Copper powder

Abstract

The present invention provides copper powder that can be suitably used in copper paste. [Solution] The copper powder of this invention has an average particle size D50 based on the number of particles obtained by image analysis of SEM images. SEM and the standard deviation of particle size σ D Therefore, the formula is: CV = 100 × σ D / D50 SEM The coefficient of variation CV calculated is 30% or less, and the 90% particle size D90 obtained by the laser diffraction-scattering method is 30% or less. Laser However, it is less than 0.86 μm.

Description

[Technical Field]

[0001] This invention relates to copper powder. [Background technology]

[0002] Copper powder can be mixed with an organic binder to form a copper paste, which is sometimes used in the manufacture of electronic components such as wiring or terminals for low-temperature co-fired ceramic (LTCC) substrates, or as electrode materials for multilayer ceramic chip capacitors (MLCCs).

[0003] Multilayer ceramic chip capacitors have a structure in which dielectric layers and internal electrode layers are alternately stacked, with external electrodes provided at both ends. Here, the dielectric layers are mainly composed of ceramics with high dielectric constants, such as barium titanate. On the other hand, various metal or alloy powders can be used for the internal electrode layers. Copper powder has high conductivity and is considered a promising metal powder for use in internal electrode layers because it has the potential to enable thinning or miniaturization of the internal electrode layer of multilayer ceramic chip capacitors and improve frequency characteristics.

[0004] Conventional technologies related to copper powder include those described in Patent Documents 1 and 2, for example.

[0005] Patent Document 1 states that "the volume cumulative particle size D50 measured by a laser diffraction scattering particle size distribution analyzer is 0.20 μm to 0.70 μm, the ratio of crystallite size to D50 (crystallite size / D50) is 0.15 to 0.60 (μm / μm), and the ratio of oxygen content (O content) to specific surface area is 0.10 to 0.40 (wt%·g / m²)." 2 The text describes "copper powder characterized by being..."

[0006] Patent Document 2 describes a copper powder containing copper particles, with a compacted bulk density of 1.30 g / cm³. 3 ~2.96 g / cm³ 3It is described as "copper powder" in which the 50% particle size D50 when the cumulative frequency becomes 50% in the volume-based particle size histogram of copper particles and the crystallite size D obtained using Scherrer's formula from the diffraction peak of the Cu(111) plane in the X-ray diffraction profile obtained by the powder X-ray diffraction method for the copper powder satisfy D / D50 ≥ 0.060.

Prior Art Documents

Patent Documents

[0007]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0008] By the way, in the production of the above-described multilayer ceramic chip capacitor, etc., when the copper powder is applied as a copper paste, it may be required that the surface of the coating film becomes smooth. Conventional copper powders have room for further improvement from the viewpoint of obtaining a sufficiently smooth surface in the coating film after the application of the copper paste.

[0009] An object of this invention is to provide copper powder that can be suitably used for a copper paste.

Means for Solving the Problems

[0010] The copper powder of this invention has an average particle size D50 based on the number obtained by the image analysis method of the SEM image SEM and a standard deviation σ of the particle size D From the formula: CV = 100×σ D / D50 SEM the coefficient of variation CV calculated thereby is 30% or less, and the volume-based 90% particle size D90 obtained by the laser diffraction / scattering method Laser is 0.86 μm or less.

[0011] The above copper powder has a volume-based 99% particle size D99 obtained by the laser diffraction / scattering method Laser preferably being 1.35 μm or less.

[0012] The above copper powder preferably has a cumulative ratio of the volume distribution of particles with a particle size of less than 1 μm obtained by the laser diffraction / scattering method of 95% or more.

[0013] The above copper powder has a volume-based 50% particle size D50 obtained by the laser diffraction / scattering method Laser preferably being 0.52 μm or less.

[0014] The above copper powder preferably has a BET specific surface area of 2.7 m 2 / g or more.

Advantages of the Invention

[0015] The copper powder of this invention can be suitably used in copper paste.

Modes for Carrying Out the Invention

[0016] Hereinafter, embodiments of the copper powder of this invention will be described in detail. The copper powder of one embodiment of this invention has a coefficient of variation CV of 30% or less and a 90% particle size D90 Laser being 0.86 μm or less.

[0017] Here, the above coefficient of variation CV is the number-based average particle size D50 obtained by an image analysis method for an image of a scanning electron microscope (SEM) SEM and the standard deviation σ of the particle size D From the formula: CV = 100×σ D / D50 SEMThis is calculated as follows. The smaller the coefficient of variation CV, the more uniform the particle size and the more monodisperse the particles are. This is thought to contribute to improving the surface smoothness of the coating film formed by applying copper paste. On the other hand, since the image analysis method mainly measures the particle size of primary particles, it is not possible to evaluate whether or not aggregation occurs. It is presumed that large aggregated particles in copper powder reduce the smoothness of the coating film surface. In contrast, the copper powder of this embodiment further has a volume-based 90% particle size D90 obtained by laser diffraction and scattering method. Laser The particle size is 0.86 μm or less. In laser diffraction / scattering, the powder is dispersed in a solvent for measurement, but it is not possible to completely disperse the primary particles. Therefore, it is an effective method for understanding the tendency of aggregated particles. 90% particle size D90 Laser A small value means that there are almost no or very few coarse aggregated particles. The copper powder of this embodiment has a coefficient of variation CV and a 90% particle size D90 Laser Since each of these values ​​is below a predetermined level, it can be said that the particle size distribution is relatively sharp and large aggregation is suppressed. As a result, it is thought that the coating film after application of the copper paste will have a smooth surface. However, this invention is not limited to such theories.

[0018] Such copper powder can be produced, for example, by a gas-phase reduction method or any other method, and then subjected to a predetermined classification process on a raw material powder mainly containing copper particles. Details will be described later.

[0019] (Particle size) The coefficient of variation (CV) of the copper powder particle size is 30% or less, preferably 28% or less, and more preferably 26% or less. If the coefficient of variation (CV) exceeds 30%, the particle size distribution will not be very sharp, and the surface of the copper paste coating will become rough. There are no particular disadvantages to having a coefficient of variation (CV) that is too small, but it is typically 7.5% or more, or even 6% or more.

[0020] The coefficient of variation CV above is obtained from the number-based average particle diameter D50 obtained by image analysis of SEM images. SEMThe standard deviation of particle size σ obtained by a similar image analysis method D This value is calculated by dividing by and expressing it as a percentage. More specifically, a scanning electron microscope (JSM-7500F from JEOL Ltd. or an equivalent device) is used to acquire a 15,000x magnification SEM image containing approximately 500 copper powder particles. By analyzing this SEM image using image analysis software (Macview 4.0 from Mountec Co., Ltd. or an equivalent software), the particle size of each particle contained in the SEM image can be obtained. This then gives the standard deviation σ of the particle size. D This allows us to determine the average particle size D50, where the cumulative distribution based on the number of particles in the resulting particle size distribution graph is 50%. SEM In this case, "particle diameter" refers to the diameter of the smallest circle that encloses the projected image of the particle in the SEM image.

[0021] Also, 90% of copper powder particle size D90 Laser The particle size is 0.86 μm or less, preferably 0.75 μm or less, and more preferably 0.70 μm or less. 90% particle size D90 Laser If the particle size exceeds 0.86 μm, it indicates that a certain amount of coarse aggregated particles are present, which reduces the smoothness of the copper paste coating surface. 90% particle size D90 Laser There are no particular disadvantages to it being too small, but 90% particle size D90 Laser It is preferably 0.45 μm or larger, and more preferably 0.50 μm or larger.

[0022] Copper powder has a particle size of 99% D99. Laser The particle size is preferably 1.35 μm or less, and more preferably 1.16 μm or less. This is expected to significantly reduce the amount of coarse particles and further improve the smoothness of the coating surface. 99% particle size D99 Laser There are no particular disadvantages to it being too small, but 99% particle size D99 Laser Preferably, the particle size is 0.55 μm or larger, and more preferably 0.70 μm or larger.

[0023] Copper powder with a particle size of 50% D50 LaserThe particle size is preferably 0.52 μm or less, and more preferably 0.45 μm or less. Furthermore, 0.25 μm to 0.35 μm is preferable. 50% particle size D50 Laser If the particle size is too small, there is a possibility that many fine particles will aggregate easily, and if it is too large, there is a concern that the coating film's smoothness will deteriorate due to the large amount of coarse particles.

[0024] The above-mentioned 90% particle size D90 Laser , 99% particle size D99 Laser and 50% particle size D50 Laser These terms refer to the particle sizes at which the cumulative distribution based on volume reaches 90%, 99%, or 50%, respectively, in the particle size distribution graph obtained by measurement using the laser diffraction / scattering method based on JIS Z8825 (2013) with a laser diffraction / scattering particle size distribution analyzer (LA-960 laser diffraction / scattering particle size distribution analyzer manufactured by Horiba, Ltd. or an equivalent device).

[0025] It is preferable that the copper powder has a cumulative volume distribution percentage of particles with a particle size of less than 1 μm, as measured by laser diffraction and scattering, of 95% or more, and more preferably 97% or more. In this case, since there are sufficiently few particles with a particle size of 1 μm or more, it is thought that the coating film smoothness after application of the copper paste will be good. The cumulative volume distribution percentage of particles with a particle size of less than 1 μm can be determined, as described above, by calculating the cumulative volume distribution percentage for the particle size range of less than 1 μm when measuring the volume-based particle size distribution using a laser diffraction and scattering particle size distribution analyzer.

[0026] (BET specific surface area) The BET specific surface area of ​​copper powder is 2.7 m². 2 It must be 3.0m or more, and also 3.0m 2 / g~4.3m 2 It is preferable that it be / g.

[0027] If the BET specific surface area is too large, there is a concern that many fine particles will aggregate easily. On the other hand, if the BET specific surface area is too small, there is a risk that many coarse particles will be present, resulting in poor coating smoothness.

[0028] BET specific surface area measurement is performed using the BET method (gas adsorption method), and a fully automatic specific surface area measuring device (Macsorb®) manufactured by Mountec Co., Ltd. can be used.

[0029] (composition) Copper powder primarily contains Cu (copper), with a Cu content of, for example, 98% by mass or more, typically 98.5% by mass or more. Copper powder may contain at least one impurity selected from the group consisting of carbon, oxygen, and chlorine. The Cu content can be obtained by dividing the Cu content by the analytical values ​​of each impurity element described below.

[0030] Copper powder may contain chlorine (Cl). The Cl content of copper powder is preferably 100 ppm by mass or less, and more preferably 50 ppm by mass or less. The Cl content is measured by coulometric titration using a chlorine analyzer TOX2100H (manufactured by Mitsubishi Chemical Corporation).

[0031] Copper powder may contain oxygen (O). The O content of copper powder is preferably 10,000 ppm by mass or less, and more preferably 8,000 ppm by mass or less. The O content is measured by the inert gas fusion-infrared absorption method.

[0032] Copper powder may contain carbon (C). The C content of copper powder is preferably 4000 ppm by mass or less, and more preferably 3000 ppm by mass or less. The C content is measured by combustion-infrared absorption spectroscopy.

[0033] (Manufacturing method) First, prepare a raw material powder mainly containing copper particles. The raw material powder can be a commercially available or pre-made product, or it may be prepared by a gas-phase method such as gas-phase reduction, or by a liquid-phase method. Here, we will explain in detail the preparation of the raw material powder by gas-phase reduction as an example, but this is not the only method.

[0034] In the gas-phase reduction method, a solid copper raw material, mainly consisting of elemental copper, is brought into contact with chlorine gas in a chlorination step to obtain copper chloride gas, and a reduction step is performed in which the copper chloride gas is brought into contact with a reducing gas and reacted. More specifically, copper chloride gas can be generated in the chlorination step and then supplied to the reduction step to come into contact with a reducing gas, thereby continuously reducing the copper chloride gas.

[0035] In the chlorination process, the copper raw material is heated to a temperature below its melting point, for example, 800°C to 1000°C, while chlorine gas is supplied to bring the high-temperature copper raw material into contact with the chlorine gas. At this time, an inert gas for dilution may be supplied along with the chlorine gas to adjust the amount of chlorine that comes into contact with the copper raw material. In this way, the copper in the copper raw material is chlorinated to produce gaseous copper chloride, i.e., copper chloride gas. However, if copper chloride gas is already available or obtained by another method, the chlorination process may be omitted.

[0036] In the reduction process, the copper chloride gas is brought into contact with a reducing gas at a temperature of, for example, 1000°C to 1300°C to reduce the copper chloride in the copper chloride gas to copper. Examples of reducing gases include hydrogen, hydrazine, ammonia, and methane. In addition to the reducing gas, chlorine gas or an inert gas for dilution may also be supplied. When the copper chloride gas comes into contact with the reducing gas, copper atoms are generated at that moment, and ultrafine particles are generated and grow as the copper atoms collide with each other to form copper particles.

[0037] The copper particles obtained in the reduction process are rapidly cooled to a predetermined temperature while blowing in an inert gas such as nitrogen as needed to suppress aggregation. The copper particles are then separated and recovered using a bag filter or the like. This yields a raw material powder consisting of aggregates of copper particles.

[0038] For the above-mentioned raw material powder, it is important to perform a classification process to separate particles of a predetermined particle size from the raw material powder. Furthermore, it is preferable to perform a dispersion process before the classification process to disperse the raw material powder by eliminating any aggregation that may be present in it.

[0039] The dispersion process can be either dry dispersion or wet dispersion, or one of both. Wet dispersion methods include those utilizing rotary shear and those utilizing high-pressure jetting and impact. In the rotary shear method, a slurry containing powder is passed through a narrow gap in the disk while the disk is rotated at high speed, and a shear force is applied to the aggregated particles in the powder, thereby dispersing the powder. In the high-pressure jetting and impact method, the slurry containing powder is ejected at high pressure in opposing directions and collided with other particles to disperse the powder. In manufacturing copper powder according to this embodiment, the high-pressure jetting and impact method is preferred because it can achieve a better dispersion state. Regarding dry dispersion, the method utilizing rotary shear is not preferred; a method in which particles accelerated by high pressure are rotated and collided with each other is preferred.

[0040] When the raw material powder after the dispersion process is subjected to the classification process, the classification is performed with the aggregated particles already removed during the dispersion process. This results in a sharper particle size distribution in the final copper powder, and a sufficient reduction or elimination of aggregated particles. When the dispersion process is wet classification, the classification process can be carried out after washing, drying, and crushing under a predetermined pressure.

[0041] In the classification process, airflow classification can preferably be employed. This allows for the effective separation of particles of the desired size, even with raw material powders that have small particle sizes, such as those targeted here. On the other hand, in the case of wet classification, with raw material powders that have relatively small particle sizes, even if a predetermined dispersant is used during the wet dispersion process, particles of a predetermined particle size may not be sufficiently separated during the subsequent wet classification process, resulting in copper powder of the desired particle size.

[0042] In dry classification using an airflow classifier, for example, when raw material powder is supplied from the hopper of the airflow classifier, the raw material powder is drawn into a swirling airflow within the classification chamber, where it is separated into coarse and fine powder. For example, by collecting and obtaining the fine powder, the material from which the coarse powder has been separated can be obtained.

[0043] After the classification process, copper powder is obtained after sieving as needed. This copper powder tends to have uniform particle size and suppressed aggregation, making it suitable for use in copper paste. [Examples]

[0044] Next, we fabricated a prototype of the copper powder of this invention and confirmed its effects, which are described below. However, this explanation is for illustrative purposes only and is not intended to be an exhaustive list.

[0045] 25 kg of raw material powder, mainly consisting of copper particles, was prepared by a gas-phase reduction method. Copper powder was produced by subjecting this raw material powder to the predetermined dispersion, crushing, and classification conditions shown in Table 1.

[0046] [Table 1]

[0047] For each copper powder obtained as described above, the Cu, Cl, C, and O content, and the average particle size D50 were determined according to the measurement method described earlier. SEM , coefficient of variation CV, 50% particle size D50 Laser , 90% particle size D90 Laser , 99% particle size D99 Laser The cumulative percentage of the volume distribution of particles with a particle diameter of less than 1 μm (<1 μm) and the BET specific surface area were calculated. These results are shown in Table 2.

[0048] [Table 2]

[0049] Furthermore, copper pastes were prepared using each copper powder, and tests were conducted to evaluate the surface of the coating film obtained by applying the copper paste. Specifically, 7 parts by weight of copper powder, 0.3 parts by weight of ethyl cellulose, 2.7 parts by weight of terpineolure, and 0.07 parts by weight of dispersant were weighed and manually mixed, and then processed a total of nine times using a three-roll roller. The paste prepared with the three-roll roller was used to form a coating film using a 15 μm applicator. The coating film was dried in a vacuum drying oven under a nitrogen atmosphere at 150°C for 1 hour. Surface roughness Sa, line roughness Ra, and waviness Sdr of the dried coating film were measured by arbitrarily observing a 1.1 mm square area using a scanning white light interference microscope VS1330 (manufactured by Hitachi High-Tech Corporation). The results are shown in Table 3.

[0050] The criteria for a good evaluation were a surface roughness Sa of 0.07 μm or less, a line roughness Ra of 0.07 μm or less, and a waviness Sdr of 0.15% or less. Regarding surface and line roughness, when forming the electrode film of the internal electrodes of MLCCs, poor smoothness can lead to short circuits between electrodes after lamination. Therefore, it is considered preferable for the surface to be as smooth as possible (no protrusions). Considering the current situation where the thickness of the leading dielectric layer is 1 μm or less due to thinning, it is considered that the ideal thickness is 1 μm or less, preferably 0.7 μm or less, and even more preferably 0.5 μm or less. Furthermore, regarding waviness Sdr, if waviness occurs during lamination, it is feared that voids will be created in the laminated layer, causing voids or increasing the thickness. Therefore, it is required that there be no waviness (smoothness). Based on this, the preferred value for waviness Sdr was set to 0.15 μm or less.

[0051] [Table 3]

[0052] Tables 2 and 3 show that the copper powders of Examples 1 to 4 all have a coefficient of variation CV and a 90% particle size D90. Laser Since each of these values ​​was within the specified range, the evaluation of the coating surface was good. The copper powders in Comparative Examples 1 and 2 had a 90% particle size of D90. LaserDue to the large coefficient of variation (CV), the surface roughness Sa, waviness Sdr, and linear roughness Ra of the coating film were high, resulting in a low evaluation. The copper powder in Comparative Example 3 received a low evaluation due to its high coefficient of variation (CV).

[0053] Based on the above, it is suggested that the copper powder of this invention can be suitably used in copper paste.

Claims

1. It is copper powder, Average particle diameter D50 obtained by image analysis of SEM images based on the number of particles. SEM and the standard deviation of particle size σ D Therefore, the formula is: CV = 100 × σ D / D50 SEM The coefficient of variation CV calculated is 30% or less. Volume-based 90% particle size D90 obtained by laser diffraction and scattering methods Laser However, the copper powder is 0.86 μm or smaller.

2. D99, the 99% particle size relative to volume, obtained by laser diffraction and scattering methods. Laser The copper powder according to claim 1, wherein the particle size is 1.35 μm or less.

3. The copper powder according to claim 1 or 2, wherein the cumulative percentage of the volume distribution of particles with a particle diameter of less than 1 μm obtained by laser diffraction / scattering is 95% or more.

4. Volume-based 50% particle size D50 obtained by laser diffraction / scattering method Laser The copper powder according to claim 1, wherein the particle size is 0.52 μm or less.

5. BET specific surface area is 2.7 m² 2 The copper powder according to claim 1 or 2, wherein the amount is 1 / g or more.

Images