Aluminum wiring material and method for manufacturing same

Abstract

Provided are: an aluminum wiring material having superior long-term reliability of bonding in temperature cycles, superior shape conformability to a wedge tool, and superior vibration resistance; and a method for manufacturing the same. An aluminum wiring material comprising aluminum having a purity of less than 99.9 mass%, wherein, in a cross section perpendicular to the longitudinal direction thereof: the orientation ratio of the <112> crystal orientation having an angle difference of 10° or less is 45% or less; the ratio of the number of precipitated particles having an aspect ratio of 4 or more to the total number of precipitated particles is 0.5% to 16%; the load stress ratio is 0.4 to 0.9; and the yield strength ratio is more than 1.0, but not more than 2.3. The total amount of iron and / or silicon contained with respect to the total amount of the aluminum wiring material is 0.01 mass% to 1 mass%, and the total amount of gallium and / or vanadium contained is 50 ppm by mass to 800 ppm by mass.

Description

Aluminum wiring material and its manufacturing method

[0001] The present invention relates to an aluminum wiring material, and more particularly to an aluminum wiring material for electrical conduction that has excellent bonding reliability at high temperatures, excellent followability during bonding, and high strength against vibration, and a method for manufacturing the same.

[0002] Conventionally, aluminum wiring materials have been used as wiring materials for connecting electrodes and bus bars in lithium ion batteries and as bonding wiring materials for power semiconductors.

[0003] Lithium-ion batteries are used as drive batteries (power sources) in hybrid vehicles (HVs), plug-in hybrid vehicles (PHVs, PHEVs), and electric vehicles. Hybrid vehicles and plug-in hybrid vehicles are powered by a gasoline engine and an electric motor. Electric vehicles are powered solely by an electric motor. Hybrid vehicles and electric vehicles run by driving the electric motor with electricity stored in the lithium-ion battery. In addition to lithium-ion batteries, other drive batteries include lead-acid batteries and nickel-metal hydride batteries, but because lithium-ion batteries are light, small, high-output, and have a long lifespan, they are increasingly being used in electric vehicles that prioritize fuel efficiency.

[0004] Cylindrical batteries are often used as lithium batteries for electric vehicle drive batteries due to their small size, stable shape, and ease of battery replacement. Cylindrical batteries have a structure in which an electrolyte solution and other components are sealed in a cylindrical metal case. A battery pack using cylindrical batteries consists, for example, of an outer case connected to lead members (bus bars), multiple cylindrical batteries housed in the outer case, and wiring materials connecting the electrodes of the cylindrical batteries to the lead members. Aluminum wiring materials made of aluminum alloys are widely used as wiring materials for cylindrical batteries because of their advantages in terms of conductivity and cost.

[0005] Power semiconductors are also known as power chips, power semiconductors, power devices, power elements, and power semiconductor elements. They consist of power chips and bonding wiring materials that connect the power chips to each other and to external electrodes. Power semiconductors drive motors, charge batteries, and supply power to operate microcontrollers and large-scale integrated circuits (LSIs). The main power chips used in transportation applications, such as automobiles, are insulated-gate bipolar transistors (IGBTs) and metal-oxide-semiconductor field-effect transistors (MOSFETs), which are suitable for high power. Relatively thick aluminum wiring with a diameter of 40 μm to 700 μm is often used for bonding wiring materials for power semiconductors to carry large currents. A power module is a package that integrates multiple power semiconductors into a power supply circuit.

[0006] When joining a bus bar to wiring material or bonding wiring material for power semiconductors, a common method is to hold the tip of the wiring material with a gripping device called a wedge tool (sometimes simply called a "tool"), apply ultrasonic waves to the wiring material, and then press the wiring material together with the wedge tool against a lead material or electrode, crushing and deforming it, thereby joining the lead material or electrode to the wiring material.

[0007] Electric vehicles are used in a variety of external environments, including cold regions, tropical regions, and regions with high humidity and salinity. For example, when an electric vehicle is driven in a cold or tropical region, it is exposed to low and high temperatures and their temperature fluctuations for long periods of time. To ensure the safety of electric vehicles, the power modules and drive batteries installed in electric vehicles, as well as the aluminum wiring materials connected to them, are required to be compatible with all possible environments in which electric vehicles will be used, such as those mentioned above.

[0008] In addition to temperature changes in the external environment, electric vehicle power modules and drive batteries are also subject to temperature changes due to heating and cooling caused by electrical current. Electric vehicles undergo a cycle of turning off electricity when stopped and turning it on when starting up. This causes the power chip to heat up and reach high temperatures due to the current flow, and then rapidly cools down when the current is turned off. Aluminum wiring materials bonded to the power chip and drive battery are also subjected to temperature cycles of heating and cooling. When an electric vehicle is driven for long periods of time, frequent acceleration and braking also cause these temperature cycles to be repeated frequently. During this process, for example, thermal stress occurs due to the difference in thermal expansion coefficients between the power chip and the aluminum wiring material, which can lead to metal fatigue in the joint between the power chip and the aluminum wiring material and in the aluminum wiring material. This increases the possibility of peeling or fracture of the joint and the occurrence of cracks in the aluminum wiring material. Therefore, there is a need for aluminum wiring materials that can maintain long-term reliability of the joint even when such temperature cycles of heating and cooling are repeated.

[0009] Furthermore, power transistors, a type of power semiconductor, are used in inverters. Inverters are installed in energy-saving home appliances (hereinafter also referred to as "energy-saving") such as air conditioners, refrigerators, and washing machines. The power transistor switches the current on and off, and the inverter controls the rotation speed of the motor that drives the home appliance. By using an inverter, unnecessary motor movement can be reduced, achieving energy savings while, for example, maintaining a stable room temperature in an air conditioner. In the switching operation of a power transistor, similar to the above, a temperature cycle of heating and cooling of the wiring material is repeated by turning on and off current.

[0010] Conventionally, in order to prevent destruction or failure of power modules under high temperature conditions, technologies for avoiding chip damage, technologies for improving the strength of bonding wires, technologies for improving the long-term reliability of bonding parts, etc. have been considered (see, for example, Patent Documents 1 to 5).

[0011] Furthermore, with the miniaturization of home appliances, attempts are being made to miniaturize, thin, and increase the density of power semiconductors. This reduces the bonding space for aluminum wiring materials, resulting in greater bending of the aluminum wiring materials. Therefore, aluminum wiring materials are required to have the ability to conform to such bending. When aluminum wiring materials with poor conformability are used, when the wiring material is bent laterally at a predetermined angle (e.g., 45°) using a wedge tool, the wiring material cannot conform to the lateral bending movement of the wedge tool, and a portion of the wiring material may become detached from the alligator jaws of the wedge tool. If a second bonding is performed with a portion of the wire detached, the bonding position may deviate from the intended position and contact another electrode, potentially causing a short circuit. Furthermore, if the wire becomes detached from the tool, the tip of the wedge tool may come into contact with the semiconductor device, destroying the semiconductor device. In particular, conventional wiring materials designed for long-term reliability generally tend to have high tensile strength, yield strength, and hardness (hereinafter collectively referred to as "strength"). On the other hand, wiring materials with higher strength tend to have poorer conformability. Therefore, attempts have been made to achieve both long-term reliability and wire followability (see, for example, Patent Document 5).

[0012] The invention described in Patent Document 1 is characterized in that "an aluminum alloy thin wire for ultrasonic bonding of semiconductor devices, which consists of 0.2 to 2.0 mass % iron (Fe) and the remainder being aluminum (Al) with a purity of 99.99 mass % or more, wherein iron (Fe) is dissolved in an aluminum (Al) matrix of the aluminum alloy thin wire at a rate of 0.01 to 0.05%, and the wire-drawn matrix structure in the cross section of the aluminum alloy thin wire is a homogeneous fine recrystallized structure on the order of several μm, with iron (Fe)-aluminum (Al) intermetallic compound particles uniformly crystallized at the interface and inner surface of the structure." Patent Document 1 describes that by adding a process of solution treatment and rapid cooling before thermal refining, the amount of iron (Fe) dissolved in the aluminum (Al) matrix is ​​increased to 0.052%, which is the solid solubility limit at 650°C, and that subsequent normal cold continuous wire drawing and subsequent thermal refining can refine the crystal grain size of the Al-Fe alloy wire, and that by highly purifying Al, dynamic recrystallization occurs during bonding, thereby avoiding chip damage (see paragraph 0013 of the specification).

[0013] The invention described in Patent Document 2 is "an aluminum alloy thin wire for ultrasonic bonding to an aluminum pad of a semiconductor element, the wire comprising 0.01 to 0.2 mass% iron (Fe), 1 to 20 mass ppm silicon (Si), and the remainder being a high-purity aluminum (Al) alloy. The aluminum alloy thin wire is an alloy consisting of 0.01 to 0.2 mass% iron (Fe), 1 to 20 mass ppm silicon (Si), and the remainder being aluminum (Al) with a purity of 99.997 mass% or more, the amount of dissolved Fe being 0.01 to 0.06%, the amount of precipitated Fe being 7 times or less the amount of dissolved Fe, and the wire has a fine structure with an average crystal grain size of 6 to 12 μm." Patent Document 2 describes that maintaining the ratio of the amount of precipitated Fe to the amount of dissolved Fe within a certain range stabilizes the recrystallization temperature, and that adding a small amount of Si improves strength, resulting in stable thermal shock test results (see paragraph 0012 of the specification).

[0014] The invention described in Patent Document 3 is characterized in that "the wiring is made of Al or an Al alloy, has an average crystal grain size of 0.01 to 50 μm in a cross section perpendicular to the axis of the wiring, and is characterized in that, when the crystal orientation is measured on the cross section perpendicular to the axis of the wiring, the ratio of the crystal orientation in the longitudinal direction of the wiring to the <111> crystal orientation, which has an angle difference of 15° or less with respect to the longitudinal direction of the wiring, is 30 to 90%." Patent Document 3 also describes that the reliability of the joints in the semiconductor device can be ensured even when the semiconductor device is used for a long time in a high-temperature environment (see paragraph 0012 of the specification).

[0015] The invention described in Patent Document 4 is characterized in that "by mass%, Fe is contained in an amount of 0.02 to 1%, and further, at least one of Mn and Cr is contained in an amount of 0.05 to 0.5% in total, with the remainder consisting of Al and inevitable impurities, and the total amount of solute Fe, Mn, and Cr is 0.01 to 1%." Patent Document 4 describes that by containing a predetermined amount of one or both of Mn and Cr in addition to Fe, and by setting the total amount of solute Fe, Mn, and Cr to 0.01 to 1% in the solution heat treatment and subsequent quenching treatment, the recrystallization temperature of the wire is increased, and the progress of recrystallization of the bonding wire can be sufficiently suppressed even when the semiconductor device is used for a long period of time in a high-temperature environment, and a decrease in the strength of the wire can be prevented (see paragraph 0014 of the specification). Furthermore, it is described that in a cross section (C cross section) perpendicular to the longitudinal direction of the bonding wire, the area ratio of crystals in which the angle difference between the crystal <111> orientation and the longitudinal direction of the wire is within 15° (<111> orientation area ratio) is preferably 30 to 90%, and that this allows recrystallization due to the thermal refining treatment during wiredrawing to proceed appropriately, softening the wire and preventing chip cracking during bonding, deterioration of the bondability of the joint, etc. (see paragraph 0026 of the same specification).

[0016] The invention described in Patent Document 5 is "an aluminum wire made of an aluminum alloy having an aluminum purity of 99% by mass or more, containing a total of 0.01% by mass or more and 1% by mass or less of iron and silicon relative to the total amount of the aluminum alloy, characterized in that the (111) orientation index is 1 or more, the (200) orientation index is 1 or less, and the area ratio of precipitated particles is 0.02% or more and 2% or less in a cross section of the aluminum wire perpendicular to the wire axis. Patent Document 5 describes that by containing a predetermined amount of iron and silicon and setting the (111) orientation index to 1 or more and the (200) orientation index to 1 or less in a cross section perpendicular to the wire axis, a long life is achieved in a power cycle test (see paragraph 0044 of the specification), and further, by setting the area ratio of precipitated particles to 0.02% or more and 2% or less, it is possible to prevent the wire from coming off a wedge tool during bonding wire bonding for power semiconductors (see paragraph 0050 of the specification).

[0017] Japanese Patent Application Laid-Open No. 2013-258324 Japanese Patent Application Laid-Open No. 2014-129578 International Publication No. 2020 / 184655 Japanese Patent Application Laid-Open No. 2020-059886 International Publication No. 2022 / 163606

[0018] As examples of the above-mentioned small power semiconductors, power semiconductors with a thickness of 65 μm and IGBTs with a thickness of up to 50 μm have been known. In addition to miniaturizing power semiconductors, efforts have also been made to miniaturize and thin semiconductor packages. This is because miniaturizing and thinning semiconductor packages can lead to weight reductions in electric vehicles, for example, and as a result, improved fuel efficiency. Specifically, IGBT packages with heights (thicknesses) of approximately 20 to 40 mm became widespread in the 2000s. As of 2014, IGBT packages with a height of 14 mm have also been used. More recently, next-generation general-purpose packages with a height of 5 mm have also emerged. A semiconductor package includes a power semiconductor, bonding wiring material, and, if necessary, a resin for encapsulating these components.

[0019] Furthermore, the development of electric vehicles and home appliances with higher performance and more functionality is also progressing. To achieve high performance and multi-functionality, a high driving force is required, and therefore a larger current is supplied to the power semiconductor. Therefore, power semiconductors aiming for higher performance and multi-functionality often use thicker aluminum wiring material, which can easily pass a large current. For example, aluminum wiring material with a wire diameter of about 300 μm has traditionally been used, and some products are now using aluminum wiring material with a wire diameter of about 500 μm.

[0020] Reducing wiring resistance in semiconductor packages is also important from the perspective of energy conservation in electric vehicles and home appliances. To reduce wiring resistance, i.e., the power consumed by the electrical resistance of aluminum wiring materials, it is effective to reduce the length of the wiring material by joining the loop with a low height (low loop) and a short loop length (short loop). Here, loop height refers to the height from the lowest point of the wiring material joint to the highest point of the loop, and loop length refers to the linear distance between both ends of the first and second joints.

[0021] Furthermore, as mentioned above, conventional wiring materials designed for long-term reliability tend to have higher strength, while higher strength wiring materials have poorer vibration resistance. Increasing the elongation of aluminum wiring materials by reducing their alloying components improves vibration resistance, but this results in lower fracture strength and reduced long-term reliability of the bond. In particular, in recent years, higher long-term bond reliability has been required, and as aluminum wiring materials have become stronger, achieving both vibration resistance and reliability has become more difficult. Similarly, achieving both long-term bond reliability and wedge tool compliance has become increasingly difficult. Figure 1 shows enlarged photographs of the first and second bonded sections of a bonding example (Bonding Example 1) in which tool disengagement occurred due to wiring material with insufficient wedge tool compliance, and a bonding example (Bonding Example 2) in which tool disengagement did not occur. The upper and lower photographs on the left side of Figure 1 show the first and second bonded sections formed by wiring material that experienced wedge tool disengagement. The upper and lower photographs on the right side of Figure 1 show normal first and second bonded sections formed by wiring material that did not experience wedge tool disengagement. For example, the bonding material of bonding example 1 does not have sufficient conformability, and non-adhesion occurs at the first bonding portion due to tool separation.

[0022] Electric vehicles are often used in a variety of environments, including areas with poorly maintained roads and uneven terrain. For example, when an electric vehicle travels on uneven terrain, the unevenness of the road surface constantly exposes not only the vehicle body but also the various components and devices installed in the electric vehicle to vibration. Therefore, there is concern that power semiconductors using aluminum wiring material with insufficient vibration resistance may malfunction while the electric vehicle is in operation. Furthermore, the power semiconductors are also subjected to temperature cycles while the electric vehicle is in operation. Therefore, to ensure safety, aluminum wiring materials installed in electric vehicles must have high vibration resistance and long-term bonding reliability.

[0023] The inventors of the present invention have been working hard to develop aluminum wiring materials that are small, thin, have high driving force, and are energy-efficient, aiming to apply them to products with higher performance and functionality. As a result, they discovered that when aluminum wiring materials are bonded to power chips in a small bonding space, bending stress at the neck portion of the aluminum wiring material (the rising portion of the wiring material bonded to the semiconductor) increases as the diameter of the wiring material increases.

[0024] Furthermore, in power semiconductor packages that are exposed to temperature cycles of heating and cooling caused by electric current, the temperature cycles can cause breakage of the joints, particularly cracks on the lower side of the neck (the side that is joined to the chip), and cracks on the upper side due to bending stress on the neck.The researchers also found that, depending on how these cracks occur, they can cause breakage of the wiring material.

[0025] In recent semiconductor packages, which have become smaller and thinner, increased bending stress has been found to cause cracks due to vibration, particularly on the upper side (opposite the chip bonding side), which can lead to breakage of the wiring material. Here, upper cracks in wiring material will be described with reference to FIGS. 5 to 8 . FIG. 5 is a diagram showing an observation region R of an aluminum wiring material when observing upper cracks. In FIG. 5 , an aluminum wiring material 51 is second-bonded (wedge-bonded) to an aluminum pad 53 on the surface of a TEG (Test Element Group) chip 52. FIG. 6 shows a photograph of the portion of the neck rise of the aluminum wiring material taken with a scanning electron microscope (SEM-BEI) backscattered electron image (BEI), corresponding to the observation region R, after wedge-bonding the aluminum wiring material to the TEG chip with a low loop, sealing the aluminum wiring material with resin, and then repeatedly heating and cooling the aluminum wiring material with an electric current and applying vibrations a certain number of times. FIG. 7 is a magnified photograph of a portion of FIG. 6 . 6 and 7 show that small cracks exist in region P shown above the wiring material in the rising neck portion. In contrast, FIG. 8 is a photograph of aluminum wiring material in which no cracks were observed in observation region R. In recent semiconductor packages that have become smaller and thinner, fine cracks such as those shown in FIGS. 6 and 7 can occur above the neck portion. Cracks can also occur below the neck portion. When cracks progress both above and below the neck portion, the combined effect of these cracks can accelerate the progression of the cracks, potentially increasing the risk of neck fracture.

[0026] The present invention has been made to solve the above-mentioned problems, and an object of the present invention is to provide an aluminum wiring material that has excellent long-term bonding reliability under temperature cycles, excellent wedge tool compliance, and excellent vibration resistance. Another object of the present invention is to provide a method for manufacturing an aluminum wiring material that has excellent long-term bonding reliability, excellent bending compliance, and excellent vibration resistance.

[0027] The present invention provides an aluminum wiring material made of aluminum with a purity of less than 99.9% by mass, wherein in a cross section perpendicular to the longitudinal direction of the aluminum wiring material, the ratio of crystal orientations <112> with an angle difference of 10° or less with respect to the longitudinal direction is 45% or less, the ratio of the number of precipitated particles with an aspect ratio of 4 or more to the total number of precipitated particles in the cross section is 0.5% to 16%, the load stress ratio defined in the following (1) is 0.4 to 0.9, where: load stress ratio = fatigue limit / tensile strength (1), and the yield strength ratio defined in the following (2) is greater than 1.0 and less than 2.3, where: yield strength ratio = maximum stress / 0.2% yield strength (2). The aluminum wiring material contains, relative to its total weight, at least one element selected from iron (Fe) and silicon (Si) in a total amount of 0.01 mass% to 1 mass% and at least one element selected from gallium (Ga) and vanadium (V) in a total amount of 50 mass ppm to 800 mass ppm. [2] The aluminum wiring material according to [1], characterized in that the average crystal grain size in a cross section perpendicular to the longitudinal direction of the aluminum wiring material is 40 μm or less. [3] The aluminum wiring material according to [1] or [2], characterized in that the aluminum wiring material further contains one or more elements selected from magnesium (Mg), copper (Cu), nickel (Ni), zinc (Zn), chromium (Cr), manganese (Mn), titanium (Ti), zirconium (Zr), tungsten (W), and scandium (Sc). [4] The aluminum wiring material according to any one of [1] to [3], wherein the wire diameter of the aluminum wiring material is 15 μm or more and 700 μm or less. [5] The aluminum wiring material according to any one of [1] to [4], wherein a cross section perpendicular to the longitudinal direction of the aluminum wiring material is circular, oval, or elliptical. [6] The aluminum wiring material according to any one of [1] to [5], wherein the aluminum content in the aluminum wiring material is 99 mass% or more and less than 99.9 mass% with respect to the total amount of the aluminum wiring material.

[0028] [7] A method for manufacturing an aluminum wiring material, comprising the steps of: preparing an aluminum alloy having a purity of less than 99.9% by mass, the aluminum alloy containing, relative to a total amount, at least one element of iron (Fe) and silicon (Si) in a total amount of 0.01% by mass or more and 1% by mass or less, and at least one element of gallium (Ga) and vanadium (V) in a total amount of 50 ppm by mass or more and 800 ppm by mass or less; and wiredrawing the aluminum alloy, wherein in a cross section perpendicular to the longitudinal direction of the aluminum wiring material, an orientation ratio of a crystal orientation <112> having an angular difference with respect to the longitudinal direction of the aluminum wiring material of 10° or less is 45% or less; a ratio of the number of precipitated particles having an aspect ratio of 4 or more to the total number of precipitated particles in the aluminum wiring material is 0.5% or more and 16% or less; and a load stress ratio defined in the following (1) is 0.4 or more and 0.9 or less, where load stress ratio = fatigue limit / tensile strength (1) (2) A method for producing an aluminum wiring material, wherein a yield strength ratio defined in (2) below is greater than 1.0 and is not greater than 2.3, Yield strength ratio=maximum stress / 0.2% yield strength. [8] A method for manufacturing an aluminum wiring material, comprising the steps of: preparing an aluminum alloy having a purity of less than 99.9% by mass, the aluminum alloy containing, relative to a total amount, at least one element of iron (Fe) and silicon (Si) in a total amount of 0.01% by mass or more and 1% by mass or less, and at least one element of gallium (Ga) and vanadium (V) in a total amount of 50 ppm by mass or more and 800 ppm by mass or less; and wiredrawing the aluminum alloy under conditions of an area reduction rate (working rate) per die of 5% to 40%, wherein in a cross section perpendicular to the longitudinal direction of the aluminum wiring material, an orientation ratio of crystal orientation <112> having an angular difference with respect to the longitudinal direction of 10° or less is 45% or less; a ratio of the number of precipitated particles having an aspect ratio of 4 or more to the total number of precipitated particles is 0.5% or more and 16% or less; and a load stress ratio defined in the following (1) is 0.4 or more and 0.9 or less; Load stress ratio = fatigue limit / tensile strength (1) A yield strength ratio defined in the following (2) is greater than 1.0 and not greater than 2.3, Yield strength ratio = maximum stress / 0.2% yield strength (2) A method for producing an aluminum wiring material.

[0029] In this specification, the symbol "to" indicates a range of values ​​including the values ​​before and after it.

[0030] According to the present invention, it is possible to provide an aluminum wiring material that has excellent long-term bonding reliability under temperature cycles, excellent wedge tool compliance, and excellent vibration resistance. Furthermore, according to the present invention, it is possible to provide a method for manufacturing an aluminum wiring material that has excellent long-term bonding reliability, excellent bending compliance, and excellent vibration resistance. This can contribute to the miniaturization of power semiconductors and the like that use aluminum wiring material.

[0031] 6 is a magnified photograph (left) of a joint made of wiring material that has experienced wedge tool disengagement, and a photograph (right) of a normal joint made of wiring material that has not experienced wedge tool disengagement. It is a graph of an S-N curve. It is a diagram that schematically shows an example of the cross-sectional shape of an aluminum wiring material. It is a diagram that schematically shows a simulated power semiconductor device for evaluation, for evaluating power cycle life and vibration fatigue in a resin-encapsulated state. It is a diagram that shows an observation region R of an aluminum wiring material when observing an upper crack. It is a photograph of an aluminum wiring material that has experienced a crack on the upper side. It is a photograph of an aluminum wiring material that has not experienced a crack on the upper side, which is an enlarged photograph of a portion of FIG. 6.

[0032] An aluminum wiring material according to an embodiment of the present invention is described below. The aluminum wiring material according to the embodiment is made of aluminum with a purity of less than 99.9% by mass, and in a cross section perpendicular to the longitudinal direction of the aluminum wiring material, the orientation ratio of the crystal orientation <112>, which has an angle difference with the longitudinal direction of the aluminum wiring material of 10° or less, is 45% or less.

[0033] The inventors fabricated multiple wiring material samples using aluminum of different compositions and manufacturing conditions, and evaluated the long-term reliability of the bonding (hereinafter also referred to as "power cycle life") for each sample, as described below. They also carefully observed the cross-sectional structure perpendicular to the longitudinal direction of the wiring material samples. As a result, they discovered a correlation between the abundance ratio of a specific crystal orientation and the power cycle life. Specifically, they discovered that when the abundance ratio of the <112> crystal orientation, which has an angle difference of 10° or less with respect to the longitudinal direction, is within a predetermined range, the power cycle life is significantly superior.

[0034] Controlling the proportion of the <111> crystal orientation is widely practiced in the manufacturing process of aluminum wiring materials, including those used for conducting semiconductor packages. Controlling the proportion of the <111> crystal orientation has enabled the practical application of wire products with excellent processability and softness. However, in the aluminum wiring materials used for conducting semiconductor packages, which have become smaller and thinner in recent years, it has been found that controlling the proportion of the <111> crystal orientation alone is insufficient to achieve the aforementioned problems of improving long-term reliability of bonding under temperature cycles, improving wedge tool compliance, and improving vibration resistance. Therefore, the inventors have concluded that mixing specific orientations that have higher deformation resistance than the <111> orientation due to the nature of the metal structure and optimizing their proportions is important to solving the above-mentioned problems. They have then discovered that controlling the proportion of the <112> orientation, particularly among these specific orientations, within a certain range can solve the above-mentioned problems. In other words, by maintaining the ratio of the <112> crystal orientation within a certain range and simultaneously adjusting the other properties of the wiring material, the synergistic effect of these factors makes it possible to maintain vibration resistance and wedge tool compliance while achieving a long life, particularly in terms of power cycles.

[0035] Specifically, in the aluminum wiring material of this embodiment, among the crystal orientations in a cross section perpendicular to the longitudinal direction, the abundance ratio of the <112> crystal orientation, which has an angle difference with the longitudinal direction of 10° or less, is 45% or less, thereby achieving an excellent power cycle life and also achieving high vibration resistance, which will be described later. From these viewpoints, the abundance ratio of the <112> crystal orientation is preferably 10% or more and 40% or less, and more preferably 15% or more and 35% or less.

[0036] (Evaluation of Power Cycle Life) The power cycle test is a test in which current is passed through the aluminum wiring material so that the surface temperature of the power chip bonded to the aluminum wiring material reaches 150°C, and then the current is stopped and the chip is cooled down to a surface temperature of 25°C. In other words, this test repeats cycles of rapid cooling and rapid heating with a temperature difference of 125°C, and evaluates the number of cycles until a problem occurs in the operation of the power chip. The number of cycles at which current can no longer be passed is determined as the sample life (power cycle life) of the power cycle test. The longer the power cycle life, the better the long-term reliability of the bond.

[0037] Next, the inventors carefully observed the metallographic structure of cross sections perpendicular to the longitudinal direction of wiring material samples fabricated in the same manner as above, and discovered that the shape (aspect ratio of the major axis to the minor axis when viewed as an ellipse) and the distribution of aspect ratios of precipitated particles on the aluminum-based matrix varied depending on the composition and manufacturing conditions of the wiring material sample.

[0038] The inventors hypothesized that the presence of precipitated particles with a specific aspect ratio among these precipitated particles in a cross section perpendicular to the longitudinal direction is related to the trackability of the wiring material. Based on this hypothesis, photographs of precipitated particles taken with SEM-BEI (scanning electron microscope backscattered electron imaging) were analyzed using image processing software, and the precipitated particles in the images were quantitatively evaluated. The quantitative evaluation method will be described in detail later, but it was discovered that the aspect ratio and aspect ratio distribution of the precipitated particles observed in a cross section perpendicular to the longitudinal direction of the wiring material are related to the trackability of the wiring material.

[0039] In the aluminum wiring material of this embodiment, the ratio of the number of precipitated particles having an aspect ratio of 4 or more to the total number of precipitated particles in a cross section perpendicular to the longitudinal direction is 0.5% or more and 16% or less. In terms of improving the tracking ability, the ratio of the number of precipitated particles having an aspect ratio of 4 or more is preferably 0.7% or more and 12% or less, and more preferably 1% or more and 8% or less.

[0040] The precipitated particles are particles with a size (maximum particle length) of about 30 μm or less, and are shaped like chunks, rings, plates, needles, roughly spherical particles, irregular shapes, etc. These are thought to include particles that crystallize or precipitate during the manufacturing process, as well as particles contained in the aluminum raw material. The precipitated particles are also thought to include any one or more of an aluminum-iron alloy, an intermetallic compound, an aluminum-iron-silicon alloy, an intermetallic compound, and a precipitate of elemental silicon.

[0041] The number of precipitated particles and the distribution of the aspect ratios of the precipitated particles can be controlled by the composition of the wiring material (the content ratio of iron and silicon), the heat treatment temperature, time, timing of the heat treatment, wire drawing conditions, etc. When selecting measurement locations, it is preferable to ensure the objectivity of the measurement data. As a method for this, it is preferable to obtain measurement samples from the wiring material to be measured at intervals of, for example, 3 to 5 m in the longitudinal direction of the wiring material, and provide them for measurement. However, if it is difficult to obtain samples at this interval, for example, when the wiring material is short, shorter intervals may be used as long as the objectivity of the measurement data can be ensured.

[0042] The number of precipitated particles and the distribution of their aspect ratios can be observed as follows. The observation surface of the specimen is exposed by embedding polishing, and then buffing and milling are performed to remove surface irregularities and contamination (to achieve an extremely clean surface). The precipitated particles are identified at 1,000x magnification using a field emission scanning electron microscope (FE-SEM, Hitachi High-Tech SU8220), and then observed using backscattered electron imaging (BED-C) with an accelerating voltage of 5 kV and a working distance (WD) of 10 mm. Specifically, the precipitated particles are observed over a wide area of ​​the cross section at 400x magnification, and one to three average locations of the particle distribution are then observed at 1,000x magnification. If it is difficult to identify the average location at 1,000x magnification, it is recommended to observe approximately three locations at 1,000x magnification. When observing multiple locations in this way, the average value of the observed locations should be used as the aspect ratio value. Furthermore, if the entire sample needs to be observed, it can be observed at 400x magnification, but for samples with fine particles, it is preferable to observe at 1,000x magnification rather than 400x magnification. In this regard, the observer can select the observation magnification appropriately depending on the wire diameter and particle size, but from the perspective of measurement accuracy, observation at 1,000x magnification can obtain more accurate values. In other words, observation of precipitated particles is preferably performed at a magnification of 400x or more and 1,000x or less, with higher magnifications within this range being more preferable. Furthermore, in this FE-SEM observation, the images obtained are adjusted so that the precipitated particles and the aluminum substrate are clearly visible. Furthermore, using the EDX (BRUKER Flat QUAD) attached to the same device, it was confirmed by mapping and point analysis of several locations that the precipitate particles detected in the FE-SEM image at 1,000x magnification were a different element (or alloy) from the aluminum base material, and areas with the same contrast (brightness) were considered to have the same element (or alloy) composition.In image data (TIFF file, data size 1M or more, 1280 x 1024 pixels or more) obtained by FE-SEM observation of a cross section perpendicular to the wire axis of aluminum wiring material, the composition of the aluminum wiring material and other areas differs, so precipitate particles with an atomic weight (density) smaller than aluminum, for example, are displayed as pixels with a high brightness value.In this case, regions other than precipitate particles (in the aluminum matrix) are displayed with low brightness values. The brightness values ​​of this FE-SEM image are normalized to a range of 0 to 1 to make it easier to distinguish precipitate particles. A brightness threshold value (e.g., 0.95) that allows for separation of precipitate particles from other regions is then determined using a histogram or the like, and the image is binarized to 1 or 0 at the threshold. Regions with a brightness value of 0 are thus displayed as black, and regions with a brightness value of 1 are displayed as white. The number of regions observed as precipitate particles (regions with a brightness value of 1) in this processed image that has been binarized to 1 or 0, and the aspect ratio of each region, are measured using an image analysis program (e.g., an image analysis program using MATLAB). From the calculated number of precipitate particles and the aspect ratio of each precipitate particle (aspect ratio distribution), the ratio of the number of precipitate particles with an aspect ratio of 4 or more to the total number of precipitate particles is calculated. The area observed as a precipitated particle was approximated to an ellipse, and the value of the major axis / minor axis of the ellipse was calculated as the aspect ratio of the precipitated particle. The elliptical approximation of the precipitated particle can be performed by the least squares method using the central coordinates of the precipitated particle and the coordinates of each point on the periphery of the precipitated particle area. Note that if it is difficult to identify the precipitated particle using FE-SEM, it may be possible to binarize the image using EDX mapping, but it is preferable to identify the precipitated particle using FE-SEM.

[0043] Furthermore, it has been found that the fine crystal grains in the crystalline structure of the aluminum wiring material of this embodiment increase the elasticity of the crystalline structure, which, in combination with the abundance rate of the <112> orientation ratio, further improves wedge tool followability. Specifically, in the aluminum wiring material of this embodiment, the average crystal grain size in a cross section perpendicular to the longitudinal direction is preferably 40 μm or less, which makes it easier to obtain wedge tool followability. The average crystal grain size is preferably 30 μm or less, and more preferably 20 μm or less. The average crystal grain size is usually 1 μm or more.

[0044] The electron backscatter diffraction pattern (EBSP) method can be used to measure and analyze the cross-sectional crystalline structure of aluminum wiring material. The EBSP method allows the crystalline structure of the cross-section of even thin wires such as wiring materials used in semiconductor bonding to be measured with high accuracy and sufficient reproducibility.

[0045] In the EBSP method, it is usually difficult to measure the crystal orientation with high precision when the sample surface has large irregularities or curves. Therefore, it is useful to smooth the surface of the cross section of the wiring material in order to observe it with EBSP. Methods for smoothing the surface of the cross section include mechanical polishing, chemical polishing, and FIB processing. These methods can remove residual strain on the surface of the cross section and obtain a smooth surface.

[0046] If the sample is properly pretreated as described above, the EBSP method can measure and analyze the crystal orientation of the cross section of the wiring material with high precision. By measuring the crystal orientation at at least three different locations, and preferably at least ten different locations, it is possible to obtain an average value that takes variation into account.

[0047] (Load Stress Ratio) The aluminum wiring material of this embodiment has a load stress ratio of 0.4 or more and 0.9 or less, as defined by the following (1): Load stress ratio = fatigue limit / tensile strength (1)

[0048] It is known that vibrations caused by a vehicle running or a person walking occur randomly in three axes (X, Y, and Z), and the vibration frequency is often 40 Hz or less. For example, aluminum wiring materials installed in electric vehicles are subjected to tension, compression, and bending in three axes over a long period of time due to the above-mentioned stresses. The inventors have investigated methods for evaluating performance against such complexly directional vibration loads. As a result, they have found that the above-mentioned vibration resistance correlates with the load stress ratio expressed by the above formula (1).

[0049] Generally, when steel materials are subjected to repeated stress at a constant amplitude, they will fracture due to fatigue after a certain number of cycles. The S-N curve shown in Figure 2 is a two-axis graph with the magnitude of this repeated stress on the Y axis and the number of cycles on the X axis. As the repeated stress is reduced, the S-N curve becomes horizontal around 10 to 10, and fracture may not occur even with an infinite number of repeated stress cycles. The value of the repeated stress at this point is called the fatigue limit. On the other hand, the S-N curve for non-ferrous metals such as aluminum does not have a horizontal point, as shown in Figure 2, and therefore does not have a fatigue limit. Therefore, the strength at a certain number of cycles is considered to be the fatigue limit. For example, the repeated stress that does not fracture up to 10 to 10 is called the 10 to 7 time strength. Conventionally, the stress amplitude at a number of cycles around 10 to 10 is sometimes referred to as the fatigue limit. Therefore, in this embodiment, the 10 to 7 time strength is used as the fatigue limit.

[0050] The fatigue limit is usually expressed as the maximum load stress (MPa). The fatigue limit of the present invention can be measured as follows: a tensile fatigue test is performed at room temperature (e.g., 25°C) using a dedicated tensile fatigue testing machine (ElectroPlus E3000 manufactured by Instron, hereinafter referred to as the "measuring device") in accordance with the spirit of JIS Z 2273:1978. An evaluation sample is obtained by cutting an aluminum wiring material to a length slightly longer than 100 mm to a measurement length of 100 mm. In order to calculate the value of the maximum load stress described below, the 0.2% proof stress of the aluminum wiring material is previously obtained by a tensile test. The evaluation sample is then fixed to the measuring device, and loading at the maximum load stress and unloading at the minimum load stress are repeated a maximum of more than 10 times at a rate of 5 Hz (5 times per second) (tensile fatigue test). The maximum load stress here is 90% of the 0.2% proof stress or any value less than 90%, and the minimum load stress is 10 MPa (fixed value). With 90% of the 0.2% proof stress as the maximum value, any value less than 90% is set in ascending order, and the above loading and unloading are repeated at three or more different load cycles. The results are plotted on a semi-logarithmic graph with the x-axis representing the load cycle and the y-axis representing the maximum load stress σ, and the x-axis representing a logarithmic scale (logarithm base 10) to create an S-N curve. In the tensile fatigue test, the test conditions are adjusted so that the interval between the load cycles plotted on the graph is 1 or more in logarithmic display and 10 or more in real display in the x-axis direction. An approximate line is obtained from the plotted points using the least squares method, and the maximum load stress at which fracture does not occur at a load cycle of 10 to the power of 7 is determined using this approximate line. The maximum applied stress at which the number of repeated loads is 10 to the power of 7 and does not lead to fracture is the fatigue limit (MPa). If it is difficult to evaluate the number of repeated loads is 10 to the power of 7 or more, for example, the maximum applied stress at the number of repeated loads is 10 to the power of 3 and 10 to the power of 6 and does not lead to fracture may be plotted at three or more points on a semi-logarithmic graph of the S-N curve, an approximation line is found from the plotted points by the least squares method, and the fatigue limit at the number of repeated loads is estimated from the approximation line.In this case, too, the test conditions are adjusted so that the interval between plotted points in the x-axis direction is 1 or more in logarithmic notation, and 10 or more in real notation. Furthermore, to perform this tensile fatigue test more accurately, it is important to satisfy the following three conditions: (i) both ends of the measurement length (the pulled portion) are fixed and do not move during the test; (ii) both ends of the measurement length are pulled straight toward each other in one axial direction; and (iii) the fracture site during the tensile test is not in the chuck portion.

[0051] Next, the tensile strength will be described. The tensile strength can be obtained by dividing the maximum load (kgf) when both ends of an evaluation sample are pulled using a tensile testing device by the cross-sectional area of ​​the wiring material. This maximum load is automatically calculated by converting the tensile force into an electrical signal in the tensile testing device. The load stress ratio is calculated as the value of fatigue limit / tensile strength as shown in the above formula (1), but it can also be calculated as the value of maximum load in the tensile fatigue test / maximum load in the tensile test.

[0052] In the aluminum wiring material of this embodiment, a load stress ratio of 0.4 to 0.9 significantly suppresses damage due to vibration at the neck portion of the wiring material bonded to a semiconductor chip, resulting in improved vibration resistance. The load stress ratio is preferably 0.5 to 0.9, more preferably 0.6 to 0.9, and even more preferably 0.7 to 0.9. A load stress ratio within this range can enhance the vibration resistance effect.

[0053] (Yield Strength Ratio) The aluminum wiring material of this embodiment has a yield strength ratio defined by the following (2) of more than 1.0 and not more than 2.3: Yield Strength Ratio = Maximum Stress / 0.2% Yield Strength (Stress) (2)

[0054] By setting the load stress ratio within the above range and the yield strength ratio of the aluminum wiring material to be greater than 1.0 and not greater than 2.3, the toughness of the aluminum wiring material can be increased and the vibration resistance can be further improved. The yield strength ratio is preferably 1.2 or more and 2.1 or less, and more preferably 1.4 or more and 1.9 or less.

[0055] The yield strength ratio is a value obtained by dividing the maximum stress (MPa) by the 0.2% yield strength (stress) (MPa). First, the maximum load (kgf) and 0.2% strain load (kgf) of the wiring material are determined by the above-mentioned tensile test. The maximum stress (MPa) is obtained by dividing the maximum load by the cross-sectional area of ​​the wiring material. Furthermore, the 0.2% strain load is obtained by dividing the 0.2% strain load by the cross-sectional area of ​​the wiring material. That is, maximum stress = maximum load / cross-sectional area, and 0.2% yield strength (stress) = 0.2% strain load / cross-sectional area. These can be applied to the above formula (2) to calculate the yield strength ratio. Alternatively, the yield strength ratio may be calculated as the value of the maximum load / 0.2% strain load in the tensile test.

[0056] (Evaluation of Vibration Resistance) Using an ultrasonic bonding device (REBO7 wire bonder manufactured by Ultrasonic Industries), a wiring material sample is bonded to two aluminum plates at the first and second bonding portions, respectively, so that the loop height is 1 mm and the length is 7 mm, to obtain an evaluation sample. Using this evaluation sample, a vibration fatigue test is performed using a dedicated vibration fatigue testing machine (for example, CV-101M manufactured by AS ONE Corporation) manufactured in accordance with JIS standards, and vibration resistance is evaluated. In the vibration fatigue test, vibration is applied to each sample at a frequency of 50 Hz, an acceleration of 1 G, and an amplitude of 0.099 mm until fracture. In the vibration fatigue test, the greater the total number of vibrations until fracture, the better the vibration resistance.

[0057] As described above, electric vehicles are sometimes driven in areas where roads are poorly maintained and uneven, and power semiconductors installed in electric vehicles are required to be free from failure due to vibration. The aluminum wiring material of this embodiment has excellent vibration resistance due to the above-described configuration, and can therefore suppress failure when exposed to vibration for a long period of time. Furthermore, power semiconductors are subjected to the above-described vibrations under temperature cycles caused by current. However, the aluminum wiring material of this embodiment, due to the above-described configuration, is significantly less likely to break or break even when subjected to a combination of temperature cycles and vibrations. Therefore, when used in electric vehicles, high safety can be achieved.

[0058] Ultrasonic bonding, a method for joining aluminum wiring materials, involves applying ultrasonic waves to the aluminum wiring materials while applying a load to destroy oxide films and other impurities on the joining surfaces, deforming the wiring materials and joining them. Therefore, the thicker the oxide films and other impurities on the surfaces of the bus bars and electrodes to be joined, the higher the load and energy of the ultrasonic waves applied to the aluminum wiring materials to remove them. However, applying ultrasonic waves at high loads and high energy can easily reduce the strength of the wiring materials, particularly near the joint boundary, due to the vibrations, resulting in a decrease in the long-term reliability of the bond. However, the aluminum wiring material of this embodiment can achieve excellent vibration resistance and power cycle life as described above, so that the long-term reliability of the bond can be maintained even when the aluminum wiring materials are ultrasonically bonded at high loads and high energy.

[0059] Furthermore, because low-loop joints have lower vibration resistance than high-loop joints, failures of aluminum wiring materials and their joints have traditionally been suppressed by maintaining a predetermined loop height. The aluminum wiring material of this embodiment has excellent vibration resistance, allowing for joints with a lower loop height, thereby contributing to the miniaturization of power semiconductors. In IGBTs, MOSFETs, and other devices, wiring materials and semiconductor chips are generally encapsulated with resin. However, the encapsulating resin does not completely adhere to the wiring material or semiconductor chip, resulting in gaps between the resin and the wiring material or semiconductor chip. Due to these gaps, resin-encapsulated power semiconductors require vibration resistance similar to that of non-resin-encapsulated power semiconductors. In wiring materials that have undergone thermal fatigue due to temperature cycling, even small vibrations caused by small gaps between the resin and the wiring material can cause failure. However, the aluminum wiring material of this embodiment has a long power cycle life and excellent vibration resistance, thereby suppressing fractures due to temperature cycling and vibration even in resin-encapsulated power semiconductors.

[0060] The aluminum constituting the aluminum wiring material of this embodiment contains, relative to its total amount, at least one element of iron (Fe) and silicon (Si) in a total amount of 0.01 mass% or more and 1 mass% or less, and at least one element of gallium (Ga) and vanadium (V) in a total amount of 50 mass ppm or more and 800 mass ppm or less.

[0061] The aluminum wiring material of this embodiment contains these elements, and the above-mentioned distribution of precipitated particles can be obtained by adjusting the manufacturing conditions. Therefore, the aluminum wiring material of this embodiment can obtain excellent vibration resistance and long-term bonding reliability. Iron, silicon, gallium, and vanadium may be contained in the raw materials of the aluminum wiring material in advance, or may be added during the manufacturing process of the aluminum wiring material.

[0062] In the aluminum wiring material of this embodiment, a longer power cycle life can be obtained by having the total amount of at least one of iron and silicon be 0.01% by mass or more. The total amount of iron and silicon is preferably 0.02% by mass or more, more preferably 0.05% by mass or more, even more preferably 0.1% by mass or more, and even more preferably 0.13% by mass or more, because this facilitates achieving a long life. Furthermore, the total amount of iron and silicon is more preferably 0.9% by mass or less, even more preferably 0.8% by mass or less, in order to maintain electrical conductivity while achieving a long power cycle life.

[0063] In the aluminum wiring material of this embodiment, the amount of iron is preferably 0.01% by mass or more, more preferably 0.03% by mass or more, even more preferably 0.05% by mass or more, particularly preferably 0.1% by mass or more, and even more particularly preferably 0.13% by mass or more, relative to the total amount of the aluminum alloy. The amount of iron is preferably 0.95% by mass or less, more preferably 0.9% by mass or less, relative to the total amount of the wiring material. The amount of silicon is preferably 0.01% by mass or more, more preferably 0.05% by mass or more, relative to the total amount of the aluminum alloy. The amount of silicon is preferably 0.5% by mass or less, more preferably 0.4% by mass or less. By combining the preferred ranges of iron and silicon described above, the synergistic effect of iron and silicon makes it easier to achieve long-term reliability of the bond.

[0064] In the aluminum wiring material of this embodiment, the upper limit of the content of at least one of gallium and vanadium is not particularly limited, but is approximately 1000 ppm by mass. The total amount of gallium and vanadium is preferably 800 ppm by mass or less, which makes it easier to suppress the maximum temperature reached by the aluminum wiring material when current is applied. The contents of gallium and vanadium may be 100 ppm by mass or more, or 150 ppm by mass or more, relative to the total amount of the wiring material. The contents of gallium and vanadium may be 700 ppm by mass or less, or 600 ppm by mass or less, relative to the total amount of the wiring material. When only one of gallium and vanadium is contained in the wiring material, the content of either one may be within the above range. When both gallium and vanadium are contained, the total amount of gallium and vanadium may be within the above range.

[0065] For example, if the maximum temperature reached when an aluminum wiring material with a purity of 99.99% by mass is used and current is applied is 150° C., if the gallium and vanadium contents are 1000 ppm by mass or less, the maximum temperature reached can be kept to 160° C. or less. In other words, based on an aluminum wiring material with a purity of 99.99% by mass, the temperature rise due to heat generation when current is applied can be kept to within about 10° C.

[0066] The aluminum wiring material of this embodiment may contain one or more trace elements selected from the group consisting of magnesium (Mg), copper (Cu), nickel (Ni), zinc (Zn), chromium (Cr), manganese (Mn), titanium (Ti), zirconium (Zr), tungsten (W), and scandium (Sc) in addition to iron, silicon, gallium, and vanadium. The aluminum wiring material of this embodiment may also contain inevitable impurities other than the above-described trace elements. The trace element content is a total of 0.1 mass% or less with respect to the entire wiring material. The trace element content is, for example, 0.0001 mass% or more with respect to the total amount of the aluminum alloy. These trace elements may be pre-contained in the raw materials for the aluminum wiring material, may be inevitable impurities in the aluminum raw materials, or may be added during the manufacturing process of the aluminum wiring material. The trace elements are preferably 0.005 mass% or more, more preferably 0.01 mass% or more, and even more preferably 0.015 mass% or more.

[0067] The content ratios of elements contained in the aluminum wiring material of this embodiment are generally measured by chemical analysis such as inductively coupled plasma (ICP) optical emission spectroscopy, but are not limited to this. For example, secondary ion mass spectroscopy (SIMS), glow discharge mass spectroscopy (GDMS), or energy dispersive X-ray spectroscopy (EDX) can also be used to measure the content ratios of elements contained in the aluminum wiring material of this embodiment.

[0068] The aluminum wiring material of this embodiment is preferably made of an aluminum alloy having an aluminum purity (amount of aluminum relative to the total amount of the aluminum wiring material) of 99% by mass or more. This provides sufficient conductivity and can achieve even better vibration resistance and long-term bonding reliability. The aluminum purity of the aluminum wiring material is preferably 99.9% by mass or less. An aluminum purity of 99.9% by mass or less allows the aluminum to contain sufficient amounts of iron, silicon, gallium, vanadium, and trace elements, contributing to improved long-term reliability. The purity of the aluminum alloy is more preferably less than 99.9% by mass, and even more preferably 99.8% by mass or less.

[0069] The wire diameter of the aluminum wiring material of this embodiment is typically 15 μm or more and 700 μm or less, preferably 70 μm or more and 600 μm or less, and more preferably 100 μm or more and 500 μm or less. The wire diameter of the aluminum wiring material is the diameter if the cross-sectional shape is a perfect circle, and the diameter is the length of the major axis if the cross-section is other shapes (circular, elliptical, oval, polygonal, or polygon-like shapes as shown below). However, if the cross-section is polygonal or polygon-like, the length may be measured as width or thickness. FIG. 2 schematically shows an example of the cross-sectional shape of an aluminum wiring material. As shown in FIG. 3, the cross-sectional shape of the aluminum wiring material may be circular (FIG. 3a), elliptical (e.g., FIG. 3b), oval (e.g., FIG. 3c), rectangular (e.g., FIG. 3e), triangular (e.g., FIG. 3d), or polygon-like (e.g., FIG. 3f, FIG. 3g), etc. When the cross section of the aluminum wiring material is elliptical or oval, the length of the major axis is preferably 0.3 mm to 4 mm, more preferably 0.5 mm to 3 mm, and the length of the minor axis is preferably 0.05 mm to 0.5 mm, more preferably 0.1 mm to 0.5 mm. When the cross section of the aluminum wiring material is rectangular, the length of the long side is preferably 0.3 mm to 4 mm, more preferably 0.5 mm to 3 mm, and the length of the short side is usually preferably 0.05 mm to 0.5 mm, more preferably 0.1 mm to 0.5 mm. Note that a wiring material whose cross section is polygonal, polygon-like, or oblate elliptical (a shape obtained by compressing an ellipse in a uniaxial direction) is called a ribbon, and a wiring material whose cross section is circular, elliptical, or elliptical is called a wire.

[0070] (Method for manufacturing aluminum wiring material) Next, an example of a method for manufacturing an aluminum wiring material according to an embodiment will be described. Note that the method for manufacturing an aluminum wiring material is not limited to the manufacturing method described below. It is desirable to adjust the conditions appropriately in consideration of the weight of the aluminum wiring material to be manufactured and the processing capacity of the heat treatment furnace.

[0071] Molten aluminum is prepared by dissolving iron and silicon together in high-purity aluminum of 99% by mass or higher. The purity of the high-purity aluminum used as a raw material may be 99.9% by mass or higher, or even 99.99% by mass or higher. A heating furnace such as an arc heating furnace, a high-frequency heating furnace, a resistance heating furnace, or a continuous casting furnace is used for melting. Although atmospheric melting is acceptable, the molten aluminum in the heating furnace may be melted while maintained in a vacuum or in an inert gas atmosphere such as argon or nitrogen in order to prevent contamination with oxygen or hydrogen from the atmosphere. Furthermore, when melting aluminum, it is preferable to perform an inclusion removal step in which an inert gas such as argon (Ar) gas or nitrogen gas is bubbled into the molten aluminum, followed by filtration using a filter. In the inclusion removal step, silicon dioxide (SiO ) is removed by bubbling with an inert gas. ２ ) and aluminum oxide (Al ２ O ３ Inclusions such as argon (Ar) and argon (Ar) can be agglomerated onto the surface of the molten aluminum, and then the inclusions can be removed from the molten aluminum by a subsequent filtration process. Removing the inclusions makes it easier to adjust the load stress ratio to 0.4 or higher. Using argon (Ar) gas in the inclusion removal process makes it easier to adjust the yield strength ratio and <112> orientation ratio to a higher range while increasing the load stress ratio, thereby further improving the long-term reliability and vibration resistance of the joint under temperature cycles.

[0072] Furthermore, the greater the bubbling gas flow rate and the longer the bubbling time in the inclusion removal process, the greater the inclusion removal effect, making it easier to adjust the load stress ratio to 0.4 or higher. On the other hand, if the gas flow rate is too high or the bubbling time is too long, the inclusion removal effect will be limited and then useless. Regarding the bubbling method, in order to prevent impurities from leaching into the molten aluminum, it is preferable to use a steel pipe, which has excellent corrosion resistance at high temperatures, and bubble the inert gas through the pipe. One filtration method is to pass the molten aluminum through a ceramic foam filter with a porosity of approximately 0.8 to 0.9.

[0073] Table 1 shows the difference in the load stress ratio, yield strength ratio, and <112> orientation ratio of aluminum wiring material with a final wire diameter when an inclusion removal process was performed (with bubbling) and when it was not performed (without bubbling), with other manufacturing conditions being the same. In the experimental example shown in Table 1, the effects of bubbling on composition A, which contains 0.15% by mass of iron (Fe), 0.03% by mass of silicon (Si), 0.009% by mass of gallium (Ga), 0.012% by mass of vanadium (V), and the remainder being aluminum and unavoidable impurities, and composition B, which contains 0.8% by mass of iron (Fe), 0.2% by mass of silicon (Si), 0.03% by mass of gallium (Ga), 0.03% by mass of vanadium (V), and the remainder being aluminum and unavoidable impurities, were compared. As shown in Table 1, when the inclusion removal process is performed on the molten aluminum, it is easier to adjust the load stress ratio, proof stress ratio, and <112> orientation ratio to fall within the ranges of this embodiment, compared to when the inclusion removal process is not performed.

[0074]

[0075] The molten aluminum material is solidified by continuous casting from a heating furnace to a predetermined diameter to produce an ingot, which is then set in an extruder and extruded to a predetermined wire diameter. In the manufacturing method for aluminum wiring material according to the embodiment, a scratch-prevention treatment step is preferably performed to prevent scratches on the original wire extruded to a predetermined wire diameter. The scratch-prevention treatment step facilitates adjusting the yield strength ratio to greater than 1.0 and less than 2.3. The scratch-prevention treatment can be performed by applying a treatment solution containing a surfactant as a main component to the surface of the original wire. The use of a surfactant not only reduces surface corrosion of the original wire compared to an oil-based treatment solution, but also suppresses surfactant seizure and high-temperature corrosion on the surface of the aluminum wiring material during the subsequent heat treatment step, making it easier to adjust the yield strength ratio to the above range. Furthermore, the scratch-prevention treatment step facilitates adjusting the <112> orientation ratio to 45% or less and adjusting the load stress ratio to a range of 0.4 to 0.9 while adjusting the yield strength ratio to the above range. Examples of surfactants contained in the treatment liquid for the scratch prevention treatment step include alcohol-based solvents such as ethanol, methanol, butanol, n-propyl alcohol, phenol, ethylene glycol, tridecanol, and glycerin. When a surfactant is used as a lubricant in the subsequent wiredrawing step, it is preferable to use the same type of surfactant as in the wiredrawing step. Furthermore, the concentration of the surfactant used in the scratch prevention treatment step is not particularly limited, but in terms of making it easier to adjust the proof stress ratio to more than 1.0 and not more than 2.3, when a surfactant is used in the subsequent wiredrawing step, it is preferable that the surfactant have the same concentration as the surfactant in the wiredrawing step.

[0076] The wire obtained through the above process is drawn into an intermediate wire with a diameter of 5.0 mm. The diameter of the intermediate wire is typically approximately 7 to 330 times the final wire diameter. Next, the drawn wiring material (intermediate wire) may be subjected to a homogenization process (homogenization process). In the homogenization process, the intermediate wire is heat-treated in an atmospheric furnace at 400 to 500°C for approximately 30 to 240 minutes, and then rapidly cooled. By undergoing the homogenization process, it becomes easier to achieve both the desired orientation ratio of the <112> crystal orientation and the load stress ratio, particularly when the amount of elements other than aluminum in the aluminum wiring material is small. The rapid cooling method in the homogenization process is not particularly limited and can be performed by immersing the heated intermediate wire in cold water stored in an immersion tank or by showering the heated intermediate wire with cold water. Homogenization treatment, performed at a lower temperature and for a shorter time than solution treatment, allows elements other than aluminum to be uniformly dispersed in the aluminum matrix, making it easier to achieve both the desired crystal orientation <112> and the load stress ratio. This improves the long-term reliability and vibration resistance of the joint while also providing sufficient bending compliance. In addition to the homogenization treatment, the manufacturing method for the aluminum wiring material of this embodiment may also include solution treatment to dissolve elements other than aluminum in the aluminum matrix. The intermediate wire after homogenization treatment may also be subjected to intermediate heat treatment before wiredrawing. For example, the intermediate heat treatment involves heating the intermediate wire after homogenization treatment at 240 to 300°C for approximately 30 to 140 minutes, followed by air cooling. The intermediate heat treatment allows for fine adjustment of the <112> orientation ratio, load stress ratio, and yield strength ratio. The intermediate wire rod is heated in the intermediate heat treatment by passing it through a heated atmosphere heated to a predetermined temperature, or by a batch type heat treatment by heating the wire rod in a closed furnace.

[0077] The intermediate wire is subjected to homogenization treatment, solution treatment, and intermediate heat treatment as necessary, and then wiredrawn to the final wire diameter. In the wiredrawing process, the wiring material is passed through multiple carbide or diamond dies in sequence to gradually reduce the wire diameter of the wiring material. To achieve a <112> orientation ratio of 45% or less, the area reduction rate (processing rate) per die is preferably 5% to 40%. Furthermore, the wiredrawing process is preferably performed while cooling the dies. Cooling the dies prevents the temperature of the wiring material from increasing due to friction between the dies and the wiring material, thereby making it easier to adjust the <112> orientation ratio of the aluminum wiring material to 45% or less and achieving a uniform <112> orientation ratio throughout the aluminum wiring material. An effective die cooling method is to inject a coolant containing cold water into the die inlet. Simultaneously with die cooling, a lubricant primarily composed of a surfactant may be introduced into the die, which facilitates adjusting the load stress ratio within the above range. According to the method of cooling the die by injecting a coolant into the die inlet, the coolant can be mixed with a lubricant before supplying it, allowing the lubricant to efficiently flow into the die. Regarding die cooling, for example, when the coolant temperature is 20°C or less and the flow rate of the sprayed coolant is high, the cooling effect is high and the <112> orientation ratio tends to be uniform. On the other hand, if the flow rate of the coolant is too high, the cooling effect is limited and waste is likely to occur.

[0078] The wiring material drawn to the final wire diameter is subjected to a final heat treatment, which mainly removes distortions in the metal structure remaining inside the wiring material, thereby adjusting the mechanical properties of the wiring material.

[0079] Final heat treatment methods include running heat treatment, in which the wiring material is passed through a heated atmosphere heated to a predetermined temperature, and batch heat treatment, in which the wiring material is heated in a sealed furnace. In this embodiment, the final heat treatment is preferably performed as a batch heat treatment at 150°C to 400°C for approximately 30 minutes to 60 minutes. By performing the final heat treatment under the above conditions, the ratio of the number of precipitated particles having an aspect ratio of 4 or more can be adjusted to 0.5 to 5. Specifically, the higher the final heat treatment temperature and the longer the final heat treatment time, the larger the ratio of the number of precipitated particles having an aspect ratio of 4 or more. Meanwhile, the lower the final heat treatment temperature and the shorter the final heat treatment time, the smaller the ratio of the number of precipitated particles having an aspect ratio of 4 or more.

[0080] The wiring material subjected to the final heat treatment is preferably subsequently air-cooled. Air-cooling facilitates uniformly adjusting the <112> orientation ratio of the aluminum wiring material to 45% or less. This is because, although water-cooling has a greater cooling effect than air-cooling, air-cooling does not cause contamination or defects on the surface of the aluminum wiring material, and is easier to uniformly and precisely control the <112> orientation ratio to 45% or less while achieving the above-mentioned yield strength ratio and load stress ratio. Air-cooling methods include, for example, a method in which the aluminum wiring material after the final heat treatment is left at room temperature to slowly cool, and a method in which cold air is sprayed onto the surface of the aluminum wiring material. Among these, a method in which air or cold air is sprayed onto the surface of the aluminum wiring material is preferred because it facilitates uniformly and stably adjusting the <112> orientation ratio to the desired range. When air-cooling is performed by spraying air or cold air onto the surface of the aluminum wiring material, the lower the temperature of the air or cold air and the higher the spray speed, the greater the cooling effect, making it easier to stably obtain a <112> orientation ratio within the above-mentioned range.

[0081] In the manufacturing method of the aluminum wiring material of the embodiment described above, by adjusting the conditions for the inclusion removal process in the molten metal, the homogenization process in the original wire, the die cooling in the wire drawing process, the final heat treatment, and the air cooling after the final heat treatment using the above-mentioned techniques depending on the amount and type of elements other than aluminum in the aluminum wiring material, it is possible to adjust the <112> orientation ratio, average crystal grain size, load stress ratio, proof stress ratio, and ratio of the number of precipitated particles having an aspect ratio of 4 or more within preferred ranges.

[0082] (Uses of Aluminum Wiring Material) The aluminum wiring material of this embodiment can be used in all applications in which wiring materials made of iron-based materials, copper-based materials, and aluminum-based materials have conventionally been used. Specifically, the aluminum wiring material of this embodiment can be suitably used as conductive materials for power semiconductors, conductive materials such as electric wires and cables, battery materials, windings used in generators and motors, etc. Conductive materials for power semiconductors include semiconductor bonding wires and bonding ribbons, etc. Battery materials include meshes and nets for current collectors, and bonding wiring materials for connecting bus bars to batteries, etc.

[0083] Next, examples will be described, but the present invention is not limited to the following examples.

[0084] The aluminum wiring materials of the examples were prepared as follows. High-purity aluminum ingots with a purity of 99.9% by mass or higher were prepared, and iron, silicon, gallium, and vanadium were added to produce aluminum alloys with the compositions shown in Table 2 for the wiring materials of each example. The aluminum alloys of each composition were melted under atmospheric pressure. To melt the aluminum alloys, argon (Ar) gas was bubbled into the molten aluminum alloy through a steel pipe, and inclusions precipitated on the surface of the molten alloy were filtered using a ceramic foam filter with a porosity of 0.85. The flow rate of the argon (Ar) gas was 7 L / min, and the molten alloy flux during filtration was 22 kg / min. Subsequently, wire was obtained by continuous casting and extrusion molding. The surface of the wire obtained by extrusion molding was coated with an aqueous solution of ethanol, a surfactant, for scratch prevention. The resulting wire was then drawn to an intermediate wire diameter of 5 mm. Thereafter, this intermediate wire during wiredrawing was subjected to intermediate heat treatment at 240°C to 300°C for approximately 30 to 140 minutes. The intermediate wire after the intermediate heat treatment was cooled in air. The intermediate wire before and after the intermediate heat treatment were subjected to cold wiredrawing by die drawing. In the wiredrawing, the area reduction rate (processing rate) per diamond die was 5% to 40%. In addition, in the wiredrawing by die drawing, a lubricating liquid containing the same surfactant as the surfactant described above at the same concentration as the treatment liquid used in the scratch prevention treatment was poured into the diamond die to cool the diamond die. In the production of the aluminum wiring material of the example, the total processing rate up to the final wire diameter was 99.36%, and the wire was drawn to the final wire diameter (400 μm). Furthermore, the aluminum wiring material processed to the final wire diameter was subjected to a final heat treatment in a batch furnace or a salt bath at a temperature between 150°C and 400°C for 30 to 60 minutes. After the final heat treatment, the aluminum wiring material was air-cooled by spraying cold air onto it, and then the air-cooled aluminum wiring material was rewound onto a spool by a rewinding machine every 300 m. This resulted in an aluminum wiring material with a circular cross section.

[0085] The aluminum wiring materials of the comparative examples were produced as follows: In the same manner as in the examples, aluminum ingot with a purity of 99.9% by mass or more was prepared, and Fe, Si, Ga, V, etc. were added so as to obtain the composition shown in Table 2. The aluminum wiring materials of the comparative examples were produced by changing the manufacturing conditions, such as the heat treatment temperature and time at the intermediate wire diameter and the final wire diameter, the processing rate from each wire diameter to the next wire diameter, the cooling rate after the intermediate heat treatment, and the area reduction rate of each die, outside the ranges of the above examples.

[0086] (Measurement of Crystal Orientation and Average Crystal Grain Size of Wiring Material Cross Section) The crystal orientation of the cross section of the aluminum wiring material in the examples and comparative examples was measured as follows. Aluminum wiring material was cut into lengths of several centimeters to prepare multiple evaluation samples. The evaluation samples were attached straight and flat to a metal (Ag-plated frame) plate, taking care to prevent stretching or sagging. The evaluation sample and the metal plate were then placed in a cylindrical mold with the metal plate at the bottom of the cylinder. An embedding resin was then poured into the mold, and a curing agent was then added to harden the resin. The cylindrical resin containing the hardened evaluation sample was then roughly polished with a polisher to expose a vertical cross section (hereinafter also referred to as a transverse section) of the wiring material in the longitudinal direction. The cut surface was then polished to a final finish, followed by ion milling to remove residual strain on the polished surface and obtain a smooth surface. The ion milling device was fine-tuned so that the cut surface of the wiring material was perpendicular to the longitudinal direction of the wiring material.

[0087] The cross section of the wiring material of the evaluation sample (i.e., the polished surface of the evaluation sample) was attached to the sample stage of a field emission scanning electron microscope (FE-SEM, JEOL JSM-7800F) so that it was parallel to the sample stage, and the crystal orientation and average crystal grain size of the cross section were measured using the FE-SEM at a magnification of 200x, an acceleration voltage of 15 keV, a measurement area of ​​approximately 420 x 420 μm, and a measurement interval (Step Size) of 1.2 μm. The crystal orientation thus obtained was analyzed using dedicated software for crystal orientation analysis (OIM analysis, manufactured by TSL, hereinafter also referred to as dedicated software).

[0088] The crystal grain discrimination conditions of the dedicated software were set, and the crystal orientation ratio of the measured sample, i.e., the proportion of each crystal orientation, was analyzed. The crystal grain discrimination conditions of the dedicated software can usually be set depending on the purpose of the analysis, but since this has almost no effect on the analysis results of the crystal orientation ratio, in this example and comparative example, the conditions were set so that when the orientation difference is 10° or less and two or more pixels are connected, it is recognized as one crystal grain, and the proportion of crystal grains with the <112> orientation among them was calculated.

[0089] Furthermore, in analyzing EBSP measurement data, there may be regions where the crystal orientation cannot be measured due to the roughness of the measurement surface, residual strain due to polishing, contamination, the presence of an oxide film, etc. For this reason, a reliability level was set, and using this reliability level as a reference, the area of ​​only the identified crystal orientations within the measurement area was used as a population, and the orientation ratios were automatically calculated using dedicated software. In other words, the orientation ratios were calculated by excluding areas where the crystal orientation could not be measured and areas where the crystal orientation could be measured but the reliability of the measurement results was low. Here, parameters may be provided in the dedicated software, and the reliability can be set using these parameters. Specifically, several parameters such as the Confidential Index (CI value) and Image Quality (IQ value) can be used to select a criterion for determining whether the reliability is met depending on the condition of the sample, the purpose of the analysis, etc. For example, the CI value was set to 0.1 or more, and areas with a CI value of less than 0.1 were excluded. Furthermore, using the cleanup process included in the dedicated software, orientation data of pixels with low measurement reliability was replaced with data from surrounding pixels with normal measurements, thereby complementing the measurement results of areas with low measurement reliability. This method is effective for removing scattered pixels with low measurement reliability. However, excessive cleanup processing increases noise in the projected image. Therefore, for example, when the orientation difference is 10° or less and two or more pixels are connected, a single crystal grain is recognized. The grain dilation method was performed once, and the grain CI standardization method was also performed once.

[0090] The <112> orientation ratios with an angle difference of 10° or less with respect to the longitudinal direction and the average crystal grain sizes in the cross sections perpendicular to the longitudinal direction of each of the aluminum wiring materials of the Examples and Comparative Examples, determined as described above, are shown in Table 2 below.

[0091] (Measurement of aspect ratio and aspect ratio distribution of precipitated particles) A ​​sample was obtained using the above-described sampling method for a wire after thermal refining treatment with a final wire diameter of 400 μm. A cross section (transverse section) perpendicular to the wire axis was observed for the particle size of the precipitated particles using an FE-SEM (field emission scanning electron microscope, JEOL JSM-7800F) at a magnification of 400x to observe precipitated particles over a wide range of the cross section, and then 1 to 3 average locations of the particle distribution were observed at a magnification of 1,000x. The SEM imaging conditions were an acceleration voltage of 5 kV, a working distance (WD) of 10 mm, and a backscattered electron image (BED-C) was selected. In image analysis, the brightness value of the captured SEM image was normalized to a range of 0 to 1, and then binarized using a threshold of 0.95. Areas with brightness values ​​higher than the threshold were identified as precipitated particles.

[0092] In addition, in the image analysis, adjacent pixels in eight neighborhoods of an area on an image recognized as a particle were determined as one particle. The eight neighborhoods refer to eight directions, up, down, left, right, and 45° rotation from a predetermined area recognized as a particle, and a contact area in any of these eight directions is defined as one particle. Note that if only the predetermined area is recognized as a particle and there is no contact area in any of the eight directions, it is not determined to be a particle.

[0093] Next, the aspect ratio (the ratio of the major axis to the minor axis (major axis / minor axis) when the shape of the precipitated particle is approximated as an ellipse circumscribing the precipitated particle) and number of precipitated particles in the entire cross section perpendicular to the wire axis direction were measured, and the ratio of the number of precipitated particles with an aspect ratio of 4 or more to the total number of precipitated particles was expressed as a percentage. This ratio was almost the same at the tip, rear end, and middle part of the wire. The ratio of the number in each example is shown in Table 2. The ratio of the total number of precipitated particles to the number of precipitated particles with an aspect ratio of 4 or more for Example 3 and Comparative Example 9 is shown in Table 4.

[0094] (Measurement of Load Stress Ratio) The load stress ratio was determined by the following method. Each aluminum wiring material of the Examples and Comparative Examples was used to obtain an evaluation sample with a measurement length of 100 mm. Using this evaluation sample, a tensile fatigue test was performed using a dedicated tensile fatigue testing machine (ElectroPlus E3000, manufactured by Instron) at room temperature of 15 to 28°C to measure the fatigue limit (MPa). Both ends of the evaluation sample were pulled, and loading at the maximum load stress and unloading at the minimum load stress were performed at a rate of 5 Hz (5 times per second) with a maximum number of repetitions exceeding 107. The maximum load stress at three or more different load repetitions was plotted on a semi-logarithmic graph of an S-N curve, and an approximate line was obtained from the plotted points using the least squares method. The fatigue limit (MPa) after 107 repetitions was determined using this approximate line.

[0095] In addition, each aluminum wiring material in the examples and comparative examples was cut to a length slightly longer than 100 mm to obtain a separate evaluation sample. Tensile strength, i.e., maximum load (kgf), was measured by a tensile test at room temperature of 15 to 28°C. The maximum load in the tensile test can be measured using a tensile testing device. The maximum load was calculated by continuously pulling the aluminum wiring material of the evaluation sample at a speed of 20 mm / min and a load cell rating of 100 N with a measurement length of 100 mm using a tensile testing device (e.g., an autograph manufactured by Shimadzu Corporation, model AGS-5kNX), and the maximum load was calculated as the maximum value until breakage. The maximum load is the force applied to the wiring material pulled at the above speed, and is usually calculated automatically by converting the pulling force into an electrical signal using a load cell. Taking into account the variation in measurement results, the maximum load was calculated as the average value of three pieces. The maximum load was divided by the cross-sectional area of ​​the wiring material to determine the tensile strength (MPa).

[0096] The load stress ratio was calculated from the fatigue limit and tensile strength measured as described above using the following formula: Load stress ratio = fatigue limit / tensile strength The load stress ratios of the aluminum wiring materials of the examples and comparative examples calculated as described above are shown in Table 2 below.

[0097] (Measurement of Yield Strength Ratio) Each aluminum wiring material of the Examples and Comparative Examples was cut to a length slightly longer than 100 mm to obtain an evaluation sample for measuring the yield strength ratio. A tensile test was performed under the same conditions as described above to determine the maximum load (kgf) and 0.2% strain load (kgf) of each sample, which were then divided by the cross-sectional area of ​​each sample to calculate the yield strength ratio using the following formula: Yield Strength Ratio = Maximum Stress / 0.2% Yield Strength (Stress) The yield strength ratios of each aluminum wiring material of the Examples and Comparative Examples obtained as described above are shown in Table 2 below.

[0098] (Measurement of Element Concentration) The element concentrations in each aluminum wiring material in the Examples and Comparative Examples were measured as follows. Approximately 0.5 g of the manufactured wiring material was dissolved. The concentration of each element in the solution was determined by high-frequency inductively coupled plasma atomic emission spectroscopy (Shimadzu Corporation, ICPE-9000). The elements measured here other than aluminum were iron (Fe), silicon (Si), gallium (Ga), vanadium (V), magnesium (Mg), copper (Cu), nickel (Ni), zinc (Zn), chromium (Cr), manganese (Mn), titanium (Ti), zirconium (Zr), tungsten (W), and scandium (Sc), and their contents are shown in Table 2 below. In the table below, the abbreviation "mass" means mass.

[0099] The compositions of the aluminum wiring materials of the Examples and Comparative Examples obtained above are shown in Table 2. Next, the following performance evaluations were carried out on the aluminum wiring materials obtained above.

[0100] (Power Cycle Test) Each of the aluminum wiring materials of the examples and comparative examples was bonded to the aluminum alloy electrodes of a power chip using an ultrasonic bonding device (a REBO7 wire bonder manufactured by Ultrasonic Industries). For each aluminum wiring material, the ultrasonic energy and pressure during bonding were set so that the longitudinal length of the wiring material at the bonded portion was 500 μm. After bonding the aluminum wiring material, the power chip was subjected to a current, current flow time, and cooling time set so that the maximum temperature (Tjmax) of the power chip was 150°C and the minimum temperature (Tjmin) was 25°C, i.e., the temperature difference between the maximum and minimum temperatures (ΔTj) was 125°C. A power cycle test was performed using the current flow time of approximately 7 seconds, the current stop time of approximately 13 seconds, and one cycle of approximately 20 seconds.

[0101] The number of cycles at which electricity could no longer be passed was defined as the life in the power cycle test (power cycle life). Samples with a power cycle life of 60,000 cycles or more were judged to have a life exceeding the target and were therefore rated as "S" (excellent). A power cycle life of 40,000 cycles or more but less than 60,000 cycles was judged to be at the target level and therefore rated as "A" (good). A power cycle life of 20,000 cycles or more but less than 40,000 cycles was judged to have passed the test and was therefore rated as "B" (passable). A sample with a power cycle life of less than 20,000 cycles was judged to have failed the test and was therefore rated as "C" (fail). The evaluation of the power cycle test (thermal cycle) of the aluminum wiring materials in each of the examples and comparative examples is shown in Table 3.

[0102] (Vibration Resistance) Each aluminum wiring material of the examples and comparative examples was bonded to two aluminum plates at the first and second bonding portions using an ultrasonic bonding device (REBO7 wire bonder manufactured by Ultrasonic Industries) so that the loop height was 1 mm and the length was 7 mm, thereby obtaining an evaluation sample. The loop length was the linear distance between both ends of the first and second bonding portions. Using this evaluation sample, a vibration fatigue test was performed using a dedicated vibration fatigue testing machine (CV-101M manufactured by AS ONE Corporation) manufactured in accordance with JIS standards, and vibration resistance was evaluated. For the vibration fatigue test, vibration was applied to each evaluation sample at a frequency of 50 Hz, an acceleration of 1 G, and an amplitude of 0.099 mm until fracture. In the vibration fatigue test, evaluation samples that exceeded 50,000 total vibration cycles until fracture were rated as "S," meaning they met the target. Samples that exceeded 10,000 total vibration cycles until fracture but were less than 50,000 total vibration cycles were rated as "A," meaning they met the target level. Samples that exceeded 5,000 total vibration cycles until fracture but were less than 10,000 total vibration cycles were rated as "B," meaning they met the target level. Samples in which the total number of vibrations until fracture was less than 5,000 failed were marked with a "C." Table 3 shows the evaluation of the vibration fatigue test for the aluminum wiring materials in each of the examples and comparative examples.

[0103] (Tool Disengagement Evaluation) Next, an evaluation sample of aluminum wiring material with a wire diameter of 400 μm was used, and an ultrasonic bonding device (K&S ASTERION wire bonder) was used to bond two aluminum plates with a first joint and a second joint, with a distance of 5 mm between them. The evaluation sample was bent horizontally to the axial direction of the wiring material, aiming for a 45° angle, and the second joint was bonded. The bonding conditions were set so that the ultrasonic energy and pressure were optimal for the evaluation sample. The bond tool used was a Kulicke & Soffa model number 127591-16, and the dimensions of the alligator jaws that grip the wiring material were a width (inner diameter) of 0.5 mm, a depth (height) of 0.2 mm, and a length (depth) of 1.0 mm.

[0104] The occurrence of a tool disconnection defect was determined by observing the condition of the wiring material at the second bond. Using an evaluation sample, 30 bonding runs (a combination of the first and second bonding runs counted as one run) were performed, and if the wiring material was not bonded or if there was even one contact mark with the tool due to partial contact, as shown in the lower left of Figure 3, it was judged as a failure and marked with an "X." If the wiring material was properly bonded, as shown in the lower right of Figure 3, it was judged as a pass and marked with an "O." These results are shown in Table 3.

[0105] The wiring materials of Example 1 and Comparative Example 11 were subjected to a corrosion resistance test in a high humidity environment as follows. Using an ultra-accelerated life tester, the test was conducted for up to 1,000 hours under conditions of 121°C and 100% RH (saturation). The corrosion layer thickness was measured by preparing a wire cross section using a cross-section milling device, and then observing the corrosion layer using an FE-SEM. As a result, the corrosion layer thickness was found to be approximately 5 μm in all cases, a level that was confirmed to be acceptable for use as a wiring material.

[0106] (Overall Evaluation) If the above three evaluations were only "S" and "○", the overall evaluation was judged to be "Excellent", meaning excellent. If the above three evaluations were only "S", "A" or "○", the overall evaluation was judged to be "Good", meaning good. If there was any other combination of evaluations and there was no evaluation of "C" or "X", the overall evaluation was judged to be "Fair", meaning passing the test. If there was even one evaluation of "C" or "X", the overall evaluation was judged to be "Fail", meaning failing the test. These results are shown in Table 3.

[0107] (Power Cycle and Vibration Fatigue Tests of Resin-Sealed Chips) A simulated power semiconductor device for evaluation was fabricated to evaluate the power cycle life and vibration fatigue of a resin-sealed chip. FIG. 4 shows a schematic diagram of the simulated power semiconductor device 200 for evaluation. As shown in FIG. 4 , two stepped aluminum terminals 202 and 203 and an IGBT chip 204 were die-bonded to a tungsten carbide metal plate 201 measuring 45 mm in length, 40 mm in width, and 2 mm in thickness. A metal spacer was placed on the back surface of one of the aluminum terminals 203 to provide an appropriate protrusion. The IGBT chip 204 and terminal 203 were then bonded using aluminum wiring materials of the example and comparative examples. A loop was then formed so that the height from the top of the loop to the surface of the terminal 203 was 1 mm, the height from the surface of the IGBT chip 204 to the top of the loop was 10 mm, and the distance between both end faces of the wiring material joint, i.e., the length of the wiring material, was 7 mm at the shortest and 20 mm at the longest. Thereafter, a wall was formed around the IGBT chip using a commercially available resin dam material, and the inside of the wall, the IGBT chip and the bonding area for the wiring material, were filled with a silicone gel-based sealing resin.

[0108] The simulated power semiconductor device fabricated by the above method was subjected to a power cycle test and a vibration fatigue test as thermal vibration fatigue tests. In other words, in order to reproduce loads similar to those caused by thermal stress and vibration in an actual device, one cycle of the power cycle test (4,320 cycles of 20 seconds x 4,320 times = 24 hours) and a subsequent vibration test (24 hours) of the semiconductor device itself were counted as one cycle, and a current test was performed for each cycle. The vibration test was conducted under the conditions of the vibration fatigue test described above.

[0109] The number of cycles at which current could no longer be passed was defined as the life in the test, and samples with a life of 12 cycles or more were rated "excellent," meaning that they exceeded the target life, samples with a life of 8 cycles or more but less than 12 cycles were rated "good," meaning that they met the target level, samples with a life of 4 cycles or more but less than 8 cycles were rated "passable," and samples with less than 4 cycles were rated "fail," meaning that they failed the test. Table 5 shows the evaluation of the thermal vibration fatigue test for the aluminum wiring materials in each example and comparative example. Note that Comparative Examples 9, 10, and 11 failed the tool removal evaluation and were not evaluated. The results are shown in Table 5. Note that in Table 5, some of the compositions of the wiring materials are omitted.

[0110] (Evaluation of Rolled Wiring Materials) Furthermore, samples prepared under the same processing conditions as in Example 1 using aluminum having the composition shown in Table 6 were rolled into ribbons with long sides of 2 mm and short sides of 0.2 mm. The resulting ribbon-shaped wiring materials were loop-bonded between two aluminum plates in the same manner as in Example 1. The ribbon-shaped wiring materials were bonded and evaluated under the same conditions as in Example 1, except that the second bonding step was performed with the aim of bending the wiring materials horizontally at an angle of 20° relative to the wiring material axial direction. The results are shown in Table 6. Furthermore, aluminum wiring materials having the composition shown in Table 7 were prepared in the same manner as in Example 1, and the resulting samples were rolled into ribbons with long sides of 2 mm and short sides of 0.2 mm. The resulting ribbon-shaped wiring materials were subjected to the "thermal vibration fatigue test of resin-encapsulated chips" described above. The results are shown in Table 7.

[0111]

[0112]

[0113]

[0114]

[0115]

[0116]

[0117] From the above, the aluminum wiring material of the example can simultaneously achieve improved vibration resistance, improved heat resistance, and improved followability. The aluminum wiring material for power semiconductors of the example can greatly contribute to the development of the power electronics industry, the automobile industry, the electric railway industry, the electric power industry, etc.

Claims

1. An aluminum wiring material made of aluminum with a purity of less than 99.9% by mass, wherein in a cross section perpendicular to the longitudinal direction of the aluminum wiring material, the orientation ratio of the crystal orientation <112> having an angle difference of 10° or less with respect to the longitudinal direction is 45% or less, the ratio of the number of precipitated particles having an aspect ratio of 4 or more to the total number of precipitated particles in the cross section is 0.5% to 16%, and the load stress ratio defined in the following (1) is 0.4 to 0.9, where: load stress ratio = fatigue limit / tensile strength (1); and the yield strength ratio defined in the following (2) is greater than 1.0 and less than 2.3, where: yield strength ratio = maximum stress / 0.2% yield strength (2) The aluminum wiring material is characterized in that it contains at least one element of iron (Fe) and silicon (Si) in a total amount of 0.01 mass% or more and 1 mass% or less relative to the total amount of the aluminum wiring material, and contains at least one element of gallium (Ga) and vanadium (V) in a total amount of 50 mass ppm or more and 800 mass ppm or less.

2. The aluminum wiring material according to claim 1, characterized in that the average crystal grain size in a cross section perpendicular to the longitudinal direction of the aluminum wiring material is 40 μm or less.

3. The aluminum wiring material according to claim 1 or 2, further comprising one or more elements selected from magnesium (Mg), copper (Cu), nickel (Ni), zinc (Zn), chromium (Cr), manganese (Mn), titanium (Ti), zirconium (Zr), tungsten (W) and scandium (Sc).

4. The aluminum wiring material according to claim 1 or 2, wherein the wire diameter of the aluminum wiring material is 15 μm or more and 700 μm or less.

5. The aluminum wiring material according to claim 1 or 2, wherein a cross section perpendicular to the longitudinal direction of the aluminum wiring material is circular, oval, elliptical or flattened elliptical.

6. The aluminum wiring material according to claim 1 or 2, wherein the aluminum content in the aluminum wiring material is 99 mass % or more and less than 99.9 mass % with respect to the total amount of the aluminum wiring material.

7. A method for manufacturing an aluminum wiring material, comprising the steps of: preparing an aluminum alloy having a purity of less than 99.9% by mass, the aluminum alloy containing at least one of iron (Fe) and silicon (Si) in a total amount of 0.01% by mass or more and 1% by mass or less, and at least one of gallium (Ga) and vanadium (V) in a total amount of 50 ppm by mass or more and 800 ppm by mass or less; and drawing the aluminum alloy, wherein in a cross section perpendicular to the longitudinal direction of the aluminum wiring material, an orientation ratio of a crystal orientation <112> having an angle difference of 10° or less with respect to the longitudinal direction of the aluminum wiring material is 45% or less; a ratio of the number of precipitated particles having an aspect ratio of 4 or more to the total number of precipitated particles of the aluminum wiring material is 0.5% or more and 16% or less; a load stress ratio defined in the following (1) is 0.4 to 0.9, and load stress ratio = fatigue limit / tensile strength (1) (2) A method for producing an aluminum wiring material, wherein a yield strength ratio as defined in the following (2) is greater than 1.0 and not greater than 2.3, Yield strength ratio=maximum stress / 0.2% yield strength.

8. A method for producing an aluminum wiring material, comprising the steps of: preparing an aluminum alloy having a purity of less than 99.9% by mass, the aluminum alloy containing at least one of iron (Fe) and silicon (Si) in a total amount of 0.01% by mass or more and 1% by mass or less, and at least one of gallium (Ga) and vanadium (V) in a total amount of 50 ppm by mass or more and 800 ppm by mass or less; and drawing the aluminum alloy under conditions of an area reduction rate (working rate) per die of 5% to 40%, wherein in a cross section perpendicular to the longitudinal direction of the aluminum wiring material, an orientation ratio of a crystal orientation <112> having an angle difference of 10° or less with respect to the longitudinal direction of the aluminum wiring material is 45% or less; and a ratio of the number of precipitated particles having an aspect ratio of 4 or more to the total number of precipitated particles of the aluminum wiring material is 0.5% or more and 16% or less; A method for producing an aluminum wiring material, comprising: a load stress ratio defined in (1) below is 0.4 or more and 0.9 or less, Load stress ratio = Fatigue limit / Tensile strength (1); a yield strength ratio defined in (2) below is more than 1.0 and 2.3 or less, Yield strength ratio = Maximum stress / 0.2% yield strength (2).

Images