Copper ink, electronic circuit board, and electronic circuit board manufacturing method

Abstract

Provided is a copper ink wherein bleeding is suppressed when the copper ink is printed on a substrate and thus it is possible to manufacture a highly reliable circuit pattern with few short circuits. Also provided is an electronic circuit board and an electronic circuit board manufacturing method in which this copper ink is used. This copper ink includes copper particles, wherein the thixotropic index TI(1 / 10), which is defined as the ratio (η1 / η10) of the viscosity η10 at a shear rate of 10 sec–1 to the viscosity η1 at a shear rate of 1 sec–1, is 7.5 or less, and the tanδ, which is expressed as the ratio (G" / G') of the loss modulus G" and the storage modulus G' at a strain of 0.01%, and the thixotropic index TI(1 / 10) satisfy relational expression (1): tanδ<–0.15×TI(1 / 10)+1.31.

Description

Copper ink, electronic circuit board, and method for manufacturing electronic circuit board

[0001] The present invention relates to a copper ink, an electronic circuit board, and a method for manufacturing an electronic circuit board, and more particularly to a copper ink that can be used in a screen printing method, an electronic circuit board using the same, and a method for manufacturing an electronic circuit board.

[0002] As electronic devices and information terminals become smaller and lighter, the electronic components used inside the devices are becoming smaller and smaller. This has led to a gradual decrease in the size of the wiring patterns inside the electronic components, as well as narrower widths of the wiring patterns and spacing between the wires.

[0003] Optical patterning, which involves exposure and etching processes, is a common method for forming high-resolution wiring patterns on electronic components. However, optical patterning has the drawback of excessive waste of materials such as photoresist, developer, and etching solution. Optical patterning also has the drawback of being difficult to improve process efficiency due to its complex process. Optical patterning also has the drawback of requiring the use of large-area masks, making it difficult to apply new designs to the production line within the shortest possible time.

[0004] In order to overcome the various drawbacks of optical patterning, a method for forming metal wiring by printing using ink has been developed as a method for directly applying a pattern to a substrate without a mask. One known printing method using ink is screen offset printing. This screen offset printing method involves printing a predetermined pattern by screen printing on the surface of a silicone blanket (a blanket made of silicone rubber), and then transferring the printed pattern to the substrate or film that is the intended printing target, thereby forming a pattern. Various inks that can be used for screen offset printing have been developed in the past (see, for example, Patent Documents 1 to 3).

[0005] JP 2017-069198 A JP 2020-009554 A JP 2014-507510 A

[0006] In response to the recent demand for finer circuit patterns, there is a demand for copper ink to be printed on substrates with higher resolution than ever before. For example, if the copper ink bleeds on the substrate when it is applied, short circuits may occur in the resulting circuit pattern. In particular, in the manufacture of electronic circuit boards having ultrafine wiring with a wiring line width of 20 μm or less, the characteristics of the copper ink used to print the wiring may affect the reliability of the electronic circuit board.

[0007] In view of the above problems, the present invention provides a copper ink that can suppress bleeding when printing copper ink onto a substrate and can produce a highly reliable circuit pattern with few short circuits, as well as an electronic circuit board and a method for manufacturing an electronic circuit board using the same.

[0008] In order to solve the above problems, according to one embodiment of the present invention, there is provided a copper ink containing copper particles, the copper ink having a shear rate of 1 sec -1 Viscosity η1 at shear rate of 10 sec -1 The copper ink has a thixotropy index TI(1 / 10), which is defined as the ratio (η1 / η10) of the viscosity at 0.01% strain to the viscosity at 0.15% strain at 100% strain, of 7.5 or less, and the thixotropy index TI(1 / 10) satisfies the following relationship (1): tanδ<-0.15 × TI(1 / 10)+1.31 (1).

[0009] According to another embodiment of the present invention, there is provided an electronic circuit board comprising a wiring layer using the above-mentioned copper ink.

[0010] According to yet another embodiment of the present invention, there is provided a method for manufacturing an electronic circuit board, comprising: forming a copper ink film on a blanket by screen printing using a copper ink; transferring the copper ink film from the blanket onto a substrate; and firing the copper ink film transferred onto the substrate to form a circuit on the substrate.

[0011] According to the present invention, it is possible to provide a copper ink that can suppress bleeding when printing copper ink onto a substrate and can produce a highly reliable circuit pattern with few short circuits, as well as an electronic circuit board and a method for manufacturing an electronic circuit board using the same.

[0012] 1 is a graph showing the relationship between the storage modulus G' [Pa] and the film thickness [μm] of a wiring formed on a substrate when the shear strain of the copper ink is 0.01% in a dynamic elastic modulus measurement of the copper ink according to an embodiment of the present invention.

[0034] FIG. 1 is a graph showing the flow curve of Copper Ink No. 1.

[0035] FIG. 1 is a graph showing the viscoelastic properties of Copper Ink No. 1.

[0036] FIG. 1 is a photograph of a linear pattern having a wiring line width of 20 μm printed on a PDMS blanket by screen offset printing using Copper Ink No. 1.

[0037] FIG. 1 is a photograph of a linear pattern having a wiring line width of 20 μm printed on a PI substrate by screen offset printing using Copper Ink No. 1.

[0038] FIG. 2 is a photograph of a linear pattern having a wiring line width of 15 μm printed on a PDMS blanket by screen offset printing using Copper Ink No. 7.

[0039] FIG. 3 is a photograph of a linear pattern having a wiring line width of 20 μm printed on a PI substrate by screen offset printing using Copper Ink No. 6. 1 is a graph showing the relationship between the thixotropy index (TI) of copper ink and tan δ at a strain of 0.01%. -1 1 is a graph showing the relationship between the viscosity η of the copper ink at 156 and tan δ at a strain of 0.01%.

[0013] (Copper Ink) One embodiment of the copper ink according to the present invention will be described in detail below. The copper ink according to the present invention is suitable for use in printing on a substrate. The copper ink is used in printing in which the copper ink is directly printed onto a substrate, such as screen printing through a mesh. Alternatively, as is known as screen offset printing, for example, the copper ink according to the present invention is particularly suitable as a copper ink for use in screen offset printing in which the copper ink is first printed onto the surface of a blanket in the form of a roll, plate, or block to form a copper ink film, and then the copper ink film is transferred from the blanket to a substrate.

[0014] <Thixotropy Index> Thixotropy describes the property of a substance where the viscosity changes over time, and can be expressed by the thixotropy index (TI). TI is expressed as the ratio of viscosities at different shear rates. A substance with a TI close to 1 is called a Newtonian fluid. A TI of 1 or greater is considered to be more thixotropic, with a high viscosity in the low shear region (static state) and a low viscosity in the high shear region (flowing state). The copper ink according to this embodiment preferably has a TI that allows it to be fluid with low viscosity during printing (high shear region), making printing possible.

[0015] Specifically, the copper ink according to this embodiment has a shear rate of 1 sec -1 Viscosity η1 at shear rate of 10 sec -1 When the TI (1 / 10) is higher than 7.5, bleeding may occur when a circuit pattern with a line and space (L / S) of 20 μm or less is printed by screen offset printing, which may increase the rate of short circuits (defect rate).

[0016] The lower limit of TI(1 / 10) is not particularly limited. However, a small TI(1 / 10) indicates that there is little difference between the low shear region (e.g., in a stationary state) and the high shear region (e.g., in a printed state). In order to provide a copper ink capable of accurately forming fine wiring with high connection reliability, it is preferable to have a TI(1 / 10) that has a certain degree of fluidity in a printed state and a certain degree of viscosity in a stationary state. Therefore, TI(1 / 10) is more preferably 1.5 or more, and even more preferably 2.5 or more.

[0017] The TI (1 / 10) can be measured using a dynamic viscoelasticity measuring device (rheometer). In this embodiment, a rheometer (model number: MCR102) manufactured by Anton Paar was used as the dynamic viscoelasticity measuring device, with the measurement section temperature set to 25°C, the measuring jig set to PP (parallel plate) 25, and the gap set to 0.5 mm. Using the dedicated software (RheoCompass ver. 1.20.471) provided with the dynamic viscoelasticity measuring device, the measurement mode was set to "flow curve measurement," and the shear rate was set to 0.01 to 1000 sec. -1 The viscosity was measured in the range of 1 s. -1 Viscosity at shear rate of 10 sec -1 The TI (1 / 10) was calculated based on the viscosity at the time of measurement. The amount of copper ink used for measurement was an amount that could fill the gap between the measurement jig and the device. Any copper ink that spilled out from the measurement jig was wiped off before measurement.

[0018] <Dynamic Modulus of Elasticity> The dynamic modulus of elasticity expresses dynamic viscoelastic properties as the deflection angle of the complex modulus of elasticity, focusing on the phase lag of the stress-strain characteristics of a viscoelastic material, and is used as an index for evaluating the viscoelasticity of an object. This dynamic modulus of elasticity is decomposed into two terms: the "storage modulus G'", which corresponds to the real part of the complex modulus of elasticity, and the "loss modulus G''", which corresponds to the imaginary part. The loss factor (loss tangent) tanδ is expressed as the ratio of the loss modulus G'' to the storage modulus G' (G'' / G').

[0019] When tan δ, expressed as the ratio of the loss modulus G" to the storage modulus G' (G" / G'), is 1 or less, it indicates that the material has elastic properties, and when tan δ is greater than 1, it indicates that the material has viscous properties. This embodiment focuses on the ratio of the elastic component and the viscous component of the copper ink in the elastic range where tan δ is 1 or less, and can adjust the composition of the copper ink so that it has properties favorable for forming fine wiring. This improves the ability to retain the wiring shape after printing the copper ink, making it easier to form fine wiring.

[0020] The state of a material when strained by 0.01% (hereinafter simply referred to as "strain" or "shear strain") essentially represents the viscoelastic properties of the material in a static state. The static state can be considered to represent the state of the material after printing with copper ink. Therefore, when a strain of 0.01% is applied to copper ink, copper ink that has a certain degree of elasticity (hardness) can be said to be able to maintain its shape even after printing. Such copper ink is particularly suitable for forming fine wiring.

[0021] In this embodiment, we evaluate the relationship between the printability of copper ink and the ratio of elastic and viscous components in the elastic range where tanδ is equal to or less than 1. As a result, we found that copper ink that has a certain relationship between tanδ and TI (1 / 10) at a strain of 0.01% is less likely to chip or bleed, even when printing a circuit pattern with an L / S of 20 μm or less.

[0022] <Relationship between tan δ and TI (1 / 10) at a strain of 0.01%> That is, the copper ink according to this embodiment has a shear rate of 1 sec -1 Viscosity η1 at shear rate of 10 sec -1 The thixotropy index TI(1 / 10) defined as the ratio (η1 / η10) of the viscosity η10 at a strain of 0.01% is 7.5 or less, and the thixotropy index (1 / 10) and tanδ[-], which is the ratio (G'' / G') of the storage modulus G' to the loss modulus G'' at a strain of 0.01%, satisfy the following relational expression (1): tanδ<-0.15×TI(1 / 10)+1.31 ... (1)

[0023] If the tan δ of the copper ink at a strain of 0.01% is too small, the elastic component becomes strong, making it difficult to eject the copper ink from the screen during the printing process, which may result in breaks in the printed wiring. Therefore, the tan δ of the copper ink at a strain of 0.01% is preferably greater than 0.12, more preferably 0.15 or more, even more preferably 0.20 or more, and still more preferably 0.25 or more. The upper limit of tan δ of the copper ink at a strain of 0.01% is not particularly limited, but is, for example, 1.00 or less, more preferably 0.50 or less, and even more preferably 0.40 or less.

[0024] The storage modulus G' of the copper ink at a strain of 0.01% is preferably 500 Pa or more, more preferably 900 Pa or more, even more preferably 1000 Pa or more, and still more preferably 1500 Pa or more. There are no particular limitations on the upper limit of the storage modulus G', but it is, for example, 20000 Pa or less, more preferably 10000 Pa or less, and even more preferably 5000 Pa or less.

[0025] The loss modulus G'' of the copper ink at a strain of 0.01% is preferably 250 Pa or more, more preferably 300 Pa or more, even more preferably 400 Pa or more, and still more preferably 500 Pa or more. There is no particular upper limit to the loss modulus G'', but it is, for example, 3000 Pa or less, more preferably 2000 Pa or less, and even more preferably 1000 Pa or less.

[0026] The storage modulus G' and loss modulus G" can be measured using a dynamic viscoelasticity measuring device, similar to the measurement of TI (1 / 10) described above. In this embodiment, an Anton Paar rheometer (model number: MCR102) was used as the dynamic viscoelasticity measuring device, with a measurement section temperature of 25°C, a measuring jig: PP25, and a gap of 0.5 mm. Using dedicated software (RheoCompass ver. 1.20.471) provided with the dynamic viscoelasticity measuring device, the measurement mode was set to "strain dispersion," and measurements were taken over strains from 0.01% to 100% to obtain measurement results for the storage modulus G' and loss modulus G" when the strain was 0.01%, and tan δ at a strain of 0.01% was calculated from these measurement results. The amount of copper ink used for the measurement is the same as in the above-mentioned TI (1 / 10) measurement, and is an amount that can fill the gap between the measurement jig and the device, and any copper ink that spills out of the measurement jig is wiped off before the measurement is performed.

[0027] FIG. 1 is an example graph showing the relationship between the storage modulus G' [Pa] of a copper ink at a shear strain of 0.01% and the film thickness [μm] of a wiring formed on a substrate, as measured using a viscoelasticity measuring device for a copper ink according to this embodiment. FIG. 1 shows the film thickness of a wiring layer obtained by forming a predetermined circuit pattern on a PI (polyimide) substrate using a screen offset printing method described in the Examples below, in which the copper ink according to this embodiment was used to perform screen offset printing so that the L / S of the wiring was 20 μm / 20 μm. The film thickness was measured using a confocal laser microscope (OPTELICS HYBRID+) (×50 lens) manufactured by Lasertec Corporation and dedicated software (LMeye7 ver. 7.5.2.26), measuring the film thickness of six wirings, and calculating the average value as the film thickness. The storage modulus G' at 0.01% was measured under the same conditions as those described for measuring the dynamic modulus above.

[0028] As shown in Figure 1, it can be seen that the copper ink according to this embodiment has a relationship between the storage modulus G' at a strain of 0.01% and the film thickness. As shown in Figure 1, a copper ink having a storage modulus G' of 900 Pa or more at a strain of 0.01% can achieve a film thickness of approximately 3 μm or more even when forming fine wiring with an L / S ratio of 20 μm / 20 μm or less. Therefore, the copper ink according to this embodiment can provide a copper ink that can form a circuit pattern on a substrate with a sufficient film thickness that makes wiring breakage less likely.

[0029] <Relationship between tan δ and viscosity η156 at strain 0.01%> Shear rate 156 sec -1 This roughly corresponds to the printing speed (squeegee speed) on the blanket using a screen mask in the Examples described later. If the viscosity of the copper ink during printing is low, for example, when a blanket made of polydimethylsiloxane (PDMS) is used as the blanket, the copper ink may sag during screen printing, making it difficult to form fine wiring.

[0030] In this embodiment, we evaluated the relationship between the printability of copper ink and the ratio of elastic and viscous components in the elastic range where tanδ is equal to or less than 1. As a result, we found that copper ink that has a certain relationship between the loss factor tanδ and the viscosity η156 is less likely to chip or bleed, even when printing a circuit pattern with an L / S of 20 μm or less.

[0031] That is, the copper ink according to this embodiment has a shear rate of 156 sec -1 It is preferable that the viscosity η156 [Pa·sec] is 2.0 Pa·sec or more, tan δ is greater than 0.12, and tan δ satisfies the following relational expression (2): tan δ<0.07×η156−0.05 (2).

[0032] The copper ink according to this embodiment has a shear rate of 156 sec -1 The viscosity η156 at a shear rate of 156 sec is more preferably 3.0 Pa·sec or more, even more preferably 4.0 Pa·sec or more, and even more preferably 6.0 Pa·sec or more.-1 If the viscosity η156 at the shear rate is too high, it may be difficult to uniformly eject the copper ink from the screen, making it difficult to form fine wiring. Furthermore, copper inks with extremely high viscosity η156 are generally difficult to manufacture. Therefore, the copper ink according to this embodiment has a shear rate of 156 sec. -1 The viscosity η156 at 15.0 Pa·sec or less is preferably 15.0 Pa·sec or less, more preferably 12.5 Pa·sec or less, even more preferably 12.0 Pa·sec or less, and still more preferably 11.0 Pa·sec or less.

[0033] Here, the shear rate is 156 sec -1 The viscosity η156 was measured using a rheometer (model: MCR102) manufactured by Anton Paar as a dynamic viscoelasticity measuring device, with the measuring section temperature set to 25°C, the measuring jig set to PP25, and the gap set to 0.5 mm. Using the dedicated software (RheoCompass ver. 1.20.471) attached to the dynamic viscoelasticity measuring device, the measurement mode was set to "flow curve measurement," and the shear rate was set to 156 sec. -1 The viscosity of the copper ink was measured in the same manner as in the above-mentioned TI, with the amount of copper ink used being an amount that could fill the gap between the measurement jig and the device, and the copper ink that spilled out of the measurement jig was wiped off before the measurement.

[0034] If the tan δ of the copper ink at a strain of 0.01% is too small, the elastic component becomes strong, making it difficult to eject the copper ink from the screen during the printing process and possibly causing breaks in the printed wiring, etc. Therefore, in relational expression (2), the tan δ of the copper ink at a strain of 0.01% is preferably greater than 0.12, more preferably 0.15 or greater, even more preferably 0.20 or greater, and still more preferably 0.25 or greater.

[0035] (Copper Ink Composition) The copper ink according to this embodiment contains (a) copper particles, (b) a binder resin, (c) a solvent, and (d) an additive.

[0036] (a) Copper particles Copper particles contain copper (Cu) and are often composed mostly of copper. The copper content of the copper particles, excluding copper oxide, is, for example, 98% by mass or more, and typically 99.5% by mass or more. The copper content of the copper particles can be confirmed and measured by X-ray diffraction (XRD).

[0037] The average particle size D50 of the copper particles can be, for example, 0.01 μm to 5.00 μm, preferably 0.05 μm to 1.00 μm, and more preferably 0.10 μm to 0.80 μm. The copper particles may be composed of a mixture of two or three types of copper particles with different average particle sizes D50. For example, copper particles having an average particle size D50 of 0.01 μm to 5.00 μm, copper particles having an average particle size D50 of 0.05 μm to 1.00 μm, and copper particles having an average particle size D50 of 0.10 μm to 0.80 μm may be mixed to produce copper particles. The method for measuring the particle size D50 of copper particles will be described below.

[0038] The content of copper particles in the copper ink can be, for example, 60.0% by mass to 90.0% by mass, preferably 65.0% by mass to 85.0% by mass, and more preferably 70.0% by mass to 80.0% by mass.

[0039] (b) Binder Resin The binder resin is not particularly limited, but may include, for example, one or a mixture of two or more resins selected from the group consisting of copolymer polyester, polyester, ethyl cellulose, acrylic resin, and epoxy resin. In particular, the copper ink according to this embodiment preferably includes an acrylic resin.

[0040] Examples of acrylic resins include resins containing (meth)acrylate units, such as polymers or copolymers of acrylic monomers such as alkyl (meth)acrylic esters, (meth)acrylic acid, and (meth)acrylamide, as well as copolymers of the above acrylic monomers with monomers such as styrene and maleic anhydride.

[0041] Considering the various properties required for a copper ink capable of printing fine wiring with an L / S ratio of 20 μm / 20 μm or less on a substrate with high precision, the weight-average molecular weight Mw of the resin is preferably 5,000 or more, more preferably 30,000 or more, and even more preferably 50,000 or more. Specifically, the weight-average molecular weight Mw of the resin is preferably 5,000 or more and 500,000 or less, more preferably 30,000 or more and 400,000 or less, and even more preferably 50,000 or more and 300,000 or less.

[0042] The content of the binder resin in the copper ink can be, for example, 1.0% by mass to 25.0% by mass, preferably 2.0% by mass to 22.0% by mass, and more preferably 5.0% by mass to 20.0% by mass.

[0043] When preparing the copper ink, the binder resin may be used in the form of a solution by mixing it with a solvent, which will be described later. Considering the properties required for a copper ink that can print fine wiring on a substrate with high precision and little bleeding, the solids ratio of the binder resin in the binder solution can be adjusted to between 20.0% by mass and 80.0% by mass, more preferably between 25.0% by mass and 75.0% by mass, and even more preferably between 30.0% by mass and 70.0% by mass.

[0044] Furthermore, the resin solids ratio of the binder resin to the copper particles in the copper ink [mass % vs Cu] is preferably 1.0 mass % to 15.0 mass %, more preferably 2.0 mass % to 10.0 mass %, and even more preferably 3.0 mass % to 8.0 mass %.

[0045] (c) Solvent The solvent mainly serves as a viscosity adjuster or diluent (solvent) to dissolve the binder resin and allow it to blend with the copper particles. The solvent is not particularly limited, but examples include terpenes such as terpineol and dihydroterpineol, ethers such as ethylene glycol butyl ether, diethylene glycol methyl ether, and diethylene glycol ethyl ether, and esters such as diethylene glycol monobutyl ether acetate, ethylene glycol ethyl ether acetate, diethylene glycol butyl ether acetate, ethyl lactate, butyl lactate, γ-butyrolactone, and propylene carbonate. These solvents can be used alone or in combination of two or more.

[0046] As described above, the solvent is contained in the binder resin so that the solids content of the binder resin is 20.0% by mass to 80.0% by mass. In addition, in the preparation of the copper ink, a solvent may be further added to the mixture of copper particles, binder resin, and additives as a solvent for dissolving and blending the copper particles, binder resin, and additives. The amount of solvent added together with the copper particles, binder resin, and additives when preparing the copper ink can be adjusted to, for example, a range of 0.5% by mass to 20.0% by mass depending on the desired properties of the copper ink.

[0047] (d) Additives Additives are substances added to improve the dispersibility or thixotropy of copper particles in the copper ink. Additives for improving dispersibility include, for example, polyether phosphate esters or mixtures of higher fatty acids and amines. Additives for improving thixotropy include, for example, polyolefins. The additives are diluted with the above-mentioned solvents as needed.

[0048] Considering the properties required of a copper ink that can print fine wiring on a substrate with high precision and little bleeding, the additive ratio to copper particles in the copper ink [mass % vs Cu] is preferably 10.0 mass % or less, more preferably 5.0 mass % or less, and even more preferably 3.0 mass % or less.

[0049] In addition to the materials described above, various additives may be added to improve the stability and printability of the ink. For example, a leveling agent, a viscosity modifier, a rheology control agent, an antifoaming agent, an anti-sagging agent, etc. Rheology control agents are used to control the rheology of the copper ink to prevent settling during storage of the copper ink, prevent sagging of printed matter, and improve ease of application.

[0050] (Method of Producing Copper Ink) Copper particles, a binder resin, a solvent, and an additive are prepared, mixed together, and then kneaded to produce copper ink.

[0051] The order in which the copper particles, binder resin, solvent, and additives are mixed together is not limited. For example, the additives may be mixed with the solvent, or in some cases, the additives may be mixed with the solvent beforehand, and then mixed with the copper particles and binder resin. The binder resin may be mixed with the solvent beforehand, and then mixed with the copper particles and additives. The copper particles used in the production of copper ink may be purchased or otherwise obtained. Alternatively, copper particles having predetermined properties may be prepared in advance by a liquid phase method such as a known chemical reduction method or disproportionation method.

[0052] According to this embodiment, by preparing appropriate copper particles, binder resin, solvent, and additives in appropriate ratios, bleeding when printing copper ink onto a substrate can be suppressed, thereby obtaining a copper ink that can produce highly reliable circuit patterns with few short circuits.

[0053] (Electronic Circuit Board and Method for Manufacturing Electronic Circuit Board) An electronic circuit board having a wiring layer formed using the copper ink according to this embodiment can be manufactured by a screen printing method, more preferably a screen offset printing method, using the copper ink. The method for manufacturing the electronic circuit board is not particularly limited. For example, when using screen offset printing as the method for manufacturing the electronic circuit board, a copper ink film is formed on a blanket using the copper ink. The copper ink film is then transferred from the blanket to a substrate. Furthermore, the copper ink film transferred to the substrate is fired to form a circuit on the substrate, thereby obtaining an electronic circuit board having a wiring layer using the copper ink according to this embodiment.

[0054] A polyimide substrate is preferred as the substrate. A blanket made of polydimethylsiloxane (PDMS) is preferred as the blanket. In a method for manufacturing an electronic circuit board using screen offset printing, for example, a screen mask is used in which a pattern plate made of Ni or the like on which a predetermined wiring pattern is formed is attached to a screen mesh. The number of meshes per inch of the screen mesh can be, for example, 400 to 900. The thickness of the screen mesh can be, for example, 15 μm to 42 μm.

[0055] Copper ink is placed on a screen mask, coated with a doctor blade, and then squeegeeed onto a blanket to form a copper ink film on the blanket. Urethane squeegee rubber or the like can be used as the squeegee. The squeegee speed is not limited to the following. To accurately print fine wiring with an L / S ratio of 20 μm / 20 μm or less, the speed is set to, for example, 5 mm / s to 50 mm / s. The squeegee pressure can be set to, for example, 0.5 mm to 3.0 mm, and the clearance can be set to, for example, 0.1 mm to 1.0 mm.

[0056] The copper ink film printed on the blanket is then transferred onto a substrate. The transfer conditions can be, for example, a transfer speed of 5 mm / s to 50 mm / s, a push-in amount of 50 μm to 400 μm, and a delay time of 0 sec to 60 sec. The copper ink film transferred onto the substrate is baked in an inert gas atmosphere at 200°C to 400°C for 10 to 60 minutes, thereby forming a circuit on the substrate. The delay time refers to the time from when a predetermined amount of copper ink (one cycle of copper ink) has been printed on the blanket until the transfer of the copper ink to the substrate begins.

[0057] Examples of the present invention will be described below together with comparative examples. These examples are provided for a better understanding of the present invention and its advantages, and are not intended to limit the present invention.

[0058] (Preparation of Copper Inks) Copper particles A to D, binder resins E to K, additives L to N, and solvents O to S were weighed into a 100 mL container so as to obtain the mixing ratios shown in Tables 1 and 2. This was stirred at 2000 rpm for 1 minute using a rotation / revolution mixer (Awatori Taro, ARE-310, manufactured by Thinky Corporation), and then degassed at 2000 rpm for 1 minute. The degassed mixture was passed through a three-roll mill (80E, manufactured by EXAT Co., Ltd.) for 10 passes at a roll gap of 5 μm and a roll rotation speed of 130 rpm, to obtain copper inks Nos. 1 to 27 and 29 to 30.

[0059] (Copper particles) Average particle size D 50 Copper particles A and B having an average particle size of 0.18 μm and average particle size D 50 Copper particles C having an average particle size of 0.15 μm and copper particles D 50 Copper particles D having a particle size of 0.77 μm were prepared. The average particle size D50 of copper particles A to D was determined by measuring the particle size of each copper particle using a laser diffraction / scattering particle size distribution analyzer. The average particle size D50 refers to the particle size at which the cumulative volume-based frequency of each copper particle reaches 50% in the particle size histogram (particle size distribution graph) obtained, and was measured in accordance with JIS Z8825 (2013). Specifically, 1 g of copper slurry was added to 10 mL of 2-propanol and subjected to ultrasonic irradiation for 10 minutes (ultrasonic cleaner US-3KS, 120 W, 38 kHz, manufactured by SND Corporation). 5 mL of the resulting dispersion was mixed with 15 mL of an aqueous solution of hexasodium metaphosphate, and the mixture was subjected to ultrasonic irradiation for 20 minutes to obtain a dispersion for measurement. The aqueous solution of hexasodium metaphosphate was prepared by dissolving 10 g of hexasodium metaphosphate manufactured by Junsei Chemical Co., Ltd. in 5 L of water. The laser diffraction / scattering particle size distribution measuring device used for the measurement was a MASTERSIZER 3000 manufactured by Malvern, and the measurement was carried out at a temperature of 40° C., with a stirrer speed of 2000 rpm and while irradiating 12 W of 40 kHz ultrasonic waves.

[0060] (Binder Resin) The following binder resins were used as binder resins E to K. Binder resin E: ER2300 manufactured by Kusumoto Chemical Co., Ltd. (30.0 mass% acrylic resin, 70.0 mass% terpineol, weight average molecular weight 200,000) Binder resin F: ER2602 manufactured by Kusumoto Chemical Co., Ltd. (70.0 mass% acrylic resin, 30.0 mass% butyl carbitol acetate (BCA), weight average molecular weight 5,000) Binder resin G: KFA2000 manufactured by GOO Chemical Co., Ltd. (35.0 mass% acrylic resin, 65.0 mass% terpineol, weight average molecular weight 250,000) Binder resin H: S-LEC ASM-3010 manufactured by Sekisui Chemical Co., Ltd. (35.0 mass% acrylic resin, 65.0 mass% terpineol, weight average molecular weight 50,000) Binder resin I: S-LEC manufactured by Sekisui Chemical Co., Ltd. ASM-3010 (acrylic resin 48.0 mass%, terpineol 52.0 mass%, weight average molecular weight 50,000) Binder resin J: M4200 (acrylic resin 35.0 mass%, terpineol 65.0 mass%, weight average molecular weight 200,000) manufactured by Soken Chemical Co., Ltd. Binder resin K: M4210 (acrylic resin 30.0 mass%, dihydroterpineol 70.0 mass%, weight average molecular weight 20,000) manufactured by Soken Chemical Co., Ltd.

[0061] (Additives) The following dispersants or thixotropic agents were used as additives L to N. Dispersant L: ED-152 (polyether phosphate ester) manufactured by Kusumoto Chemicals Co., Ltd. Dispersant M: ED-120 (mixture of higher fatty acid and amine) manufactured by Kusumoto Chemicals Co., Ltd. Thixotropic agent N: ET4010 (polyolefin) manufactured by Kusumoto Chemicals Co., Ltd.

[0062] (Solvents) The following solvents were used as solvents O to S. Solvent O: terpineol (specific gravity 0.934, boiling point 213°C) Solvent P: γ-butyrolactone (specific gravity 1.13, boiling point 204°C) Solvent Q: ethylene glycol (specific gravity 1.11, boiling point 197°C) Solvent R: ethylene glycol monobutyl ether (specific gravity 0.902, boiling point 171°C) Solvent S: propylene carbonate (specific gravity 1.2, boiling point 240°C)

[0063] Tables 1 and 2 show the mixing ratios of copper particles A to D, binder resins E to K, additives L to N, and solvents O to S for copper inks No. 1 to 8, 10 to 16, 18 to 27, and 29 to 30 used in the preparation of the copper inks. While the total of copper particles, binder resin, additives, and solvents equals 100%, some inks in Tables 1 and 2 do not add up to 100% due to rounding to the nearest tenth. In Tables 1 and 2, "Copper Particle Content" refers to the content (mass %) of copper particles in the prepared copper ink. "Resin Solids Ratio" refers to the ratio of the resin solids of the binder resin to the mass of copper particles in the copper ink (mass % vs. Cu). "Additive Ratio" refers to the ratio of additives to copper particles in the copper ink (mass % vs. Cu).

[0064]

[0065]

[0066] The evaluation results of the obtained copper inks Nos. 1 to 8, 10 to 16, 18 to 27, and 29 to 30 are shown in Tables 3 and 4.

[0067]

[0068]

[0069] The measurement equipment and conditions for evaluating the copper ink were as follows. Conditions not specified were the same as those described above. Dynamic viscoelasticity measuring equipment: Rheometer (model: MCR102) manufactured by Anton Paar. Temperature of the measuring part of the equipment: 25°C. Measuring jig: PP25 (parallel plate). Gap: 0.5 mm.

[0070] <Measurement of Viscosity and Thixotropy Index> Using the dynamic viscoelasticity measuring device, the measurement mode was set to "flow curve measurement," and the viscosity and thixotropy index were measured at each shear rate [sec -1 In the flow curve measurement, the shear rate of the copper ink was 0.01 sec. -1 From 1000 seconds -1 The change in viscosity η was measured when the viscosity was changed to 1. An example of the flow curve for copper ink No. 1 is shown in Figure 2.

[0071] <Measurement of storage modulus G', loss modulus G'', and loss factor tan δ at 0.01% strain> Using the dynamic viscoelasticity measuring device described above, the storage modulus G', loss modulus G'', and loss factor tan δ shown in Table 3 were determined as the dynamic modulus of the copper ink by setting the measurement mode to "strain dispersion" and the rotation speed (frequency) of the plate of the viscoelasticity measuring device to 1 Hz. In the strain dispersion mode measurement, the storage modulus G', loss modulus G'', loss factor tan δ, complex modulus, and shear stress were measured when shear strains of 0.01% to 100% were applied to the copper ink. FIG. 3 shows an example of measurement of copper ink No. 1. In FIG. 3, the storage modulus G' and loss modulus G'' gradually decrease as the strain increases, and at a strain of approximately 10%, G' = G'' (= tan δ = 1). It can be seen from FIG. 3 that in a substantially static state with a strain of 0.01%, the storage modulus G' is larger than the loss modulus G''.

[0072] <Evaluation of Defect Rates When Printing on PDMS Blankets or PI Substrates> Screen offset printing was performed using copper inks Nos. 1 to 8, 10 to 16, 18 to 27, and 29 to 30, and the printability was confirmed. A commercially available printing device (manufactured by Mino Group Co., Ltd., model number SO-1010) was used.

[0073] The substrate used for printing was a PI substrate (Apical, polyimide substrate, manufactured by Kaneka Corporation), and the blanket was the aforementioned polydimethylsiloxane (PDMS) blanket roll (Fujikura Composites, Type: #700-STD). A 32 cm x 32 cm screen mask with a linear pattern formed on a 2 cm x 2 cm area with L / S of 20 μm / 20 μm, 15 μm / 15 μm, and 10 μm / 10 μm was used. A Ni pattern plate was used as the mask, and this was attached to a screen mesh by electroplating. The mesh count per inch was 500, the mesh material was stainless steel, the mesh wire diameter was 13 μm, and the total thickness was 31 μm. For screen offset printing, copper ink was placed on the screen, coated (filled) with a doctor blade, and then squeegeeed onto the blanket. The printing conditions were as described above.

[0074] The copper ink circuit patterns printed on the substrate were observed with an optical microscope, and images were obtained. Examples of the obtained images are shown in Figures 4 to 7. Five prints were performed using each copper ink, and images of arbitrary regions of each resulting wiring pattern were obtained. The percentage of straight lines in which adjacent straight lines were connected horizontally among the total number of straight lines present in the five obtained images was defined as the "defect rate," and the average defect rate was calculated. When printing onto the PDMS blanket, a defect rate of 5% or less was rated "Good," a defect rate of more than 5% was rated "Fair," and a defect rate of unprintable was rated "Poor." When printing onto the PI substrate, a defect rate of 7% or less was rated "Good," a defect rate of more than 7% was rated "Fair," and a defect rate of untransferable was rated "Poor." In this evaluation, "unprintable" refers to cases where printing onto the PDMS blanket using copper ink was not possible. "Untransferable" refers to both cases where printing onto the PDMS blanket was possible but printing from the PDMS blanket to the polyimide substrate was not possible, and cases where printing onto the PDMS blanket was not possible.

[0075] Figure 4 shows an image after printing on a blanket using copper ink No. 1 (Example) with a line / space ratio of 20 μm / 20 μm. Figure 5 shows an image after transferring from the blanket to a PI substrate using copper ink No. 1 (Example) with a line / space ratio of 20 μm / 20 μm. In both of the examples shown in Figures 4 and 5, there are no connections between adjacent straight lines. Therefore, the defect rate for both is 0%. Note that Figure 4 does not show short circuits between adjacent lines in the horizontal direction, but rather several breaks within the same line. However, these breaks are largely dependent on the ink printing conditions and are therefore excluded from the defect rate evaluation. Figure 6 shows a photograph after transferring to a blanket using copper ink No. 7 (Example) with a line / space ratio of 15 μm / 15 μm. In this example, it was confirmed that adjacent straight line patterns were connected horizontally in 20 of the 160 straight lines, resulting in a defect rate of (20 / 160) x 100 = 12.5%, and the evaluation was Fair. Figure 7 shows a photograph after printing on a PI substrate using copper ink No. 6 (comparative example) with an L / S ratio of 20 μm / 20 μm. In this example, printing was successful, so the evaluation was Fair, but because all lines were connected horizontally, the defect rate was 100%.

[0076] (Evaluation) <Relationship between tan δ and TI(1 / 10) at 0.01% Strain> In Table 3, copper inks Nos. 1, 2, 3, 11, 14, 18, 5, 13, 16, 27, 10, 23, 24, 15, 26, 7, and 12 according to the examples, which have a TI(1 / 10) of 7.5 or less and a tan δ (represented by the ratio of the storage modulus G' to the loss modulus G" (G" / G') at 0.01% strain) and a thixotropy index TI(1 / 10) that satisfy the relational expression (1), show that when a linear pattern with an L / S ratio of 20 μm / 20 μm was produced, the defect rate was kept to 5% or less when printed on both a PDMS blanket and a PI substrate. Copper inks No. 29 and 30 were evaluated for linear patterns transferred from a PDMS blanket to a PI substrate when the L / S was 15 μm / 15 μm, rather than forming a linear pattern with a 20 μm / 20 μm L / S. It can be seen that copper inks No. 29 and 30 suppressed the printing defect rate on the PI substrate to 5% or less, even when the L / S was 15 μm / 15 μm. These results suggest that even when copper inks No. 29 and 30 are used to print wiring patterns with a 15 μm / 15 μm L / S on a PDMS blanket, the defect rate is likely to be 5% or less. Figure 8 is a graph showing the relationship between tan δ and TI (1 / 10) at a strain of 0.01% for copper inks Nos. 1 to 8, 10 to 16, 18 to 27, and 29 to 30. The straight line in the graph represents relational expression (1). Copper ink Nos. 1 to 30, which are comparative examples that do not satisfy relational expression (1), are also highly likely to achieve a defect rate of 5% or less. Nos. 8, 4, 6, 19, 20, 21, 22, and 25 were all inferior to the Examples in printability onto the PDMS blanket and transferability onto the PI substrate.

[0077] <Relationship between tan δ and viscosity η156 at strain 0.01%> FIG. 9 shows the relationship between tan δ, which is expressed as the ratio (G″ / G′) of the storage modulus G′ to the loss modulus G″ at a strain of 0.01%, and the viscosity η156 at a shear rate of 156 sec. -19 is a graph showing the relationship between the viscosity η156 of the copper ink at 100°C and the viscosity η156 of the copper ink at 100°C. As can be seen from Fig. 9, copper inks No. 5 and 16 are copper inks according to the examples, but do not satisfy relational expression (2). Copper ink No. 1 according to the example, which satisfies relational expression (2), has TI(1 / 10) and tan δ at a strain of 0.01% that are relatively close to those of copper inks Nos. 5 and 16, but its printability on the PDMS blanket and its transferability to the PI substrate were superior to those copper inks.

Claims

1. A copper ink containing copper particles, -1 Viscosity η1 at shear rate of 10 sec -1 a thixotropy index TI(1 / 10) defined as the ratio (η1 / η10) of a storage modulus G' to a loss modulus G'' at a strain of 0.01% (G'' / G'), and the thixotropy index TI(1 / 10) satisfy the following relational expression (1): tanδ<-0.15×TI(1 / 10)+1.31 (1).

2. Shear rate 1 sec -1 2. The copper ink according to claim 1, wherein the viscosity η156 [Pa·sec] at η156 is 2.0 Pa·sec or more, the tan δ is greater than 0.12, and the tan δ satisfies the following relational expression (2): tan δ<0.07×η156−0.05 (2).

3. The copper ink according to claim 1, wherein the thixotropy index TI (1 / 10) is 1.5 or more.

4. Shear rate 156 sec -1 3. The copper ink according to claim 2, wherein the viscosity η156 at η156 is 15.0 Pa·sec or less.

5. The copper ink according to claim 1, which is a copper ink used in screen offset printing.

6. The copper ink according to claim 5, which is a copper ink for printing using a blanket made of polydimethylsiloxane.

7. The copper ink according to claim 5, which is a copper ink for printing on a polyimide substrate.

8. The copper ink according to any one of claims 1 to 7, comprising the copper particles, a binder resin, a solvent, and an additive.

9. The copper ink according to claim 8, wherein the copper particles have an average particle size (D50) of 0.01 μm to 5.00 μm.

10. The copper ink according to claim 8, wherein the binder resin comprises at least one selected from the group consisting of copolymer polyester, polyester, ethyl cellulose, acrylic resin, and epoxy resin.

11. The copper ink of claim 8, wherein the solvent comprises at least one selected from the group consisting of terpenes, ethers, and esters.

12. The copper ink according to claim 8, wherein the resin solids ratio of the binder resin to the copper particles is 1.0% by mass to 15.0% by mass.

13. An electronic circuit board comprising a wiring layer using the copper ink according to any one of claims 1 to 7.

14. A method for manufacturing an electronic circuit board, comprising: forming a copper ink film on a blanket by screen printing using the copper ink according to any one of claims 1 to 7; transferring the copper ink film from the blanket onto a substrate; and firing the copper ink film transferred onto the substrate to form a circuit on the substrate.

Images