main surface 2A and l1 when the heat dissipation member 1 is viewed in section β, which includes l1 and is approximately perpendicular to the main surface 2A, and let h2 be the maximum distance between a point on the curve formed by the main surface 2A and l2 when the heat dissipation member 1 is viewed in section γ, which includes l2 and is approximately perpendicular to the main surface 2A.

[0102] At this point, it is preferable to design the heat dissipation component 1 in the manner of h1 / a≥h2 / b.

[0103] More specifically, the value of (h1 / a) / (h2 / b) is preferably 1.00 or more and 1.9 or less, and more preferably 1.07 or more and 1.6 or less.

[0104] The above inequality can be explained as follows.

[0105] It can be said that h1 / a refers to the "bending per unit length in the long side direction" of the heat dissipation component 1.

[0106] Similarly, h2 / b can be said to refer to the "bending per unit length in the short side direction" of the heat dissipation component 1.

[0107] At this point, "h1 / a≥h2 / b" can be interpreted as follows: even excluding the factor that the longer side is longer than the shorter side, the "degree of curvature" in the direction of the longer side is the same as the "degree of curvature" in the direction of the shorter side, or the "degree of curvature" in the direction of the longer side is greater than the "degree of curvature" in the direction of the shorter side.

[0108] While some aspects remain unexplained, when a bent heat sink component 1 is joined to other components using screws or similar means, the rectangular heat sink component 1 is more prone to deformation (bending) in the long side direction compared to the short side direction. Therefore, it is presumed that by setting the "degree of bending" in the long side direction to be greater than the "degree of bending" in the short side direction, the heat sink component 1 as a whole can easily become flat even without applying excessive force (even with a small thread tightening force). Furthermore, it is believed that a relatively small thread tightening force can suppress the occurrence of cracks around the screws.

[0109] [Hole for threaded fastening]

[0110] Heat dissipation component 1 preferably has a hole for threaded fastening. Figure 1 , Figure 2 (Not explicitly shown in the text).

[0111] For example, when the heat dissipation component 1 is substantially rectangular, it is preferable to provide threaded fastening holes (through holes) at the periphery of the four corners of the heat dissipation component 1. In addition, depending on the length of the long side of the heat dissipation component 1, for example, a hole may be provided near the midpoint of the long side in the periphery of the heat dissipation component 1.

[0112] The diameter of the hole can be, for example, greater than 5mm and less than 9mm.

[0113] To be cautious, the means of joining the heat dissipation component 1 to other components is not limited to screws. For example, it can be joined by a special clamp or the like that allows it to be mounted to other components.

[0114] [Manufacturing Method / Materials]

[0115] The manufacturing method of the heat dissipation component in this embodiment is not particularly limited, and known methods can be used appropriately.

[0116] The heat dissipation component of this embodiment is preferably manufactured by a process including the following steps: (i) preparing a plate-shaped metal-silicon carbide composite containing aluminum or magnesium; (ii) machining (grinding, cutting, etc.) a portion of at least one side of the metal-silicon carbide composite to form the main surface 2A.

[0117] The heat dissipation component of this embodiment comprises a metal-silicon carbide composite containing aluminum or magnesium. A preferred method for manufacturing this metal-silicon carbide composite is high-pressure forging, which impregnates a porous body with metal under high pressure. More specifically, liquid forging or die casting can be used. High-pressure forging involves filling a porous silicon carbide body (preform) into a high-pressure vessel and impregnating it with molten metal containing aluminum or magnesium under high pressure to obtain the composite.

[0118] In the manufacture of the heat dissipation component of this embodiment, liquid forging is particularly preferred because it allows for stable mass production. The manufacturing method using liquid forging will be described below.

[0119] As an example, the heat dissipation component of this embodiment can be manufactured through the following process.

[0120] (Process 1) The process of forming a flat silicon carbide porous body (SiC preform).

[0121] (Step 2) The step of machining at least one side of the silicon carbide porous body into a convex curved shape.

[0122] (Step 3) The process of impregnating a porous silicon carbide body with a metal (alloy) containing aluminum or magnesium to create a metal-silicon carbide composite having a composite part containing silicon carbide and metal, and a surface metal layer on the outer surface of the composite part, and

[0123] (Step 4) Machining the surface metal layer on at least one side of the curved shape into a convex shape, thereby obtaining a heat dissipation component where f2 is more than 10 μm smaller than f1.

[0124] The above (process 1) to (process 4) will be explained in more detail.

[0125] There are no particular limitations on the manufacturing method of the silicon carbide porous body (SiC preform) in the above (step 1), and it can be manufactured using known methods. For example, silicon dioxide or alumina can be added as a binder to silicon carbide (SiC) powder as raw material, mixed, shaped, and calcined at a temperature above 800°C to manufacture it.

[0126] There are no particular restrictions on the molding method; pressure molding, extrusion molding, casting molding, etc. can be used, and adhesives for maintaining the shape can also be used as needed.

[0127] The key properties of metal-silicon carbide composites obtained by impregnating porous silicon carbide with metals containing aluminum or magnesium are thermal conductivity and coefficient of thermal expansion. Higher SiC content in the porous silicon carbide results in higher thermal conductivity and a lower coefficient of thermal expansion, making it preferable. However, if the SiC content becomes too high, it may sometimes fail to adequately impregnate the aluminum alloy.

[0128] In practical applications, a silicon carbide porous body containing at least 40% by mass of coarse SiC particles with an average particle size preferably of 40 μm or more, and a relative density of the SiC preform preferably in the range of 55% or more and 75% or less, is preferred. To prevent breakage during operation or impregnation, the strength of the silicon carbide porous body (SiC preform) is preferably 3 MPa or more in terms of flexural strength. The average particle size can be determined as follows: using a scanning electron microscope (e.g., JSM-T200 type manufactured by Nippon Egis Corporation) and an image analysis device (e.g., manufactured by AVIONICS Corporation of Japan), the particle size is determined for 1000 particles, and the average value is calculated. In addition, the relative density can be determined by Archimedes' method or the like.

[0129] Regarding the SiC powder used as a raw material for silicon carbide porous bodies (SiC preforms), it is preferable to adjust the particle size by appropriately combining coarse and fine powders. This makes it easy to simultaneously achieve both the strength of the silicon carbide porous body (SiC preform) and the high thermal conductivity of the resulting heat dissipation component.

[0130] Specifically, a mixed powder is preferably a mixture of (i) coarse SiC powder with an average particle size of 40 μm or more and 150 μm or less and (ii) fine SiC powder with an average particle size of 5 μm or more and 15 μm or less. Here, the preferred ratio of (i) to (ii) in the mixed powder is (i) 40% by mass or more and 80% by mass or less, and (ii) 20% by mass or more and 60% by mass or less.

[0131] Silicon carbide porous bodies (SiC preforms) can be obtained by debinding and calcining a molded body of a mixture obtained by adding a binder to SiC powder. If the calcination temperature is above 800°C, regardless of the calcination atmosphere, silicon carbide porous bodies (SiC preforms) with a flexural strength of 3 MPa or higher can be easily obtained.

[0132] However, in an oxidizing atmosphere, calcination of SiC is promoted at temperatures exceeding 1100°C, sometimes resulting in a decrease in the thermal conductivity of the metal-silicon carbide composite. Therefore, in an oxidizing atmosphere, calcination at temperatures below 1100°C is preferred.

[0133] The calcination time can be appropriately determined based on the size of the silicon carbide porous body (SiC preform), the amount added to the calcination furnace, the calcination atmosphere, and other conditions.

[0134] When silicon carbide porous bodies (SiC preforms) are formed into a specified shape, they can be dried piece by piece, or they can be dried using spacers such as carbon of the same shape as the preforms. This prevents warping caused by drying. Furthermore, regarding calcination, by performing the same treatment as drying, shape changes associated with changes in internal structure can be prevented.

[0135] In step 2 above, at least one side of the silicon carbide porous body (SiC preform) is machined into a convex curved shape facing outward by a cutting or grinding tool such as a lathe. This machining (cutting) is performed at the preform stage, which has the following advantages: no special cutting tools are needed after metal impregnation, and the degree of bending and flatness are easily controlled.

[0136] In the manufacturing of the heat dissipation component of this embodiment, processing can be performed on not only one side of the silicon carbide porous body (SiC preform) but also on both sides. That is, the silicon carbide porous body (SiC preform) can be processed in such a way that f2 is more than 10 μm smaller than f1 in the final heat dissipation component, and in a way that the values ​​of f1 and f2 respectively reach the desired values.

[0137] In the above (step 3), a metal-silicon carbide composite material with a composite part containing silicon carbide and metal and a surface metal layer on the outer surface of the composite part can be produced by impregnating a silicon carbide porous body (SiC preform) with a metal containing aluminum or magnesium through high pressure forging or the like.

[0138] As a method for impregnating a porous silicon carbide body (SiC preform) with a metal (alloy) containing aluminum or magnesium to obtain a metal-silicon carbide composite, for example, the following method is available.

[0139] A silicon carbide porous body (SiC preform) is placed inside a mold, and then configured such that one or more of the following—fibers, spherical particles, and fragmented particles, including alumina or silica—are in direct contact with the two surfaces of the mold, thereby forming a block.

[0140] The block is preheated to a temperature between 500°C and 650°C, and then one or more of the blocks are placed in a high-pressure vessel. Next, to prevent the temperature of the block from dropping, molten metal containing aluminum or magnesium is pressurized as quickly as possible at a pressure of 30 MPa or higher, so that the metal is impregnated into the pores of the silicon carbide porous body (SiC preform).

[0141] Thus, a metal-silicon carbide composite material can be obtained, which has a composite part comprising silicon carbide and metal, and a surface metal layer on the outer surface of the composite part.

[0142] In order to ensure that the metal in the metal-silicon carbide composite (typically an alloy of aluminum or magnesium) can fully penetrate into the voids of the preform during impregnation, it is preferable to have the lowest possible melting point.

[0143] From this perspective, aluminum alloys containing 7% to 25% silicon by mass are preferred, for example. Furthermore, the presence of 0.2% to 5% magnesium by mass further strengthens the bond between the silicon carbide particles and the metal portion, which is also preferable. Regarding the metal components in the aluminum alloy other than aluminum, silicon, and magnesium, there are no particular restrictions as long as the properties do not change drastically; for example, copper may be included.

[0144] As aluminum alloys, AC4C, AC4CH, ADC12, etc., which are used as casting alloys, can also be preferred.

[0145] To remove the strain generated during impregnation, annealing can be performed after the metal-silicon carbide composite is fabricated. Annealing for the purpose of removing strain is preferably performed at a temperature of 400°C or higher and 550°C or lower for 10 minutes or more and 5 hours or less.

[0146] If the annealing temperature is above 400°C, the strain inside the composite is fully released, which can suppress significant bending changes during the annealing process after machining. On the other hand, if the annealing temperature is below 550°C, it can prevent the aluminum alloy used in the impregnation from melting.

[0147] If the annealing time is 10 minutes or more, the strain inside the composite is fully released, which can suppress large changes in bending during the annealing process used to remove the machining strain. On the other hand, from the viewpoint of mass production, it is preferable to have an annealing time of 5 hours or less.

[0148] Furthermore, in (step 3), for example, the silicon carbide porous body (SiC preform) can be configured such that one or more of the following—fibers containing alumina or silica, spherical particles, and particles with broken shapes—are in direct contact with the surface: a surface metal layer of a specified thickness can be formed. It also has the advantages of almost no staining after impregnation and good processability when shaping.

[0149] The content of one or more materials in the surface metal layer, including fibers containing alumina or silicon dioxide, spherical particles, and fragmented particles, is preferably 0.1% by mass or more and 5% by mass or less, more preferably 0.3% by mass or more and 2% by mass or less, relative to the mass of the metal-silicon carbide composite.

[0150] If the content is 0.1% by mass or more, the thickness control of the aluminum layer becomes easier, and significant changes in the bending shape due to post-processing annealing can be suppressed. Conversely, if the content is 5% by mass or less, the aluminum alloy layer will not become too hard, making it easier to perform conventional machining.

[0151] Regarding the thickness of the surface metal layer on the surface of the metal-silicon carbide composite, in either case where only the heat dissipation side (corresponding to the main surface 2A side of the heat dissipation member 1) is machined in (step 4), or where both sides are machined, appropriate adjustments can be made after machining to ensure that the thickness difference between the surface metal layers on the inside and outside remains constant. The aforementioned inorganic fibers can be appropriately used in the adjustment.

[0152] Specifically, the average thickness of the metal layer on the mounting surface side of the circuit board (corresponding to the main surface 2B side of the heat sink 1) is preferably 10 μm or more and 300 μm or less. Furthermore, regarding the difference in average thickness between the two metal-containing layers, it is preferable that the average thickness of the thinner layer is within 50% of the average thickness of the thicker layer.

[0153] Regarding the difference in average thickness between the two metal-containing layers, it is preferable to adjust the average thickness of the metal-containing layer on the heat dissipation surface to be within 40%, and particularly preferably within 30%. This is because the difference in the coefficients of thermal expansion of the two metal-containing layers suppresses changes in the bending state.

[0154] In step (4), the surface metal layer of the metal-silicon carbide composite, which is at least machined into a convex curved shape, is machined and further annealed as needed, thereby obtaining a heat dissipation component whose f2 is more than 10 μm smaller than f1. Specifically, a suitable curved shape is formed on the heat dissipation surface of the metal-silicon carbide composite using a tool capable of precision machining (grinding, cutting, etc.) such as a lathe, and then, for example, it is heated to about 400°C to 550°C in a muffle furnace and annealed for about 2 to 6 hours.

[0155] In this embodiment, by precisely machining the surface layer of the metal-silicon carbide composite, a heat dissipation component with desired values ​​for f1, f2, etc., can be obtained. In particular, by machining not only the convex, curved sides of the metal-silicon carbide composite but also, as needed, the opposite sides, with appropriate machining (grinding, cutting, etc.), a heat dissipation component with desired f1, f2, h1 / a, h2 / b, etc., can be obtained. That is, a heat dissipation component with better heat dissipation characteristics and reliability can be obtained.

[0156] In addition, through the heat treatments such as "annealing" performed in (steps 1) to (steps 4) described above, the metal-silicon carbide composite or heat dissipation component is usually deformed as a whole. That is, even if only the heat dissipation surface of the metal-silicon carbide composite is formed into a curved shape by grinding, cutting, etc., heat treatment such as annealing usually forms a certain degree of curvature on the opposite side of the heat dissipation surface as well (i.e., f2 is usually a non-zero value). It should be noted that the degree of deformation caused by heat treatment such as annealing can be quantitatively estimated to a certain extent by conducting several preliminary tests.

[0157] In summary, in this embodiment, a suitable combination of machining and annealing conditions can yield the desired heat dissipation component, which is preferred.

[0158] The sum of the average thicknesses of the metal-containing layers on both sides after the above machining is preferably 500 μm or less, and particularly preferably 300 μm or less.

[0159] If the sum of the average thicknesses of the metal-containing layers on both sides is less than 500 μm, the thermal expansion of the heat dissipation component as a whole can be suppressed to a small extent. When a heat load is applied, the occurrence of cracks caused by the difference in thermal expansion rates between the heat dissipation component and the electrical components connected to it can be suppressed.

[0160] To be cautious, of course, the manufacturing method of the heat dissipation component in this embodiment is not limited to this.

[0161] For example, in the heat dissipation component of this embodiment, the surface metal layer has an arbitrary configuration, so in the above (step 3), the surface metal layer may or may not be necessary to form.

[0162] As another example, step 2 in steps 1 through 4 above can also be omitted. That is, the silicon carbide porous body (SiC preform) can be left unprocessed into a curved shape before metal impregnation, and the necessary machining can be performed in step 4 to obtain a heat dissipation component with the desired curvature or flatness. However, since machining is easier to perform before metal impregnation than after metal impregnation, it is preferable to pre-form a certain degree of curvature in step 2 from the viewpoint of reducing manufacturing costs.

[0163] Furthermore, as another example, it is also considered to omit step (4) from steps (1) to (4) above. That is, if the bending process is performed with sufficient precision in step (2) and the mold used for impregnation in step (3) is pre-set with the desired bending, then machining in step (4) may not be necessary. However, considering the dimensional changes caused by heating and cooling due to the impregnated metal, even if the bending process is performed with sufficient precision in step (2), it is preferable to adjust the bending shape and flatness by performing step (4).

[0164] The manufacturing method of the heat dissipation component in this embodiment may include other processes not described above.

[0165] For example, the process may include providing a hole for threaded fastening. For example, between steps (3) and (4) described above, a hole for threaded fastening, used for engagement with other components, may be provided in the metal-silicon carbide composite through machining or the like. By providing a hole for threaded fastening between steps (3) and (4), it is advantageous that this hole can be used in step (4) when fixing the metal-silicon carbide composite to a cutting or grinding tool.

[0166] Furthermore, after step (4), a grinding or sandblasting process can be performed on the surface of the heat dissipation component. This allows, for example, the average length RS of the aforementioned roughness curve element to be appropriately adjusted. m Regarding the specific methods for grinding or sandblasting, known techniques may be appropriately applied.

[0167] Furthermore, after step (4), a plating process can be performed to form a plating layer. For example, a plating layer can be formed on the surface of the heat dissipation component using known electroless Ni-P plating or Ni-B plating methods. The preferred thickness of the plating layer is as described above.

[0168] The embodiments of the present invention have been described above, but these are merely examples of the present invention, and various configurations other than those described above can be used. Furthermore, the present invention is not limited to the above embodiments, and modifications and improvements made within the scope of achieving the objectives of the present invention are all included in the present invention.

[0169] Example

[0170] The embodiments of the present invention will be described in detail based on examples and comparative examples. The present invention is not limited to these examples.

[0171] As described above, "Main Surface 2A" is the surface that joins with heat sinks, etc., and it is convexly curved outwards rather than inwards towards the heat dissipation component. "Main Surface 2B" is the surface of a component connected to electrical elements, etc.

[0172] <Manufacturing of heat dissipation components>

[0173] [Example 1]

[0174] (Formation of silicon carbide porous bodies)

[0175] First, mix silicon carbide powder A, silicon carbide powder B, and silica sol in a mixer for 30 minutes to obtain a mixture.

[0176] • Silicon carbide powder A (manufactured by Pacific Landam Co., Ltd.: NG-150, average particle size: 100μm) 300g

[0177] • Silicon carbide powder B (manufactured by Yakushima Electric Co., Ltd.: GC-1000F, average particle size: 10μm) 150g

[0178] • Silica sol (manufactured by Nissan Chemical Industries, Ltd.: Snowtex) 30g

[0179] The resulting mixture was placed into a mold and pressurized at a pressure of 10 MPa. This yielded a flat molded body with dimensions of 185 mm × 135 mm × 5.5 mm. The molded body was then calcined in air at 900 °C for 2 hours to obtain a porous silicon carbide body with a relative density (bulk density) of 65% by volume.

[0180] The main surface 2B of the silicon carbide porous body, which becomes the completed heat dissipation component, is machined using a surface grinding wheel. Then, the surface that will become the main surface 2A is machined into a convex curved shape using a lathe with a radius of 11m. At this time, adjustments are made to make the central thickness of the silicon carbide porous body 4.8mm.

[0181] For the following process, 30 identical silicon carbide porous bodies are fabricated.

[0182] (Metal impregnation)

[0183] Alumina fibers (manufactured by Tanaka Paper Industry Co., Ltd., 97% purity, sheet form) are arranged on the main surface 2A of a machined silicon carbide porous body, and are sandwiched between stainless steel plates with dimensions of 210mm×160mm×0.8mm on both sides coated with carbon, and 30 sheets are stacked.

[0184] Then, 6mm thick iron plates are placed on both sides and connected with 6 M10 bolts. The plates are then installed with a torque wrench with an installation torque of 3Nm in the face direction to form a block shape.

[0185] Then, the monolithic block is preheated to 620°C in an electric furnace, and then placed inside a preheated inner diameter container. The mixture is placed in a pressure mold. Molten aluminum alloy containing 12% by mass silicon and 1.0% by mass magnesium is injected into the pressure mold, and then pressurized at 100 MPa for 20 minutes. This impregnates the silicon carbide porous body with the aluminum alloy.

[0186] After impregnation, the stainless steel sheet is cooled to 25°C and then cut along its shape using a wet band saw to remove the clamped sheet. It is then annealed at 500°C for 3 hours to remove the strain from the impregnation process.

[0187] Thus, an aluminum-silicon carbide composite was obtained.

[0188] (Post-impregnation treatment)

[0189] The outer periphery of the resulting aluminum-silicon carbide composite was machined using an NC lathe to achieve a length and width of 190mm × 140mm. Then, through holes with a diameter of 7mm were machined at eight locations along the edge, and additional holes were machined at four locations. The countersunk hole.

[0190] In addition, the side of the aluminum-silicon carbide composite corresponding to the main surface 2A was machined using a turning center to achieve a radius (R) of 11m. After machining, it was annealed in a muffle furnace at 500°C for 4 hours to remove machining strain.

[0191] Furthermore, the aluminum-silicon carbide composite was cleaned by blasting with alumina abrasive particles under a pressure of 0.4 MPa and a conveying speed of 1.0 m / min. Then, electroless Ni-P and Ni-B plating were performed. As a result, a plating layer with a thickness of 8 μm (Ni-P: 6 μm, Ni-B: 2 μm) was formed on the surface of the composite.

[0192] The heat dissipation component is obtained through the above operations.

[0193] (Determination of f1 and f2)

[0194] For the obtained heat dissipation component, the flatness of the main surfaces 2A and 2B was measured using a VR-3000 device manufactured by KEYENCE, yielding f1 and f2 respectively. Since the main surfaces 2A and 2B could not be contained within the observation field of the device, the center (geometric centroid) of the main surfaces 2A / 2B when viewed from above was aligned with the center of the observation field of the measuring device, and the measurement was performed over a range of 190mm × 100mm.

[0195] (Determination of h1 and h2)

[0196] Using a laser three-dimensional shape measuring instrument, data related to the shapes of main surfaces 2A and 2B are obtained. By analyzing this data, h1 and h2 are calculated.

[0197] Apparatus: Laser 3D Shape Measuring Instrument (the following four components are integrated into one instrument)

[0198] XYθ platform unit: K2-300 (manufactured by Kazuki Seiki Co., Ltd.)

[0199] High-precision laser displacement gauge: LK-G500 (manufactured by KEYENCE Co., Ltd.)

[0200] Motor controller: SC-200K (manufactured by Shinzu Seiki Co., Ltd.)

[0201] AD converter: DL-100 (manufactured by Kazuki Seiki Co., Ltd.)

[0202] (Determination of RSm)

[0203] The average length RSm of the roughness curve elements of main surfaces 2A and 2B was measured using a Mitutoyo SJ-310 apparatus based on ISO 4287-1997.

[0204] [Examples 2-12]

[0205] In Examples 2 to 12, the heat dissipation components were manufactured using the same procedures as in Example 1, except for changing the lengths of the long and short sides of the heat dissipation component, the processing R described above (formation of the silicon carbide porous body), and the processing R described above (post-impregnation treatment). Furthermore, various values ​​were measured using the same procedures as in Example 1.

[0206] [Comparative Examples 1-3]

[0207] In the above-described (formation of silicon carbide porous body), except that the surface of the silicon carbide porous body, which becomes the main surface 2B of the completed heat dissipation component, is not machined using a surface grinding wheel, and the machining R in the above-described (formation of silicon carbide porous body) and the machining R in the above-described (treatment after impregnation) are changed, the heat dissipation component is manufactured through the same process as in Example 1. Then, various values ​​are measured by operating in the same manner as in Example 1.

[0208] In addition, since the surface of the main surface 2B of the completed heat dissipation component was not machined using a surface grinding wheel, the surface underwent natural bending to some extent through calcination and subsequent cooling.

[0209] The various numerical values ​​are summarized in Table 1 and Table 2.

[0210] In the table, the "E" in the h1 / a and h2 / b columns indicates the exponent; for example, 2.38E-03 means 2.38 × 10⁻⁶. -3 .

[0211]

Table 1

[0212]

[0213]

Table 2

[0214]

[0215] <Evaluation of manufacturing stability, etc., of power modules>

[0216] Ten heat dissipation components for each embodiment or comparative example are prepared and connected to simulated power components, thereby manufacturing a substrate for a simulated power module.

[0217] As a specific manufacturing sequence, using an apparatus commonly used in the manufacture of power modules, six specific locations on the main surface 2B of the heat dissipation component in each embodiment or comparative example are soldered with a ceramic substrate (a substrate with metal layers such as copper and aluminum on both sides of the ceramic plate). This yields a substrate for simulating a power module.

[0218] Then, in order to manufacture the analog power module, the analog power module is assembled into a housing using a substrate, sealed with resin, and covered to obtain the analog power module.

[0219] The obtained simulated power modules are checked for any adverse conditions that could lead to problems in mass production.

[0220] In all the analog power modules manufactured using the heat dissipation components of the various embodiments, there are no adverse conditions that would cause problems in mass production.

[0221] On the other hand, among the analog power module substrates manufactured using the heat dissipation components of Comparative Examples 1 to 3, one out of ten has the following defects that could cause problems in mass production: the jig for aligning the ceramic substrate cannot fit accurately, and the housing does not fit properly after being connected to the ceramic substrate.

[0222] The results above demonstrate that by using a plate-shaped heat dissipation component with features such as flatness satisfying specific relationships on two main surfaces, manufacturing stability (yield, etc.) can be improved when manufacturing power modules.

[0223] As a supplementary evaluation, the heat dissipation components of Examples 1-12 were joined to the heat sink using screws, and the fit between the heat dissipation components and the heat sink, heat dissipation performance, etc., were evaluated. The results showed that the fit and heat dissipation performance were good.

[0224] That is, it is shown that by using the heat dissipation components of Examples 1 to 12, a power module with good heat dissipation can be manufactured, and the yield rate when manufacturing such a power module can also be improved.

[0225] This application claims priority based on Japanese Patent Application No. 2019-013762, filed on January 30, 2019, the entire contents of which are incorporated herein by reference.

Claims

1. A heat dissipation component, which is a plate-shaped heat dissipation component comprising a metal-silicon carbide composite containing aluminum or magnesium. At least one of the two main surfaces of the heat dissipation component is convexly curved outwards from the heat dissipation component. Let the flatness of one principal surface as defined by JIS B 0621 be f1, and let the flatness of another principal surface, which is different from the one principal surface, as defined by JIS B 0621 be f2, where f2 is smaller than f1 and f2 is less than 300 μm. One main surface is the surface that engages with the heat sink, and the other main surface is the surface that connects to the power components. in, Flatness, as defined by JIS B 0621, is the minimum distance between two geometrically parallel planes that hold a planar object together.

2. The heat dissipation component as described in claim 1, wherein, One of the main surfaces and / or the other main surface has a surface metal layer containing aluminum or magnesium.

3. The heat dissipation component as described in claim 1 or 2, wherein, The f1 is greater than 100 μm and less than 700 μm.

4. The heat dissipation component as described in claim 1 or 2, wherein, The heat dissipation component is essentially rectangular. Let the length of the longer side of the rectangle be a, and the length of the shorter side be b. Let l1 be the straight line connecting the midpoints of the two shorter sides of the main face, and l2 be the straight line connecting the midpoints of the two longer sides of the main face. Let h1 be the maximum distance between a point on the curve formed by the main surface of the heat dissipation component when viewed in section containing l1 and approximately perpendicular to the main surface, and l1. Let h2 be the maximum distance between a point on the curve formed by the main surface of the heat dissipation component when viewed in section containing l2 and approximately perpendicular to the main surface, and l2. In this case, h1 / a ≥ h2 / b.

5. The heat dissipation component as described in claim 4, wherein, The value of (h1 / a) / (h2 / b) is greater than 1.00 and less than 1.

9.

6. The heat dissipation component as described in claim 1 or 2, wherein, The average length RS of the roughness curve feature of a principal surface m It is between 50μm and 250μm.

7. The heat dissipation component as described in claim 1 or 2, wherein, The average length RS of the roughness curve element of the other main surface m It is between 50μm and 200μm.

8. A method for manufacturing a heat dissipation component according to any one of claims 1 to 7, comprising the following steps: The process of preparing plate-shaped metal-silicon carbide composites containing aluminum or magnesium, and The process of machining at least a portion of one side of the composite to form the main surface.

Images