JPS6081259A - Production of cu electrically conductive coating material - Google Patents

Abstract

PURPOSE:To obtain a Cu electrically conductive coating material, suitable to printing of lead wires with a fine line width, and having thixotropic properties, by incorporating a thermosetting resin having the reductive property in reaction at ordinary temperature and thermosetting reaction of fine Cu powder in a specific amount. CONSTITUTION:(A) Fine Cu powder is incorporated with (B) a thermosetting resin having reductive property at ordinary temperature and in thermosetting reaction, e.g. URUSHI based resin, and (C) an unsaturated fatty acid and (D) 1.5wt% fine SiO2 powder, e.g. having the surfaces of the respective particles surrounded with hydrophobic molecular groups such as CH3, and having <=50mmu average diameter. EFFECT:Improved moisture resistance of films.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for producing a Cu conductive paint for printing, particularly a paint having thixotropy suitable for printing conductive wires with minute line widths.

Conventional conductive paints for printing have mainly consisted of heg powder suspended in a resin solution. Such A(7 paints) are widely used in the electronic parts industry as paints for room temperature drying or high temperature baking. This is because, in addition to this low value, the printing properties are good and the physical properties such as post-baking moisture resistance are excellent.

However, there are major drawbacks as follows. That is, when used as a substitute for connecting conductor wires and jumper wires for resistors, chimp capacitors, etc. in a high humidity environment, if the distance between the wires is small and the potential difference between the wires is large, the phenomenon of silver migration may occur. , short circuit may occur in severe cases, and the price is high.

Recently, research has been conducted to obtain a substitute for Ag paint by using Cu powder instead of A(7 powder).This is a natural trend.

However, the development of a paint comparable to paint A(7) has not yet been successful.

Our company has been conducting research from an early stage and has recently succeeded in inventing a Cu paint that can be put to practical use.

For example, 1 Japanese Patent Application No. 55-098778 (conductive paint using lacquer-based resin), 2 Japanese Patent Application No. 55-129115 (conductive film II using derivatives of hydroquinones), 55-172645 (conductive paint using lacquer-based paint), Conductive paint used) 4, 56-010843 (Conductive paint using hydroquinone derivatives) The reason for the success of Cu paint is that we researched the conditions that the binder resin should have and selected a resin that met those conditions. It's what I did. That is, as follows.

(1) It must be thermosetting.

(2) Must have reducing properties even at room temperature.

(3) Must have reducibility even during heat curing.

(4) Molecules are tightly bonded to each other by a hardening reaction.

Since it is already widely known that (1) is generally a necessary condition for a printed resistor, the explanation thereof will be omitted.

Considering that the conductive wire obtained by sintering Cu paint is also a type of low resistance wire, it is natural that the condition (1) above is required. The reason (2) is necessary is because Cu is a conductor.
This is due to the fact that the powder is extremely small, with a particle size of (1-5) μm. Then, the surface of Cu will have a high activity level, adsorb surrounding moisture, 02, So□, etc., and quickly create compounds, so the surface resistance will be low.
It becomes difficult to use as a conductive paint.

It is necessary to eliminate this drawback, and at the same time, even if the surface of the Cu powder is slightly oxidized and covered with a thin oxide film, it will be reduced during paint preparation and the contact resistance of the Cu powder will be reduced. This is because the overall conductor resistance becomes small. Condition (3) is necessary during the thermosetting reaction. If the reaction product produced when the resin transforms into a three-dimensional structure by heating is oxidizing, the Cu surface is immediately covered with an oxide film and loses its conductivity. Even if it is neutral, 02 in the air is Cu
If it comes into contact with the powder surface, an oxide film will be formed on the powder surface, impairing its conductivity. Therefore, in order for the Cu powder itself not to lose its conductivity even during a severe chemical change called a curing reaction, the curing reaction of the resin must naturally be a reducing reaction.

Many resins do not have such properties, and attempts have been made to remedy this by adding reducing substances that do not take part in the reaction, but this does not seem to yield satisfactory results. .

The fourth characteristic is that the electrical conduction mechanism of the conductive film obtained by forming a coating film with Cu conductive paint and sintering it is mostly due to the contact current flowing through the contact surface of the CV/wooder. When the resin hardens tightly, the Cu powder itself firmly contacts each other and the contact area becomes a dog.
As a result, contact resistance becomes small. Therefore, the conductivity becomes good.

All inventions related to the above application filed by the same applicant as the present application are:
This success was achieved by using a resin with such characteristics.

The above-mentioned Cu paint can be used to form printed wiring by screen printing as a substitute for conductive wiring made by etching a general O1L laminate (abbreviated as PC board). However, in recent years high-density printed resistors (PR
As the practical use of the technology (abbreviated as "C") progressed, major problems arose. The reason for this is that conventionally, even if the printed conductive wire had a small width, it was sufficient to have a width of one width. However, for high densities, the conductor width is usually 0.5 mm, and the minimum is 0.2 mm or 200 mm.
μm is required.

C = paint (even with conventional HG paint, the minimum line width is 1)
mm is considered to be the limit. The reason is as follows. FIG. 1 shows a cross-sectional view perpendicular to the length direction of a conductive wire formed on an insulating substrate 1 using Cu paint by screen printing. The shaded area 2 shows the cross section immediately after printing. Printing paint has a very hard pace!・Since it is shaped like this, it can be considered to have a roughly rectangular cross section. Generally the thickness t is (30~50)μ
m or normal.

When this is heated and cured at a high temperature of 110 DEG C. to 150 DEG C., reaction products such as H20 are generated during the process, the viscosity decreases, and a sweet "sauce-1" as shown in 6 is produced.

This phenomenon occurs similarly in the case of 1'TLC. "If the width of the drip-1 is indicated by ε as shown in the figure, in the case of a paint suitable for screen printing, the width of the sauce-1 is 6 mm (ε
The minimum value of 0.25 mm or 250 μm is the limit.

Then, the minimum value of line width B is clearly Bmi = 0.
7 van. For this reason, the minimum resistance width of PRO is set to 0.7 mm. In this case, the flat portion BO of width 13 is at most 0.2 cylinders. Therefore, if the minimum width is tentatively set to 0.5+nm, the flat portion of B will be zero, and the distribution of resistance values will actually become wider and the film thickness will become unstable, resulting in a lower yield and thus making it impossible to put it into practical use.

In the case of Cu paint, the result is similar; the resistance value differs in each part of the conductive wire, resulting in a large value as a whole, and in some cases, it may become unusable.

Therefore, the minimum line width that can be obtained even with the above-mentioned Cu paint is 1 mm, which can be said to be 0.7 mm, which is the practical minimum line width. In any case, 200μ is required for high density.
A conductive wire having a line width of 500 μm, let alone a line width of 500 μm, cannot be realized regardless of the screen printing mask.

The purpose of the present invention is to investigate the causes of the sag-1 described in detail above, and to achieve a technique for achieving a line width of at least 500 μm when CALm material is used.

In order to achieve this objective, the method for producing a Cu conductive paint according to the present invention includes: (i) a Cu conductive paint consisting of a fine Cu powder, a thermosetting resin, an unsaturated fatty acid, and a fine 5i(J)2 powder; The thermosetting resin is selected to have reducibility at room temperature and during the thermosetting reaction, and (11) the content of the 5i02 powder is 1.5% or less.

Next, the configuration of the present invention will be explained in detail.

The present inventor has previously conducted research on four-wet paints with carbon resistance. Cylindrical porcelain substrate (hereinafter referred to as 7+)
It is abbreviated as in, ). ), efficient dip (tip) painting is performed. In this case, even if the viscosity is the same,
Undue emphasis was placed on sagging of paint on porcelain terminals or a decrease in the thickness of the paint film on the edges of the terminals.

That is, as shown in FIGS. 2(a) and 2(b). (a
, ) and (b) are axial horizontal and vertical cross-sectional views of the carbon film resistor. In the figure, 4 is a bobbin, 5 is a carbon layer, and 6 is a coating film. (σ,) makes the bobbin horizontal,
(b) shows the case of drying in a vertical position, and in both cases the coating film is thin or ``sagging'' at the edges. As a result of various studies, it has been found that the above problem is related to the stringiness of the paint material rather than the viscosity.

In other words, if the stringiness is removed to some extent, the above-mentioned drawbacks can be greatly improved.

I noticed that the problem with 0W paint is essentially the same as the problem with general paints. As a result of repeated preliminary experiments, it was confirmed that it was effective, and as a result of proceeding with this research, we were able to obtain good results. Next, this point will be explained in detail.

When a highly viscous material, such as paint, is applied thickly, it maintains its original state without reducing the film thickness at the edges or "sagging", and generally has thixotropic properties (hereinafter referred to as (abbreviated as thixotropy) is said to be thick. In order to improve the thixotropy, it is sufficient to reduce the stringiness of the paint or the resin on which the paint is based.

For example, when fine powder such as ZnO2 is added, the titanium property increases. However, if the titanium properties are increased in this way, if the paint is dried or sintered at high temperatures to form a strong paint film, the addition of pigments such as the ZnO2 powder will increase the mechanical strength of the film. As a result, adhesive strength and the like naturally decrease, and furthermore, the film itself gradually becomes porous, resulting in deterioration of the essential moisture resistance.

Therefore, the problem boils down to what pigment or powder should be selected in order to improve the properties of the coating film described above while maintaining a sufficiently high titanium property.

The first consideration is that the powder itself must be moisture resistant. This is naturally required for paints. It is desirable to have hydrophobicity. Hydrophobicity and moisture resistance are often thought to be similar phenomena, but this is a misunderstanding. Hydrophobicity is a property of resistance to macroscopic water molecules, and moisture resistance is a property of resistance to water molecules in a microscopic form, and they are rarely compatible.

For example, silicone resins are generally highly hydrophobic, but they cannot be said to have excellent moisture resistance.

Next, when considering adhesion, mechanical strength, and moisture resistance properties as a paint, it is clear that the smaller the amount of pigment or inorganic powder, the better the properties.

Therefore, the technological nature of paint and its original characteristics as a paint are contradictory. However, in terms of properties as a paint, the larger the coating film t, the better.
It is true that it is better to mix pigments to slightly increase titanium properties. However, if the thickness t is the same, it is better not to mix pigments. In order to make t the same, pure resin paint can be coated in multiple coats. The properties are better in this way than in the case where a powder that increases the titanium properties is mixed.

From the above detailed explanation, we have experimentally reached the following conclusion in order to impart titanium properties to conductive paint within a practical range.

(1) If the inorganic powder is of the same type, the smaller the particle size, the better.

(2) For different types of powder, each powder is OH
Those surrounded by hydrophobic molecular groups like 3 are better.

(3) It is better for the powder itself to have a moisture respiration effect, such as rutile-type TiO3. (For example,
5-18360) (4) Generally, even if the substances are different, powders with much smaller particle sizes are more effective.

The above conclusion was also confirmed through experiments,
This seems to be acceptable based on common sense.

However, since there are no doubts regarding item 4), an example of the experimental results is shown in FIG. In the figure, the horizontal axis is the weight ratio when powders of different types and particle sizes are mixed with the paint of Example 1 (described later) used for the test, and the vertical axis is the weight ratio when powders of different types and particle sizes are mixed with the paint of Example 1 (described later), and the vertical axis is the weight ratio of powders of different types and particle sizes. The graph shows two times the average value of the drop J after printing the conductor and drying and sintering it.The curve ■ shows the curve ■ when using Al2O3 with an average diameter of 10 μm.
When rutile type Ti2O3 of 5 μm is used, ■ is 0.
The results are shown when using 01 μm 5i02. This clearly shows that the above-mentioned item (4) is effective. That is, S
? :02 gave much better results, and when the mixture amount was 1.0 wt%, "Dare-127 was 0.15".
It is found that 6% is about o, i amorous.

The next powder that can be used is At201. TZ'
There are many types such as 203°ZnO2 and CcC03, but based on the condition (4), they are limited to 5i021 types. This is because the particle size is about two orders of magnitude smaller than that of other powders.

This fact was confirmed by experiment.

Next, even if the powder is limited to SiO2, there are many types depending on the particle size. Among the things that are called white carbon, the main ones are as follows.

(a) Silicic anhydride: Aerosil (Japan Aerosil) (K
(manufactured by) A I50, 200, 300, 380, 0X50
．． T'11. ”600.JL972 etc.

Particle size 7-40mμm (b) Hydrous silicic acid dinib seal VN3 (Nihon Rika).

Particle size 1.7-4.1 μm Carplex (Shionoki Pharmaceutical■ <K made) A80°112
0th class Ultrasil (Germany, Tegunsa DECj [JSSA) VN
3 Among the above, the one with the best characteristics is the Aerosil type, and the flat 972 is particularly excellent. This will be briefly discussed later. Therefore, the characteristics of the paint using 1(972) will be mainly explained by examples.

(Example 1) Cu powder (1-5) μm 71 Note★1 Resin 1 Pyrocatechol derivative) 15 Resin 2 Phenol resin Solvent (butylcarbyl) 11 Unsaturated fatty acid (oleic acid) 6 100% 1 Here, R is given by the following formula.

R= −(Cl−1) 1−(CI(), n−C,,I
-1271-1-1 or J(nee〇〇〇 (C)I
)1 (Cl-l2), nCnl-12,++ (2
)1, m, n are arbitrary integers.

(These details are described in Japanese Patent Application No. 55-154703, or are omitted for the sake of brevity.) The following steps are used to produce a Cu paint for printing using the raw materials having the above composition. Add an appropriate amount (less than the specified value) of a solvent to the resin 1 to be kneaded and add 2 parts to it and mix thoroughly, then add oleic acid and mix further. Next, the Cu powder is treated in II2 gas at around 500 °C to sufficiently reduce it.
Return to room temperature, immediately add to the mixed solvent and mix,
Furthermore, add a small amount of +(,972) and gradually add the solvent to the mixture to make it considerably larger than the specified value and mix thoroughly.

Mixing was done using a ball mill. It is then transferred to a sowing machine and finally rolled to evaporate the solvent and bring it to the specified value. I
L-972 (manufactured by Nippon Aerosil KJ <; average particle size 16
++17t1++,) is heat treated at around 500° C. to drive out moisture and other impurities before use.

In the above case, the amount of 几-972 is taken as a parameter and its value is set to 0.0% of the weight of the Cu powder. ．． 0.2, 0; 5, 10
, 1.5% Cu conductive coatings of type 5 were prepared, and these four types of paints were tested.

The sample is prepared in exactly the same manner as in the case of a widely known printed resistor (abbreviated as pn, c), so the explanation will be omitted for the sake of simplicity. Next, regarding the dimensions of the sample, etc.
This will be explained with reference to the figures. The substrate 1 is made of Bakelite Pc boat and Tetron fabric of printing mask f4200 mesh.

The length of the printed line L is 1DO+nm, 10 lines are printed in parallel, and the line width I3° is 5 lines. The baking conditions were 14 [3°C, 1 hour].

When the line width B after completion is measured, ε and 7 are given by the following equation.

The average value 7 of the completed thickness t is 45 μI11 if the weight ratio of u-972 is 0.5% or more, and the variation is extremely small and completely zero. The measurement results are shown in FIG. According to the figure, the average value of ``dare'' was 0.38 mm before adding R,972.
5i02. Just adding 1.0% of R-972 resulted in 0.0%.
It decreases by 16mmK. As is clear from the figure, 1 (97
If 2 is increased, ε becomes smaller, but as mentioned above, problems arise in terms of moisture resistance and mechanical strength."Second, 11.97
In the case of ultrafine powder 5202 such as CI7. The conductivity of the paint decreases rapidly around 2%, making it unusable.

Therefore, the weight limit for Cu powder is up to 1.
It cannot exceed 5 sections.

Then, in order to make ε small, as is clear from FIG. 1, it is necessary to reduce the thickness of the coating film by an amount equal to 1. Next, let's think about this point. In order to simplify the description, only the experimental results will be described. In printing a conductive path as shown in Fig. 1, the line width Bo
"F: 0.8 mm, length (]-) = 50 mm jumper wire manufactured, side 11 left standing test (40 ° C, 95 ratio H, H, test time 5001-1?)
and humidity resistant load condition, test (40℃, 95% river I load current 1
The test was carried out under the conditions of 007A7A and test time 5001(r), and the rate of change δIt was measured.

δR−Fuichiyo” (x1oochi) (3) R, R no, and 1.L are the resistance values before and after the test, respectively. The results are shown in Figure 5. The horizontal axis is the film thickness t. .

The vertical axis is the value of δ1(,.

In the figure, lIi+ glands (1) and (++) indicate the results of the humidity resistance test and the 1li1 humidity load test, respectively.

Since the allowable limit of δR in both tests is generally 30%, as is clear from these results, the minimum value of the thickness t is l mi )30 , u) T' (4) from FIG. 5.

As described in next K i;ii, the maximum value of R972 usage is about 1.5%, and in this case, t=45nlμ7n
It is. And from FIG. 4, it is 27 words 0.15 tnm. Therefore, if we assume that the law of proportionality holds approximately, then t
The value of 2 to I for ≧3Qmμm is 2T++i ”” X D, 15 ”= r[l],
1 mm (5) 5 In other words, the minimum value (average value) of the "sauce" obtained in this case
・01 invitation.

Therefore, referring to Figure 1, even if you look at the thick side as Bo"0.2 inch, it is B20.2 + 0.I x 220, 4mm.
The objective of 500 μm line width = 0.5 mm can be achieved.

Curves (III) and (IV) in Fig. 5 are a high temperature storage test (100°C, 500 Hr) and a load life test (70°C), respectively.

100 mA, 5001-1r), and in this case, there is no problem even if the film thickness is 20 m or less.

(Example 2) Cu powder (1-5) μm 69 Resin 1. Ur/Resin) 16 Reson 2. Ehokin Melamine Resin Solvent Petil Calbi!・-Rule 11 Unsaturated fatty acid (lilunic acid) 3 S? : 02 (R-972) 1 100% (Example 3) Cu powder (1-5) μm 69 Solvent Butyl carbitol 105 Unsaturated fatty acid (linoleic acid) SZ: 02 (J also -972) 15 100% Implemented Example 2. The results of the above-mentioned tests in B are not significantly different. It is omitted to simplify the description.

Finally, the effect of using hydrophobic 5j02 surrounded by atomic groups will be briefly explained in the sixth drawing.

The figure shows 5Q made by printing with this paint when Aerosil II, -972 (particle size 2Qmμ7n, hydrophobic) and Aerosil NO380 (particle size 7mμm) were added at 1% weight to the C11° weight of Example 1. ,+++mX
4 II + mX 35 nohm jumper wire 40
It shows the resistance change when 10001-Ir was left in a bath at 95% [°C, 95%]t and H. tllt g< (a) l
(b) shows the cases of IL-972 and NO380, respectively. Considering the particle size, I should have gotten a better result with No. 380, but J(, -972 is considerably better. This also allows for 1 molecule to be surrounded by C113 molecules. It is possible to clearly understand the effects of

Next, the effects of the present invention will be briefly explained.

C? /In conductive paint using howder (207-5
0) If a conductive path is printed to a thickness of μ7+1, high-density printing cannot be achieved unless the edge portion is made smaller. [In order to reduce the droop-j, it would be better to add thixotropic properties by mixing inorganic micropowder, but then the electrical characteristics would deteriorate as described above. Therefore, it was confirmed that fine powder having the following characteristics is suitable.

(1) When using different types of powder, the smaller the particle shape, the better.

(2) The same can be said for powders of the same genus.

(3) The one suitable for having thixotropic properties is Haku02 powder. This is because particles with an extremely small particle size can be obtained compared to other inorganic powders.

(4) Even if the particle size is the same, a powder surrounded by hydrophobic colloids such as c■-13 is better.

(5) When using 5in2 powder, the content is Cu
It must be 1.5% or less of the weight of the powder.

(6) When SZ:02 R-972 is used, when the conductor thickness t is 25 μm, the width of the “drop” can be set to 0.1 mm. Moreover, in this case, no deterioration is observed in the electrical characteristics. Conventionally, this value was 0.25+u+n
Met. In other words, it became less than 1/2.

Therefore, it was possible to greatly contribute to higher density.

[Brief explanation of the drawing]

Fig. 1 is a cross-sectional view perpendicular to the length direction of the printed conductor, and Fig. 2 (d) and (b) are carbon film resistors, respectively):
A horizontal 1-vertical cross-sectional view in the axial direction of ? 11. Figure 5 is a graph showing the relationship between the film thickness of the printed conductor and the rate of change in resistance in humidity-resistant storage and humidity-resistant load tests. It is a graph showing the wet storage characteristics of the printed conductor after sintering. 1, 4...Bohin, 5...
・Carbon layer, 6 Coating III is shown.

Claims (1)

[Scope of Claims] 1. A Cu conductive paint comprising minute O powder, thermosetting resin, unsaturated fatty acid, and minute SiO2 powder, in which: (i) the thermosetting resin has reducibility at room temperature and during a thermosetting reaction; (11) A method for producing a Cu conductive paint, characterized in that the content of the 5i02 powder is 1.5% or less. 2 Said S? The method for producing an 0w4 electric paint according to claim 1, wherein the surface of each particle of :02 is covered with four hydrophobic molecular groups such as c■-■3. 3. The method for producing a Cu conductive paint according to claim 1 or 2, wherein the average diameter of the 5i02 particles is 50 mμm or less. 4. The method for producing a Cu conductive paint according to any one of claims 1 to 3, characterized in that a sumac resin is used as the reducing resin. 5. The method for producing a Cu conductive paint according to any one of patent scopes 1 to 4, characterized in that a pyrocatechol derivative is used as the reducing resin.