WO2023140063A1 - Composite copper member and power module including composite copper member - Google Patents

Abstract

The purpose of the present invention is to provide a novel composite copper member and a power module which includes the composite copper member. The composite copper member includes a copper member and a nickel layer which is formed on at least a partial surface of the copper member. The surface including the nickel layer has convex parts, wherein in a cross-sectional image of the composite copper member photographed using a scanning electron microscope, convex parts having a length of 50 nm to 1500 nm as measured in a direction parallel to the surface including the nickel layer are present in the quantity of 5 or more per 3.8 µm, and the nickel layer has a predetermined amount or a predetermined thickness of nickel. The power module includes the composite copper member and an insulating material which is bonded to the composite copper member at the surface including the nickel layer.

Description

Composite copper member and power module including the composite copper member

The present invention relates to a composite copper member and a power module including the composite copper member.

As a substrate for mounting power semiconductor elements such as IGBTs (insulated gate bipolar transistors) and SITs (static induction transistors) that generate a large amount of heat during operation, a module substrate with excellent heat dissipation characteristics, which is equipped with an insulating substrate and a heat sink, is used, and is called a power module. When a power module is used in a high-temperature, moisture-absorbing environment where a high voltage of several hundred to several thousand volts is applied, deterioration phenomena such as leaks and short circuits may occur and break down. This deterioration phenomenon is caused by migration in which the copper forming the heat sink is ionized, eluted, and precipitated as a copper compound at the place where it moves. In order to prevent such a deterioration phenomenon, provision of a migration prevention layer (Japanese Patent Application Laid-Open No. 2010-287844) is disclosed.

The present invention provides a novel composite copper member and a power module including the composite copper member.

One embodiment of the present invention is a composite copper member comprising a copper member and a nickel layer on at least a part of the surface of the copper member, wherein the surface on the side containing the nickel layer has protrusions, and in a cross-sectional image of the composite copper member taken by a scanning electron microscope, the number of protrusions having a length of 50 nm or more and 1500 nm or less when measured in a direction parallel to the surface on the side containing the nickel layer is 5 or more per 3.8 μm, and the nickel content of the nickel layer is 2 to ³⁵ on average. It is a composite copper member, which is mg/dm2. The amount of nickel may be a value measured by high frequency inductively coupled plasma emission spectrometry.

Another embodiment of the present invention is a composite copper member comprising a copper member and a nickel layer on at least a part of the surface of the copper member, wherein the surface on the side containing the nickel layer has protrusions, and in a cross-sectional image of the composite copper member taken with a scanning electron microscope, the number of protrusions having a length of 50 nm or more and 1500 nm or less when measured in a direction parallel to the surface on the side containing the nickel layer is 5 or more per 3.8 μm, and the nickel layer of the composite copper member is Copper is detected at a depth of 0 to 350 nm on the surface on the containing side, and

(A) at a depth of 10 nm from the surface on the side containing the nickel layer, the mass content of Ni is greater than the mass content of Cu;

(B) at a depth of 10 nm from the surface, the mass ratio of Ni to the total mass of the three components Cu, O, and Ni is 40 to 100%; and

(C) the ratio of the mass of Cu to the total mass of the three components of Cu, O, and Ni at a depth of 10 nm from the surface is 0 to 50%;

A composite copper member that satisfies any one of the group consisting of The detection of copper and the analysis of the mass of Cu, O and Ni may be performed by X-ray photoelectron spectroscopy.

Any one of the above composite copper members may further have a silicon layer on the surface containing the nickel layer. The amount of silicon in the silicon layer may be 35 μg/cm ² or more. A SixOy peak and an N compound peak may be detected from the surface of the silicon layer by TOF-SIMS surface analysis.

A further embodiment of the present invention is a power module including any one of the above composite copper members, and an insulating material adhered to the composite copper member on the surface including the nickel layer. A filler filling amount of the insulating material may be 50 wt % or more. The insulating material may be sheet resin or mold resin.

A further embodiment of the present invention is any one of the above methods for manufacturing a composite copper member, comprising the steps of: forming a copper oxide layer on the surface of the copper member by oxidizing the surface of the copper member; and forming a nickel layer on the surface of the copper member by plating the surface on which the copper oxide layer is formed with nickel.

A further embodiment of the present invention is any one of the above power module manufacturing methods, comprising the step of adhering one of the above composite copper members to the insulating material on the surface on the side containing the nickel layer.

== Cross-reference with related literature ==
This application claims priority based on Japanese Patent Application No. 2022-008217 filed on January 21, 2022, and the basic application is incorporated herein by reference.

BRIEF DESCRIPTION OF THE DRAWINGS It is a schematic diagram which shows the cross section of the power module which is the 1st Embodiment of this invention. It is a schematic diagram which shows the cross section of the power module which is the 2nd Embodiment of this invention. It is a schematic diagram which shows the cross section of the power module which is the 3rd Embodiment of this invention. In the examples of the present invention, by X-ray photoelectron spectroscopy (XPS), representative results of elemental analysis in the depth direction from the treated surface of the composite copper member (Examples 1, 2, 4, Comparative Examples 1 to 4, 8) are shown. FIG. 10 is a diagram showing representative analysis results of each composite copper member by time-of-flight secondary ion mass spectrometry (TOF-SIMS) in Example 14 of the present invention. FIG. 10 is a diagram showing representative analysis results of each composite copper member by time-of-flight secondary ion mass spectrometry (TOF-SIMS) in Example 15 of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings, but the present invention is not necessarily limited to these. The objects, features, advantages, and ideas of the present invention are clear to those skilled in the art from the description of the present specification, and those skilled in the art can easily reproduce the present invention from the description of the present specification.

The embodiments and specific examples of the invention described below show preferred embodiments of the invention, are shown for illustration or explanation, and are not intended to limit the invention to them. Based on the description herein, it will be apparent to those skilled in the art that various alterations and modifications can be made within the spirit and scope of the invention disclosed herein.

== Composite copper material ==
The composite copper member disclosed in the present specification is a composite copper member comprising a copper member and a nickel layer on at least a part of the surface of the copper member, wherein the surface of the side containing the nickel layer has convex portions, and in a cross-sectional image of the composite copper member taken by a scanning electron microscope, the number of convex portions having a length of 50 nm or more and 1500 nm or less when measured in a direction parallel to the surface of the side containing the nickel layer is 5 or more per 3.8 μm, and the nickel layer has a predetermined amount or thickness of nickel and / /. Or have a nickel alloy. This composite copper member can be suitably used for a copper substrate of a power module, and by having the structure described above, it can be firmly adhered to the insulating material of the power module. A specific example of the power module will be described later.

(1) Copper member The copper member used in the composite copper member disclosed herein is a material containing Cu as a main component, and includes, but is not limited to, copper foils such as electrolytic copper foil, rolled copper foil, and copper foil with a carrier, copper wires, copper plates, copper lead frames, copper powder, copper heat sinks, copper pillars, and the like. A material that can be electrolytically plated is preferable.

The copper member is preferably a copper member made of pure copper with a Cu purity of 95% by mass or more, 99% by mass or more, or 99.9% by mass or more, more preferably made of tough pitch copper, deoxidized copper, or oxygen-free copper, and more preferably made of oxygen-free copper having an oxygen content of 0.001% to 0.0005% by mass.

　The copper member is preferably a copper foil or a copper plate. In the case of copper foil, its thickness is 0.1 μm or more and 100 μm or less. In particular, the thickness is preferably 0.5 μm or more and 50 μm or less. In the case of a copper plate, it refers to a plate-shaped plate having a thickness of more than 100 μm. The thickness of the copper plate is not particularly limited, but is preferably 0.5 mm or more, 1 mm or more, 2 mm or more, or 10 mm or more, and preferably 10 cm or less, 5 cm or less, or 2.5 cm or less.

(2) Copper Oxide Layer The copper member may have a layer containing copper oxide or a layer made of copper oxide (generally referred to herein as a copper oxide layer) on the outermost surface. As a result, it becomes possible to more firmly adhere to the insulating material of the power module. In addition, the presence of the copper oxide layer suppresses the elution of copper ions compared to copper, making it difficult for migration to occur when a voltage is applied. Therefore, the dielectric breakdown time becomes longer. Here, the copper oxide includes copper oxide (CuO) and/or cuprous oxide (Cu ₂ O) or consists of copper oxide (CuO) and/or cuprous oxide (Cu ₂ O). The copper oxide content in the copper oxide layer is preferably 90% by weight or more, 95% by weight or more, 98% by weight or more, 99% by weight or more, or 99.9% by weight or more. In this specification, a copper member containing copper oxide is also referred to as a "copper member".

The copper oxide layer may contain metals other than copper. The contained metal is not particularly limited, but may contain at least one metal selected from the group consisting of Sn, Ag, Zn, Al, Ti, Bi, Cr, Fe, Co, Ni, Pd, Au and Pt. In particular, in order to have acid resistance and heat resistance, it is preferable to contain metals having higher acid resistance and heat resistance than copper, such as Ni, Pd, Au and Pt.

The average thickness of the copper oxide layer is preferably 500 nm or less, more preferably 447 nm or less, and even more preferably 82 nm or less. Furthermore, the average thickness of the copper oxide layer is preferably 20 nm or more, more preferably 40 nm or more, and even more preferably 55 nm or more. Although the ratio of the region where the copper oxide layer has a thickness of 500 nm or less is not particularly limited, it is preferably 50% or more and 500 nm or less, 70% or more is more preferably 500 nm or less, 90% or more is more preferably 500 nm or less, 95% or more is more preferably 500 nm or less, and 99% or more is 500 nm or less. The thickness of the copper oxide layer can be calculated by, for example, Sequential Electrochemical Reduction Analysis (SERA) at 10 measurement points in an area of 10×10 cm.

(3) Nickel layer The composite copper member disclosed herein has a layer containing nickel or a layer made of nickel (generally referred to herein as a nickel layer) on at least a portion of the surface of the copper member. Nickel may be included as an alloy.

The arithmetic mean roughness (Ra) of the surface of the composite copper member including the nickel layer is not particularly limited, but is preferably 0.01 μm or more, more preferably 0.03 μm or more, and 5.00 μm or less, more preferably 3 μm or less, and further preferably 0.83 μm or less. The arithmetic average roughness (Ra) is the average of the absolute values of Z (x) (that is, the height of the peak and the depth of the valley) in the contour curve (y = Z (x)) represented by the following formula at the reference length l.

The maximum height roughness (Rz) of the surface including the nickel layer of the composite copper member is not particularly limited, but is preferably 0.1 μm or more, more preferably 0.32 μm or more, and preferably 20.00 μm or less, more preferably 10.00 μm or less, and even more preferably 1.82 μm or less. The maximum height roughness (Rz) represents the sum of the maximum value of the peak height Zp and the maximum value of the valley depth Zv of the contour curve (y=Z(x)) at the reference length l.

Ra and Rz can be calculated by the method specified in JIS B 0601:2001 (in accordance with international standards ISO4287-1997).

In addition, in a cross-sectional image of the composite copper member taken by a scanning electron microscope, the number of protrusions having a length of 50 nm or more and 1500 nm or less when measured in a direction parallel to the surface on the side containing the nickel layer is preferably 5 or more, more preferably 7 or more, per 3.8 μm.

The elution amount of Cu ions on the nickel layer-containing side of the composite copper member is preferably 10.0 ppm or less, more preferably 5.0 ppm or less, and even more preferably 2.5 ppm or less when measured by the elution test described below. The smaller the amount of eluted Cu ions, the longer the dielectric breakdown time due to Cu migration.

The elution amount of Ni ions on the side including the nickel layer of the composite copper member is not particularly limited, but when measured by the elution test described below, it is preferably 100.0 ppm or less, more preferably 50.0 ppm or less, and even more preferably 32.0 ppm or less.

The elution test is performed by measuring the mass of Ni and Cu ions in the eluate using an ICP emission spectrometer as the amount of Cu ions and Ni ions eluted. At that time, 10 composite copper parts cut to a size of 40 mm × 18 mm were used as test pieces, and only the side containing the nickel layer was brought into contact with 20 mL of pure water, and the eluate was obtained by processing under the conditions of 121 ° C., 85% humidity, 2 atmospheres, and 60 hours.

The total content of nickel and nickel alloy in the nickel layer is preferably 90% by weight or more, 95% by weight or more, 98% by weight or more, 99% by weight or more, or 99.9% by weight or more.

The average thickness of the nickel layer is preferably 20 nm or more, more preferably 30 nm or more, and preferably 360 nm or less, more preferably 351 nm or less. The average nickel adhesion amount (mg/dm ² ) is preferably 1.5 mg/dm ² or more, more preferably 2.0 mg/dm ² or more, still more preferably 2.7 mg/dm ² or more, and preferably 35 mg/dm ² or less, more preferably 32 mg/dm ² or less. Although it is not limited to the following principle, it is thought that when the average thickness of the nickel layer is small and/or the average amount of nickel attached is small, the function as a copper protective layer becomes insufficient, and copper migration occurs due to copper elution or the like, shortening the dielectric breakdown time, or making the function as a heat-resistant or corrosion-resistant layer insufficient, and the oxidation and corrosion of copper deteriorates adhesion, solder heat resistance, and water resistance. In addition, it is believed that the thicker average thickness of the nickel layer and/or the larger average amount of nickel deposited reduces the number of protrusions and makes the surface flat, resulting in poor adhesion to the insulating material and poor solder heat resistance.

The average thickness and amount of adhesion of the nickel layer can be calculated by dissolving the nickel contained in the nickel layer in an acidic solution, measuring the mass of nickel by high-frequency inductively coupled plasma emission spectrometry (ICP analysis), calculating the volume from the density, dividing the obtained volume and mass by the area of the composite copper member where nickel is dissolved, and calculating the average thickness and the amount of adhesion per unit area. In the case of a copper foil, the total mass of nickel may be measured by dissolving the copper foil itself having the nickel layer and detecting only the amount of nickel forming the nickel layer.

In addition, copper is preferably detected at a depth of 0 to 350 nm, more preferably at a depth of 0 to 343 nm, on the surface of the composite copper member including the nickel layer. In addition, it is preferable to satisfy any one of the requirements selected from the group consisting of (A) to (C): (A) at a depth of 10 nm from the surface containing the nickel layer, the mass content of Ni is greater than the mass content of Cu; and (C) the ratio of the mass of Cu to the total mass of the three components Cu, O, and Ni at a depth of 10 nm from the surface is 0 to 50% (more preferably 0 to 40%, more preferably 0 to 33%). If no copper is detected at a depth of 0-350 nm, the nickel layer is too thick, resulting in a low number of protrusions and poor adhesion to the insulating material in power modules containing this composite copper component. If none of (A) to (C) are satisfied, either the nickel layer is not formed, or the nickel layer having an appropriate thickness for Cu is not formed. As a result, in a power module including this composite copper member, the function as a copper protective layer becomes insufficient, copper dissolves and migration occurs, the dielectric breakdown time is shortened, the heat resistance and corrosion resistance are deteriorated, and the adhesive strength is reduced. In addition, the distance from the depth at which Cu begins to be detected to the depth at which the ratio of Cu is 50% when the total of the three components Cu, O, and Ni among the detected components is 100% is preferably 20 nm or more, more preferably 50 nm or more, further preferably 51 nm or more, preferably 530 nm or less, more preferably 528 nm or less, and even more preferably 500 nm or less. This value is considered to be correlated with the length of the convex portion. The mass content of each element can be obtained by elementally analyzing the treated surface of the composite copper member by X-ray Photoelectron Spectroscopy (XPS).

In this way, there is an appropriate range for the amount of nickel adhering to the nickel layer, and by keeping the amount of nickel within that range, in a power module including this composite copper member, a composite copper member with good adhesion to insulating materials, solder heat resistance, and good dielectric breakdown time can be obtained.

(4) Structure of silicon layer The surface of the composite copper member disclosed herein on the side containing the nickel layer may further have a layer containing silicon or a layer made of silicon (herein generally referred to as a silicon layer).

The content of silicon in the silicon layer (herein referred to as silicon content) is not particularly limited, but is preferably 20.0 μg/cm ² or more, more preferably 21.9 μg/cm ² or more, or even more preferably 39.9 μg/cm ² or more. By setting the amount of silicon to these values, the heat resistance of the power module including this composite copper member is significantly improved. The amount of silicon can be calculated by dissolving silicon contained in the silicon layer with an acidic solution, measuring the mass of silicon by high frequency inductively coupled plasma emission spectroscopy (ICP analysis), and calculating the volume from the density. In the case of copper foil, the total mass of silicon may be measured by dissolving the copper foil itself having the silicon layer and detecting only the amount of silicon forming the silicon layer.

In addition to these values for the amount of silicon, it is preferable that both the SixOy peak and the N compound peak are detected from the surface of the silicon layer by TOF-SIMS surface analysis. Thereby, the heat resistance of the power module including this composite copper member is further improved. Examples of N compounds include coupling agents and silicone resins containing amino groups, mercapto groups, isocyanurate groups, ureido groups, isocyanate groups, and the like.

The arithmetic mean roughness (Ra) of the surface of the composite copper member containing the silicon layer is not particularly limited, but is preferably 0.01 μm or more, more preferably 0.03 μm or more, and preferably 5.00 μm or less, more preferably 3 μm or less, and further preferably 0.83 μm or less.

The maximum height roughness (Rz) of the surface containing the silicon layer of the composite copper member is not particularly limited, but is preferably 0.1 μm or more, more preferably 0.32 μm or more, and preferably 20.00 μm or less, more preferably 10.00 μm or less, and even more preferably 1.82 μm or less.

In addition, in a cross-sectional image of the composite copper member taken by a scanning electron microscope, the number of protrusions having a length of 50 nm or more and 1500 nm or less when measured in a direction parallel to the surface on the side containing the silicon layer is preferably 5 or more per 3.8 μm, more preferably 7 or more.

The elution amount of Cu ions on the silicon layer-containing side of the composite copper member is preferably 10.0 ppm or less, more preferably 5.0 ppm or less, and even more preferably 2.5 ppm or less. The smaller the amount of eluted Cu ions, the longer the dielectric breakdown time due to Cu migration.

The amount of Ni ions eluted from the silicon layer-containing side of the composite copper member is not particularly limited, but is preferably 100.0 ppm or less, more preferably 50.0 ppm or less, and even more preferably 32.0 ppm or less.

For these measurements, the method described in "(3) Nickel layer" can be used.

== Power module structure ==
As described above, the composite copper member disclosed in this specification can be suitably used for power modules.

This power module includes, in addition to the composite copper member, an insulating material adhered to the composite copper member on the surface of the composite copper member that includes the nickel layer.

(1) Basic Structure The structure of the power module is not particularly limited as long as the composite copper member and the organic insulating layer are in contact with each other. As an example, FIG. 1 is a schematic diagram showing a cross section of a power module of the first embodiment as a basic structure. This power module includes a base substrate 1 , an organic insulating layer 2 formed on the base substrate 1 , and a copper substrate 3 formed on the organic insulating layer 2 . The composite copper component disclosed herein is formed from base substrate 1 and/or copper substrate 3 .

In the second embodiment of the power module, as shown in FIG. 2, in addition to the basic structure of FIG. 1, semiconductor elements 4 and external electrode terminals 8 are arranged on a copper substrate 3, and the semiconductor elements 4 are connected via metal wires 6. This power module is housed in a case 9 except that the external connection portions of the external electrode terminals 8 and the external heat radiation portion of the base substrate 1 are exposed to the outside.

In the third embodiment of the power module, as shown in FIG. 3, in addition to the basic structure of FIG. 1, the semiconductor element 4 is arranged on the copper substrate 3 and the lead frame 5 is directly connected to the Cu substrate 3, and the case where it is connected to the semiconductor element 4 via the metal wire 6 is shown. This power module is housed in a case 9 except that the external connection portion of the lead frame 5 and the external heat radiation portion of the base substrate 1 are exposed to the outside.

(2) Insulating Material In the power module, the copper substrate 3 is adhered to the organic insulating layer 2 and the molding resin 7 . Although the organic insulating layer 2 and the mold resin 7 are made of an insulating material, they may be made of the same insulating material or different insulating materials.

The type of resin used for the insulating material is not particularly limited, and may be a thermoplastic resin or a thermosetting resin. Examples of thermoplastic resins include fluorine resins, liquid crystal polymers, synthetic resins such as polyethylene and polypropylene, and thermoplastic polyimide resins. A thermally conductive filler may be dispersed in the insulating material, but it is preferable to disperse it evenly. Thermally conductive fillers include fused silica (SiO ₂ ), crystalline silica (SiO ₂ ), aluminum oxide (Al ₂ O ₃ ), boron nitride (BN), aluminum nitride (AlN), silicon carbide (SiC), and the like. The content of the thermally conductive filler is preferably 50 wt % or more, more preferably 55 wt % or more, relative to the insulating material. In order to improve thermal conductivity, it is preferable that the content of the thermally conductive filler is large, but if the content of the thermally conductive filler is too large, it will lead to a decrease in adhesion to the composite copper member, so the content of the thermally conductive filler should be an appropriate amount for the insulating material.

The organic insulating layer 2 is not particularly limited, and a known organic insulating sheet can be used. Although the thickness of the organic insulating layer 2 is not particularly limited, it is preferably 1 μm or more and 1000 μm or less.

== Power module manufacturing method ==
The method for manufacturing a power module of the present embodiment includes a copper member, a composite copper member including a nickel layer on at least a part of the surface of the copper member, and an insulating material adhered to the composite copper member on the surface including the nickel layer, the surface including the nickel layer having protrusions, and in a cross-sectional image of the composite copper member taken by a scanning electron microscope, the number of protrusions having a length of 50 nm or more and 1500 nm or less when measured in a direction parallel to the surface including the nickel layer is about 3.8 μm. A method for manufacturing five or more power modules, the method including a first step of forming a nickel layer on the surface of the copper member by plating the surface of the copper member.

(1) Manufacturing method of composite copper member (1-1) Formation of copper oxide layer

Before the first step, a second step of forming a copper oxide layer on the surface of the copper member by oxidizing the surface of the copper member may be included. This oxidation treatment roughens the surface of the copper member.

Before this oxidation step, a surface roughening treatment step such as soft etching or etching may be performed, but it is not necessary. In addition, before the oxidation treatment, degreasing treatment, acid cleaning for uniformizing the surface by removing a natural oxide film, or alkali treatment for preventing acid from being brought into the oxidation process after acid cleaning may be performed, but it is not necessary. The alkali treatment method is not particularly limited, but preferably 0.1 to 10 g/L, more preferably 1 to 2 g/L alkaline aqueous solution, such as sodium hydroxide aqueous solution, at 30 to 50° C. for about 0.5 to 2 minutes.

Although the oxidation treatment method is not particularly limited, it may be formed using an oxidizing agent, or may be formed by heat treatment or anodization.

The oxidizing agent is not particularly limited, and for example, an aqueous solution of sodium chlorite, sodium hypochlorite, potassium chlorate, potassium perchlorate, etc. can be used. Various additives (eg, phosphates such as trisodium phosphate dodecahydrate) and surface active molecules may be added to the oxidizing agent. Surface active molecules include porphyrins, porphyrin macrocycles, extended porphyrins, ring contracted porphyrins, linear porphyrin polymers, porphyrin sandwich coordination complexes, porphyrin sequences, silanes, tetraorgano-silanes, aminoethyl-aminopropyltrimethoxysilane, (3-aminopropyl)trimethoxysilane, (1-[3-(trimethoxysilyl)propyl]urea) ((l-[3-(Trimethoxysilyl)propyl]u rea)), (3-aminopropyl)triethoxysilane, ((3-glycidyloxypropyl)trimethoxysilane), (3-chloropropyl)trimethoxysilane, (3-glycidyloxypropyl)trimethoxysilane, dimethyldichlorosilane, 3-(trimethoxysilyl)propyl methacrylate, ethyltriacetoxysilane, triethoxy(isobutyl)silane, triethoxy(octyl)silane, tris(2-methoxyethoxy)(vinyl)silane, chlorotrimethylsilane, methyltrichlorosilane Examples include silane, silicon tetrachloride, tetraethoxysilane, phenyltrimethoxysilane, chlorotriethoxysilane, ethylene-trimethoxysilane, amines and sugars.

Various additives (for example, phosphates such as trisodium phosphate dodecahydrate) and surface active molecules may be added to the oxidizing agent to adjust the precipitation of copper oxide. Surface-active molecules include porphyrins, porphyrin macrocycles, extended porphyrins, ring-contracted porphyrins, linear porphyrin polymers, porphyrin sandwich coordination complexes, porphyrin sequences, silanes, tetraorgano-silanes, aminoethyl-aminopropyl-trimethoxysilane, 3-aminopropyl)trimethoxysilane, 1-[3-(trimethoxysilyl)propyl]urea, (3-aminopropyl)triethoxysilane, (3-glycidyloxypropyl)trimethoxysilane, (3-chlorochlorosilane). Examples include propyl)trimethoxysilane, (3-glycidyloxypropyl)trimethoxysilane, dimethyldichlorosilane, 3-(trimethoxysilyl)propyl methacrylate, ethyltriacetoxysilane, triethoxy(isobutyl)silane, triethoxy(octyl)silane, tris(2-methoxyethoxy)(vinyl)silane, chlorotrimethylsilane, methyltrichlorosilane, silicon tetrachloride, tetraethoxysilane, phenyltrimethoxysilane, chlorotriethoxysilane, ethylene-trimethoxysilane, amines, sugars, and the like. I can.

Although the oxidation reaction conditions are not particularly limited, the liquid temperature of the oxidizing agent is preferably 40 to 95°C, more preferably 45 to 80°C. The reaction time is preferably 0.5 to 30 minutes, more preferably 1 to 10 minutes.

For the copper oxide layer, a dissolving agent may be used to adjust the protrusions on the surface of the oxidized copper member, but it is not necessary. The solubilizing agent used in this dissolving step is not particularly limited, but is preferably a chelating agent, particularly a biodegradable chelating agent, such as ethylenediaminetetraacetic acid, diethanolglycine, tetrasodium L-glutamate diacetate, ethylenediamine-N,N'-disuccinic acid, 3-hydroxy-2,2'-sodium iminodisuccinate, trisodium methylglycine diacetate, tetrasodium aspartate diacetate, disodium N-(2-hydroxyethyl)iminodiacetate, and sodium gluconate. etc. can be exemplified. Although the pH of the dissolving agent solution is not particularly limited, it is preferably alkaline, more preferably pH 8 to 10.5, still more preferably pH 9.0 to 10.5, and even more preferably pH 9.8 to 10.2.

The surface of this copper oxide layer may be reduced with a reducing agent, but it is not necessary. When the reduction treatment is performed, cuprous oxide may be formed on the surface of the layer containing copper oxide. Examples of the reducing agent used in this reduction step include dimethylamine borane (DMAB), diborane, sodium borohydride, hydrazine and the like.

The resistivity of pure copper is 1.7×10 ⁻⁸ (Ωm), whereas that of copper oxide is 1 to 10 (Ωm), and that of cuprous oxide is 1×10 ⁶ to 1×10 ⁷ (Ωm). Therefore, the layer containing copper oxide has low conductivity, and even if the amount of the layer containing copper oxide transferred to the resin substrate is large, transmission loss due to the skin effect is unlikely to occur when forming a circuit of a printed wiring board or a semiconductor package substrate using the copper member according to the present invention. .

(1-2) Formation of Nickel Layer In the first step, the surface of the copper member is plated to form a nickel layer on the surface of the copper member.

The plating method is not particularly limited, and plating can be performed by electrolytic plating, electroless plating, vacuum deposition, chemical conversion treatment, or the like, but electrolytic plating is preferred because it is preferable to form a uniform plating layer.

In the case of electrolytic plating, nickel plating including nickel alloy plating is preferable. Examples of metals formed by nickel plating include pure nickel, Ni--Cu alloys, Ni--Cr alloys, Ni--Co alloys, Ni--Zn alloys, Ni--Mn alloys, Ni--Pb alloys, and Ni--P alloys.

Examples of metal salts used for nickel plating include nickel sulfate, nickel sulfamate, nickel chloride, nickel bromide, zinc oxide, zinc chloride, diamminedichloropalladium, iron sulfate, iron chloride, chromic anhydride, chromium chloride, sodium chromium sulfate, copper sulfate, copper pyrophosphate, cobalt sulfate, and manganese sulfate.

In nickel plating, the bath composition preferably contains, for example, nickel sulfate (100 g/L or more and 350 g/L or less), nickel sulfamate (100 g/L or more and 600 g/L or less), nickel chloride (0 g/L or more and 300 g/L or less), and a mixture thereof. Good.

When electrolytic plating is applied to an oxidized copper foil surface, the copper oxide on the surface is first reduced, and an electric charge is used to become cuprous oxide or pure copper, so there is a time lag before plating, and then the metal that forms the metal layer begins to precipitate. The amount of charge varies depending on the type of plating solution and the amount of copper oxide. For example, when Ni plating is applied to a copper member, it is preferable to apply a charge of 10 C to 90 C or less, more preferably 20 C to 65 C, per area dm ² of the copper member to be electrolytically plated in order to keep the thickness within a preferable range.

Also, the current density is not particularly limited, but is preferably 0.2 A/dm ² to 10 A/dm ² . Note that the current may be changed between the time until the oxide of the copper oxide layer is partially reduced and the time during which the plating is being applied.

(1-3) Formation of silicon layer The method for manufacturing a composite copper member may or may not include a third step of forming a silicon layer on the surface of the composite copper member after the first step. The method of forming the silicon layer is not particularly limited, but water glass treatment is preferred when treating with a liquid.

Water glass is an aqueous solution of alkali metal silicate, which is represented by M ₂ O.nSiO ₂ . Any of Na, Li, and K may be sufficient as an alkali metal. M ₂ O and SiO ₂ are mixed in various ratios in the water glass, and the water glass used in the third step is not particularly limited, but n is preferably 2-4.

The specific method of water glass treatment is not particularly limited, and water glass may be applied to the copper surface with a roller or bar coater, or may be sprayed, or the copper material may be immersed in water glass. The concentration of M ₂ O.nSiO ₂ is not particularly limited, but may be 0.1% to 20%, 0.5% to 10%, or 2% to 5%. Although the reaction conditions are not particularly limited, the treatment temperature is preferably 10°C to 95°C, more preferably 20°C to 85°C. The treatment time is preferably 1 second to 10 minutes. The water glass treatment may be performed multiple times.

After the copper material has been treated with water glass, it is dried. Drying after the treatment may be carried out by blowing off moisture with air or by heating. When heating, the temperature is preferably 50° C. to 250° C., and the heating time is preferably 10 seconds to 60 minutes.

A coupling agent may be dissolved in the water glass for treating the copper material (this solution is hereinafter referred to as the mixed agent). Although the concentration of the coupling agent is not particularly limited, it is preferably 0.5%, 1%, 2%, 3%, 4%, 5%, 6%, 7%, 8% or 9% or more, and preferably 20%, 15% or 10% or less by weight.

(1-4) Coupling agent treatment The method for manufacturing a composite copper member may include a fourth step of performing coupling agent treatment after the first step, before, after, or before and after the third step, but it does not have to be included.

The coupling agent is not particularly limited, but in the case of a silane coupling agent, one having 2 or 3 hydrolyzable groups is preferable, and the hydrolyzable group is preferably a methoxy group or an ethoxy group.

Although not particularly limited, 3-mercaptopropyltrimethoxysilane, 3-aminopropyltrimethoxysilane, 3-mercaptopropyltrimethoxysilane, 2-(3,4-epoxycyclohexyl)ethyltrimethoxysilane, vinyltrimethoxysilane, 3-methacryloxypropyltrimethoxysilane, 3-isocyanatopropyltriethoxysilane, 3-ureidopropyltrialkoxysilane, 3-acryloxypropyltrimethoxysilane, and the like can be used.

Coupling agent treatment can be carried out by applying or spraying a solution in which the coupling agent is dispersed in water or an organic solvent to adsorb it. A solution in which a coupling agent is dispersed in water or an organic solvent is not particularly limited.

After adsorption, drying completes the coupling agent treatment. The drying temperature and time are not particularly limited as long as the solvent water or organic solvent is completely evaporated, but it is preferable to dry at 70 degrees for 1 minute or more, more preferably at 100 degrees for 1 minute or more, and more preferably at 110 degrees for 1 minute or more.

This coupling treatment provides the composite copper member with the N compound detected by the TOF-SIMS surface analysis as described above. The presence of both SixOy and the N compound on the surface of the silicon layer further improves the heat resistance of the power module including the composite copper member.

(2) Power module manufacturing method A power module can be manufactured using the composite copper member manufactured in (1). The manufacturing method is not particularly limited, and a well-known manufacturing method may be used.

(1) Production of composite copper member As a copper foil for producing a composite copper member, FV-WS (thickness: 18 μm) (manufactured by Furukawa Electric Co., Ltd.) was used in Example 18, and DR-WS (thickness: 18 μm) (manufactured by Furukawa Electric Co., Ltd.) was used in Examples 1 to 17 and Comparative Examples 1 to 9. A commercially available copper plate (thickness: 500 μm) was used as the copper plate.

For these copper foils and copper plates, multiple test pieces were prepared under the conditions listed in Table 1 and tested. Examples 13 to 18 were subjected to water glass treatment, Comparative Examples 1, 2, 5 and 7 were not plated, Comparative Examples 7 to 9 were not oxidized, and Comparative Example was not subjected to water glass treatment.

(1-1) Etching treatment

For Example 11 and Comparative Example 7, the solution described in Table 1 was applied to the shiny side of the copper foil (glossy side; the side that is flat when compared with the opposite side) of the copper foil before the oxidation treatment and the copper plate, and the etching treatment was performed under the conditions described in Table 1. After etching, the copper foil and copper plate were washed with water and dried.

(1-2) Oxidation treatment

For Examples 1 to 17 and Comparative Examples 1 to 6, the shiny surface of the copper foil, and for Example 18, the matte surface of the copper foil and one surface of the copper plate were oxidized using the oxidizing agent (sodium chlorite; sodium hydroxide; KBM-403 (3-glycidoxypropyltrimethoxysilane; manufactured by Shin-Etsu Chemical Co., Ltd.) mixed solution) and conditions described in Table 1. After the copper foil and copper plate were oxidized, they were washed with water and then dried.

(1-3) Reduction Treatment For Comparative Example 2, after the oxidation treatment in (2), reduction treatment was performed with the reducing agent described in Table 1 under the reaction conditions.

(1-4) Electroplating Treatment After oxidation treatment, in Examples 1 to 18 and Comparative Examples 3, 4, 6, 8, and 9, Ni electrolytic plating solution (nickel sulfate 240 g/L; boric acid 30 g/L) was used to electroplate both surfaces of the copper foil and copper plate under the conditions shown in Table 1. After electroplating, the copper foil and copper plate were washed with water and dried.

(1-5) Water glass treatment For Examples 13 to 18, after the electroplating treatment in (4), water glass treatment was performed using a 41 g/L sodium silicate aqueous solution (sodium silicate 52-57% by weight, SiO ₂ /Na ₂ O ratio (2.06 to 2.31) under the reaction conditions listed in Table 1. After immersion in a 41 g/L sodium silicate solution at 50°C or 80°C for 1 minute and washing with water, the water glass treatment was performed as shown in Table 1. was dried under the drying conditions of

(1-6) Coupling treatment Examples 5 and 12 were subjected to the electrolytic plating treatment of (4), and examples 14 to 18 were subjected to the coupling treatment after the water glass treatment of (5) using KBE-903 (3-aminopropyltriethoxysilane; manufactured by Shin-Etsu Chemical Co., Ltd.) under the conditions shown in Table 1. Specifically, the copper foil and copper plate were immersed in a 3 vol % KBE-903 solution at 23° C. for 1 minute, washed with water, and dried at 130° C. for 1 minute.

(2) Test method for composite copper members (2-1) Thickness of copper oxide layer

The thickness of the copper oxide layer on the shiny side of the treated copper foil was measured using QC-100 (manufactured by ECI) and the thickness of copper oxide and copper sulfide by the continuous electrochemical reduction method (SERA) method using the following electrolytic solution. Specifically, using a gasket diameter of 0.32 cm, a reduction reaction was caused at a constant current (90 μA/cm ² ) using an electrolytic solution (pH = 8.4; boric acid 6.18 g/L; sodium tetraborate 9.55 g/L), and the following potentials were determined as the peaks of copper oxide (CuO, Cu ₂ O) and copper sulfide (CuS). The sum of the respective thicknesses was taken as the thickness of the copper oxide layer.
CuO: -0.85V to -0.55V
_Cu2O : -0.55V to -0.30V
CuS: -1.00V to -0.85V

The obtained reduction time and the current density were substituted into the following formula to convert to the film thickness.
CuO film thickness (nm)=0.00639×current density (μA/cm ² )×reduction time (sec)×0.1
Film thickness of Cu ₂ O (nm)=0.0124×current density (μA/cm ² )×reduction time (sec)×0.1
Cu ₂ S film thickness (nm)=0.0147×current density (μA/cm ² )×reduction time (sec)×0.1

(2-2) Average thickness and amount of Ni per unit area in the Ni layer, and amount of Si per unit area The composite copper member was treated with 12% nitric acid to dissolve the treated surface containing Ni and Si, and the mass of Ni and Si in the eluate was measured using an ICP emission spectrometer 5100 SVDV ICP-OES (manufactured by Agilent Technologies). Then, the volume of Ni is calculated using the density of Ni, and the average thickness of the Ni layer and the mass per unit area are calculated by dividing the obtained mass and volume by the area of the composite copper foil on which the Ni layer is formed. Similarly, for Si, the mass per unit area was calculated.

(2-3) XPS
Elemental analysis of the surface of the treated surface of the composite copper member was performed by X-ray Photoelectron Spectroscopy (XPS).

A Quantera SXM (ULVAC-PHI) was used as a device, and the component in the depth direction was measured under the conditions described below.
X-ray source: monochromatic Al Kα (1486.6 eV)
Ar sputtering conditions:
Accelerating voltage: 1 kV
Irradiation area: 3x3mm
Sputtering speed: 2.03 nm/min (in terms of SiO2)

Representative results of elemental analysis (Examples 1, 2, 4, Comparative Examples 1 to 4, 8) are shown in FIG. and,
[1] Depth at which Cu begins to be detected [2] From the depth at which Cu begins to be detected, the ratio of Cu is 50% when the sum of the three components Cu, O, and Ni among the detected components is 100%.

(2-4) Time-of-flight secondary ion mass spectrometry (TOF-SIMS)
A TRIFT V nano-TOF (manufactured by ULVAC-Phi, Inc.) was used as the device type for the metal member thus produced, and the measurement was performed under the conditions shown below. Among the results, representative results of Examples 14 and 15 are shown in FIG.

Device model:
Primary ion: 30 kV, Bi3++
Measurement mode: bunching Raster size: 100 μm × 100 μm
Measurement time: 30 frames/1 field of view Charge neutralization: None Number of pixels: 512 x 512 pixels
GCIB:
N compound: POSI spectrum SiOx: NEGA spectrum

(2-5) Elution Test A piece of copper foil was prepared by cutting the composite copper foil into a size of 40 mm×18 mm. The untreated surface was masked with a masking tape (masking tape for plating 851A: manufactured by 3M), and the elution test was performed only on the treated surface. Ten copper foil pieces were immersed in 20 mL of pure water and subjected to a HAST test (121° C., 85% humidity, 2 atm, 60 hours) to obtain a metal eluate. Using an ICP emission spectrometer 5100 SVDV ICP-OES (manufactured by Agilent Technologies), the masses of Ni and Cu in the eluate were measured.

(2-6) Number of protrusions The number of protrusions measured in a direction parallel to the surface on the side containing the nickel layer was obtained by observing the cross section of the composite copper foil with a focused ion beam (FIB).

(2-7) Surface roughness (Ra and Rz)
For the treated surface of the composite copper foil and composite copper plate, the surface shape of the copper foil was measured using a confocal scanning electron microscope OPTELICS H1200 (manufactured by Lasertec Co., Ltd.), and Ra and Rz were calculated by the method specified in JIS B 0601: 2001 (in accordance with international standards ISO4287-1997). As measurement conditions, the scan width was 100 μm, the scan type was area, the light source was Blue, and the cutoff value was 1/5. The object lens was set to ×100, the contact lens was set to ×14, the digital zoom was set to ×1, and the Z pitch was set to 10 nm.

(2-8) Adhesion to resin (peel strength to resin)
A sheet-like resin A with a filler content of 92 wt % or a sheet-like resin B with a filler content of 55 wt % was laminated on the treated surface of the composite copper foil and thermocompression bonded using a vacuum high pressure press. Peel strength (kgf/cm) was measured by a 90° peel test (Japanese Industrial Standards (JIS) C5016) on the test piece immediately after thermocompression bonding.

(2-9) Soldering heat resistance As with the peel strength test piece, a sheet-shaped resin A with a filler loading of 92 wt% or a sheet-shaped resin B with a filler loading of 55 wt% was laminated on the treated surface of the composite copper foil. A sample was immersed in a solder bath at a temperature of 260° C. or 300° C., and the time at which the copper foil piece began to swell was visually measured and used as an index of solder heat resistance.

(2-10) Dielectric breakdown time A copper plate (thickness 0.5 mm) treated under the conditions shown in Table 1 is used as the base copper, and on the treated surface of the base copper, a sheet-like resin A with a filler content of 92 wt% or a sheet-like resin B with a filler content of 55 wt% is laminated (thickness 0.1 mm). It was etched to become a pattern of With the circuit copper side as the positive pole and the base copper side as the negative pole, 1 kV was applied under constant temperature and humidity of 85° C. and 85% humidity, and the time until dielectric breakdown occurred was measured. In this test, the dielectric breakdown time was defined as the time when a current of 1 mA or more flowed. In addition, the average value of four points was used for the dielectric breakdown time.

(2-11) Adhesion with resin (die shear strength)
A polyimide tape having a thickness of 0.05 mm was attached to the copper plate treated under the conditions shown in Table 1 to form a gap of 0.05 mm, and a molding resin having a filler content of 65 wt % was applied by dispensing. After that, a 5 mm square silicon chip with a polyimide passivation film and a plasma treatment was placed on a mold resin with a filler content of 65 wt % and cured in an oven. After that, the die shear strength was measured with a universal bond tester (manufactured by Nordson Advanced Technologies). The area of the resin-adhered portion was measured, and the adhesive strength was calculated from the shear strength/resin area. Measurements were taken at 10 different points, and the average value was used.

(3) Results Table 2 shows the above analysis results.

In all of the composite copper foils of Examples, the number of protrusions having a length of 50 nm or more and 1500 nm or less when measured in a direction parallel to the surface of the side containing the nickel layer was 5 or more per 3.8 μm, and the average amount of nickel in the nickel layer was 2 to 35 mg / dm ² .

In addition, in all the composite copper foils of the examples, copper was detected at a depth of 0 to 350 nm on the surface of the composite copper member, and at a depth of 10 nm from the surface, the mass content of Ni was larger than that of Cu. The mass ratio of Cu was 0 to 50%.

Comparative Examples 1, 2, 5, and 7 were not plated, and the Ni adhesion amount was zero. Comparative Examples 3 and 6 also had a short plating time and a very small amount of Ni adhesion (average 1.3 mg/dm ² or less, which is less than half the minimum value of the examples). In addition, in these comparative examples, the Ni ratio at a depth of 10 nm from the surface was 35% or less and the Cu ratio was 50% or more, the Ni mass content was larger than the Cu mass content, and the amount of eluted Cu ions was significantly greater than in the examples. Therefore, the dielectric breakdown time is very short.

In Comparative Example 4, since the Ni adhesion amount was too large, the unevenness was smoothed by leveling, and sufficient convex portions were not formed (3 pieces). In addition, copper is not detected at a depth of 0 to 350 nm from the surface because the protrusions are too short compared to the thickness of Ni. Therefore, the adhesiveness, heat resistance, and adhesiveness after PCT are considerably inferior to those of the examples. It is considered that this is because the amount of Ni adhered by the plating process is too large for the length of the projections formed by the oxidation process.

In Comparative Example 8, like Examples 6 and 17, the plating treatment time was long, and the Ni adhesion amount was 30 mg/dm ² or more, which is larger than the other examples. However, unlike Examples 6 and 17, the number of protrusions per 3.8 μm is very small (three) because no oxidation treatment is performed. Also, since the Ni is too thick, copper is not detected at a depth of 0 to 350 nm from the surface. Therefore, the adhesiveness, heat resistance, and adhesiveness after PCT are considerably inferior to those of the examples.

In Comparative Example 9, although the amount of Ni is within an appropriate range, the number of protrusions is very small (four) because oxidation treatment is not performed. Therefore, the adhesiveness, solder heat resistance, and adhesiveness after PCT are considerably inferior to those of the examples.

On the other hand, in the example, the adhesiveness and solder heat resistance were good even at a high temperature of 300°C, the amount of Cu ion elution was small, and the dielectric breakdown time was long.

Although it is not bound by the theory below, it is believed that in order to improve the adhesiveness, solder heat resistance, and adhesiveness after PCT, it is necessary to have an appropriate balance between the length of the protrusions formed by the oxidation treatment and the amount of Ni adhered by the plating treatment.

According to the present invention, a novel composite copper member and a power module including the composite copper member can be provided.

Claims (12)

1. A composite copper member comprising a copper member and a nickel layer on at least a part of the surface of the copper member,
   The surface on the side containing the nickel layer has a convex portion,
   In a cross-sectional image of the composite copper member taken by a scanning electron microscope, the number of protrusions having a length of 50 nm or more and 1500 nm or less when measured in a direction parallel to the surface on the side containing the nickel layer is 5 or more per 3.8 μm,
   A composite copper member, wherein the nickel layer has an average amount of nickel of 2 to 35 mg/dm ² .
2. The composite copper member according to claim 1, wherein the amount of nickel is a value measured by high frequency inductively coupled plasma emission spectrometry.
3. A composite copper member comprising a copper member and a nickel layer on at least a part of the surface of the copper member,
   The surface on the side containing the nickel layer has a convex portion,
   In a cross-sectional image of the composite copper member taken by a scanning electron microscope, the number of protrusions having a length of 50 nm or more and 1500 nm or less when measured in a direction parallel to the surface on the side containing the nickel layer is 5 or more per 3.8 μm,
   Copper is detected at a depth of 0 to 350 nm on the surface of the composite copper member that includes the nickel layer, and
   (A) at a depth of 10 nm from the surface on the side containing the nickel layer, the mass content of Ni is greater than the mass content of Cu;
   (B) at a depth of 10 nm from the surface, the ratio of the mass of Ni to the total mass of the three components Cu, O, and Ni is 40 to 100%; and (C) at a depth of 10 nm from the surface, the ratio of the mass of Cu to the total mass of the three components of Cu, O, and Ni is 0 to 50%;
   A composite copper member that satisfies any one of the group consisting of:
4. The composite copper member according to claim 3, wherein the detection of copper and the mass spectrometry of Cu, O, and Ni are performed by X-ray photoelectron spectroscopy.
5. The composite copper member according to any one of claims 1 to 4, further comprising a silicon layer on the surface containing the nickel layer.
6. The composite copper member according to claim 5, wherein the amount of silicon in said silicon layer is 35 µg/cm ² or more.
7. The composite copper member according to claim 5 or 6, wherein a SixOy peak and an N compound peak are detected from the surface of the silicon layer by TOF-SIMS surface analysis.
8. A power module comprising the composite copper member according to any one of claims 1 to 7, and an insulating material adhered to the composite copper member on the surface including the nickel layer.
9. The power module according to claim 8, wherein the filler filling amount of the insulating material is 50 wt% or more.
10. The power module according to claim 8 or 9, wherein the insulating material is sheet resin or mold resin.
11. A method for manufacturing a composite copper member according to claim 1,
    forming a copper oxide layer on the surface of the copper member by oxidizing the surface;
    forming a nickel layer on the surface of the copper member by plating the surface on which the copper oxide layer is formed with nickel;
    A manufacturing method, including:
12. A method for manufacturing a power module according to claim 8,
    A manufacturing method comprising the step of bonding the composite copper member according to any one of claims 1 to 7 to the insulating material on the surface of the side including the nickel layer.

Images