METAL PASTE

Abstract

The present invention relates to a metal paste for forming a metal wiring containing a solid content of a silver particle and kneaded with a solvent. The solid content of the metal paste contains a silver particle having prescribed particle size distribution and average particle size, and using an amine compound as a protective agent. The solvent is a mixed solvent in which two organic solvents of a solvent A and a solvent B are mixed. The solvent A is dihydroterpineol or terpineol, and the solvent B is at least one organic solvent having a boiling point of 240? C. or more. The mixed solvent has a Hansen solubility parameter distance Ra from dihydroterpineol of 3.0 MPa.sup.1/2 or less. The metal paste further contains a high molecular weight ethyl cellulose as a first additive, and a polyvinyl acetal resin as a second additive.

Claims

1. A metal paste comprising a solid content of a silver particle having a protective agent bound to the particle and kneaded with a solvent, wherein the solid content comprises silver particles including a silver particle having a particle size of 100 to 200 nm in a ratio of 30% or more in terms of the number of particles, and an average particle size of the whole silver particles is 60 to 800 nm, the protective agent bound to the silver particle is at least one amine compound having 4 or more and 8 or less carbon atoms, the solvent is a mixed solvent in which at least two organic solvents of a solvent A and a solvent B are mixed, the solvent A is at least either dihydroterpineol or terpineol, the solvent B is at least one organic solvent having a boiling point of 240? C. or more, the mixed solvent has a Hansen solubility parameter distance Ra from dihydroterpineol of 3.0 MPa.sup.1/2 or less, and the metal paste further comprises a high molecular weight ethyl cellulose having a number average molecular weight of 40000 to 90000 as a first additive, and a polyvinyl acetal resin as a second additive.

2. The metal paste according to claim 1, wherein the mixed solvent has a Hansen solubility parameter distance Ra from dihydroterpineol of 2.5 MPa.sup.1/2 or less.

3. The metal paste according to claim 1, wherein the solvent B is an organic solvent having a Hansen solubility parameter distance Ra from dihydroterpineol of 3.0 MPa.sup.1/2 or less and containing two or more ester groups in a structure thereof.

4. The metal paste according to claim 3, wherein the solvent B has a Hilderbrand solubility parameter SP value of 8.5 (cal/cm.sup.3).sup.1/2 or more and 9.5 (cal/cm.sup.3).sup.1/2 or less.

5. The metal paste according to claim 3, wherein the mixed solvent has a Hansen solubility parameter distance Ra from dihydroterpineol of 1.0 MPa.sup.1/2 or less.

6. The metal paste according to claim 1, wherein the solvent B has a Hansen solubility parameter distance Ra from dihydroterpineol of more than 3.0 MPa.sup.1/2.

7. The metal paste according to claim 6, wherein the solvent B is at least either of texanol and butyl glycol acetate.

8. The metal paste according to claim 1, wherein a content of the high molecular weight ethyl cellulose that is the first additive is 1.0% by mass or more and 3.0% by mass or less in terms of a mass ratio with respect to the whole metal paste.

9. The metal paste according to claim 1, wherein a content of the second additive is 10% by mass or more and 70% by mass or less based on a content of the first additive.

10. The metal paste according to claim 1, wherein the amine compound that is the protective agent is butylamine, 1,4-diaminobutane, 3-methoxypropylamine, pentylamine, 2,2-dimethylpropylamine, 3-ethoxypropylamine, N,N-dimethyl-1,3-diaminopropane, 3-ethoxypropylamine, hexylamine, heptylamine, N,N-diethyl-1,3-diaminopropane, benzylamine, isobutylamine, or 3-isopropoxypropylamine.

11. The metal paste according to claim 2, wherein the solvent B is an organic solvent having a Hansen solubility parameter distance Ra from dihydroterpineol of 3.0 MPa.sup.1/2 or less and containing two or more ester groups in a structure thereof.

12. The metal paste according to claim 4, wherein the mixed solvent has a Hansen solubility parameter distance Ra from dihydroterpineol of 1.0 MPa.sup.1/2 or less.

13. The metal paste according to claim 2, wherein the solvent B has a Hansen solubility parameter distance Ra from dihydroterpineol of more than 3.0 MPa.sup.1/2.

14. The metal paste according to claim 2, wherein a content of the high molecular weight ethyl cellulose that is the first additive is 1.0% by mass or more and 3.0% by mass or less in terms of a mass ratio with respect to the whole metal paste.

15. The metal paste according to claim 3, wherein a content of the high molecular weight ethyl cellulose that is the first additive is 1.0% by mass or more and 3.0% by mass or less in terms of a mass ratio with respect to the whole metal paste.

16. The metal paste according to claim 4, wherein a content of the high molecular weight ethyl cellulose that is the first additive is 1.0% by mass or more and 3.0% by mass or less in terms of a mass ratio with respect to the whole metal paste.

17. The metal paste according to claim 5, wherein a content of the high molecular weight ethyl cellulose that is the first additive is 1.0% by mass or more and 3.0% by mass or less in terms of a mass ratio with respect to the whole metal paste.

18. The metal paste according to claim 6, wherein a content of the high molecular weight ethyl cellulose that is the first additive is 1.0% by mass or more and 3.0% by mass or less in terms of a mass ratio with respect to the whole metal paste.

19. The metal paste according to claim 7, wherein a content of the high molecular weight ethyl cellulose that is the first additive is 1.0% by mass or more and 3.0% by mass or less in terms of a mass ratio with respect to the whole metal paste.

20. The metal paste according to claim 2, wherein a content of the second additive is 10% by mass or more and 70% by mass or less based on a content of the first additive.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0080] FIG. 1 is a diagram illustrating an appearance of a wiring pattern for evaluating printing drawability and continuous printability of a metal paste;

[0081] FIG. 2 shows photographs of appearances of wirings printed with metal pastes using, as a solvent, butyl glycol acetate and dihydroterpineol in a preliminary test; and

[0082] FIG. 3 shows photographs of appearances of wirings (solvent: dihydroterpineol) having printing blur in an evaluation test for the continuous printability of First Embodiment.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0083] First Embodiment: A preferred embodiment of the present invention will now be described. In the present embodiment, as preliminary study, metal pastes having the same configuration as that of the conventional technique were first produced using various solvents (single solvents) to confirm whether or not printing drawability thereof were suitable. This preliminary study is conducted for confirming deviation of the width of a formed wiring from a designed width for presuming the cause. In the present embodiment, referring to results of the preliminary study, a plurality of solvents were selected based on the Hansen solubility parameter to produce metal pastes, and printing drawability and continuous printability thereof were evaluated.

Production of Metal Paste

[0084] First, a silver particle having the particle size distribution of the present invention was produced by a pyrolysis method. A wet state material obtained by adding 102.2 g of silver carbonate (silver content: 80.0 g) to 37.3 g of water (36.4% by weight with respect to 100 parts by mass of the silver carbonate) was used as a silver compound raw material. To this silver compound, 3-methoxypropylamine, used as an amine compound of a protective agent, was added in a 6-fold amount in terms of a molar ratio to the mass of silver to produce a silver-amine complex.

[0085] With a water content in this silver-amine complex adjusted, a homogenizing agent (DMF) was further added thereto. In this reaction system, a decomposition temperature was set to 130? C., and heating was performed from room temperature (at a heating speed of 10? C./min) to decompose the silver-amine complex at 130? C., and thus, silver particles were precipitated. The heating was continued until generation of carbon dioxide was stopped. In this heating step, a liquid in which the silver particles were suspended was obtained. The reaction solution was washed with methanol added thereto, and the resultant was centrifuged to collect the silver particles. The silver particles had an average particle size of 100 nm, and included silver particles having a particle size of 100 to 200 nm in a ratio of 80% or more in terms of the number of particles.

[0086] The thus obtained silver particles were added as a solid content to the solvent, and a high molecular weight ethyl cellulose (manufactured by Dow Chemical, ETHOCEL? STD100 (number average molecular weight: 63420)) used as the first additive was further added thereto and the resultant was kneaded to obtain a metal paste. The content of the silver particles corresponding to the solid content was 70% by mass, and the content of the ethyl cellulose (STD100) was 1.7% by mass with respect to the whole metal paste. In this preliminary test, texanol (trade name: Nikko NG-120), butyl glycol acetate, dihydroterpineol, and terpineol were used as the solvent, and four types of metal pastes were produced to have the same mixing ratio of the solvent. The solvents were all commercially available products. Commercially available dihydroterpineol and terpineol are presumed as mixtures of structural isomers, but the proportions thereof are not clear. In the present embodiment, as the respective parameters (?d, ?p, and ?h) of the HSPs of dihydroterpineol and terpineol, values of compounds in a form were used to calculate the distance Ra or the like. Here, the configuration of the metal paste was 70% by mass of the solid content (silver particles), 2% by mass of the high molecular weight ethyl cellulose, and a balance of the solvent. In this preliminary test, in order to focus on study of the relationship between the solvent and the printing drawability, a polyvinyl acetal resin was not added.

Confirmation of Printing Drawability of Conventional Technique (Preliminary Test)

[0087] Each of the metal pastes described above was applied/printed in a linear wiring pattern by a screen printing method. The metal paste was printed on a glass substrate from above a screen mask. The printing was performed on a glass substrate with a screen printing device (manufactured by Newlong Seimitsu Kogyo Co., Ltd., LS-150) using a screen mask having a pattern illustrated in FIG. 1 (linear patterns having designed widths of 50 ?m, 100 ?m and 200 ?m) at a printing speed of 50 mm/sec. Thereafter, the resultant was fired at 150? C. for 30 minutes to form linear metal wirings. Then, widths of the thus formed three types of wirings were measured for obtaining an average to calculate deviation. Results of this preliminary test are shown in Table 1. Besides, FIG. 2 illustrates photographs of appearances of the wirings formed with the metal pastes using, as a solvent, butyl glycol acetate and dihydroterpineol.

TABLE-US-00001 TABLE 1 Designed Wiring width/ width/ Deviation/ No. Solvent ?m ?m ?m 1 Texanol 50 92.1 42.1 2 (Nikko NG-120) 100 136.9 36.9 3 200 235.9 35.9 4 Butyl glycol acetate 50 88.1 38.1 5 100 142.0 42.0 6 200 240.9 40.9 7 Dihydroterpineol 50 56.2 6.2 8 100 102.2 2.2 9 200 209.8 9.8 10 Terpineol 50 65.1 15.1 11 100 113.6 13.6 12 200 210.5 10.5

[0088] As a result of this preliminary test, it is confirmed that the metal pastes using the same type of solvent have deviation of substantially the same value regardless of the designed width. This value of the deviation varies depending on the solvent, but the values are liable to be the same with respect to the respective solvents as described above. Based on this result, the present inventors presume that the behavior of a solvent is largely involved in the printing drawability.

[0089] Besides, it was confirmed in this preliminary test that dihydroterpineol is the optimum solvent from the viewpoint of the printing drawability. In addition, it is deemed that terpineol is a preferable solvent second to dihydroterpineol in the printing drawability. In the embodiment of the present invention, as an acceptance criterion for printing drawability, deviation of 20 ?m or less was determined as good drawability (acceptable=good) regardless of the designed width. The acceptance criterion was determined as 20 ?m or less referring to terpineol, a known solvent which was comparatively good in deviation (about 12 to 13 ?m on average) in the preliminary test, and because in consideration of forming of a wiring with a width less than 50 ?m that can be required in the future, it is presumed that the deviation needs to be less than a half of this wiring width.

Production/evaluation of Metal Paste of Embodiment using Mixed Solvent

[0090] In consideration of the results of the preliminary test, the present inventors decided to produce/evaluate a metal paste using, as a solvent, a mixed solvent of dihydroterpineol as the solvent A and the solvent B.

Selection of Solvent B

[0091] In the present embodiment, as the solvent B contained in the mixed solvent together with the solvent A, organic solvents shown in Table 2 were used. Table 2 shows the boiling point, the respective components of the Hansen solubility parameter (the dispersion term ?d, the polar term ?p, and the hydrogen bond term ?h), the HSP distance Ra from dihydroterpineol, and the HSP SP value of each solvent.

TABLE-US-00002 TABLE 2 Hansen solubility parameter Boiling point/ Number of Ra/ SP/ Solvent No. Name of solvent ? C. ester groups ?d ?p ?h MPa.sup.1/2 (cal/cm.sup.3).sup.1/2 A-1 Dihydroterpineol 208.9 16.7 3.5 6.7 8.94 B-1 Diisobutyl adipate 293.0 2 16.7 2.5 6.2 1.12 8.78 B-2 Glycerol tributyrate 305.0 3 16.3 2.5 7.0 1.32 8.74 B-3 Acetylcitric acid triethyl ester 355.7 4 16.6 3.5 8.6 1.91 9.28 B-4 Glycerol triacetate 259.0 3 16.5 4.5 9.1 2.63 9.45 B-5 Dibutyl maleate 280.0 2 16.5 6.1 7.2 2.68 9.27 B-6 Dibutyl sebacate 349.0 2 16.7 4.5 4.1 2.79 8.67 B-7 Diethyl adipate 291.3 2 16.4 6.2 7.5 2.88 9.31 C-1 Texanol 248.0 15.1 6.1 9.8 5.16 9.27 (Nikko NG-120) * Ra: HSP distance from dihydroterpineol.

[0092] As the production step of each metal paste of the present embodiment, the same silver particles as those used in the preliminary test were used, the silver particles were mixed with the solvent A, and then the solvent B was mixed therewith to obtain the metal paste. Besides, as ethyl cellulose corresponding to the first additive, the same as that described above was added in the same amount, and as the polyvinyl acetal resin corresponding to the second additive, a polyvinyl butyral resin (manufactured by Sekisui Chemical Company, Limited., S-LEC BH-S) was added in an amount of 33% by mass with respect to the ethyl cellulose. These additives were added after mixing the silver particles with the solvents. Besides, a mixing ratio between the solvent A and the solvent B was set to a mass ratio of the solvent A: the solvent B of 6:1 (as for mixing ratio in terms of volume ratio, refer to Table 3 below).

[0093] After producing the metal pastes respectively using the various solvents B, the respective metal pastes were screen printed to evaluate the printing drawability and the continuous printability. In a test for evaluating the printing drawability, the same printing device as that used in the preliminary test was used to print the pattern of FIG. 1 at a printing speed of 50 mm/sec. After applying each metal paste by printing, a firing treatment was performed at 150? C. for 30 minutes to obtain metal wirings. As the evaluation of the printing drawability, the deviation was measured in all the thus formed metal wirings (3 types of widths?3 wirings) after the printing step, and when the average of the deviation was 20 ?m or less from the designed width, the drawability was determined as good. Alternatively, when the average of the deviation was more than 20 ?m, the drawability was determined as poor.

[0094] In a test for evaluating the continuous printability, a screen mask on which 200 slits (wiring pitch: 0.3 mm) with a line width of 20 ?m (length: 100 mm) were formed as a wiring pattern was used. Each of the metal pastes was used for performing continuous 100 printing operations. A printing device used here was the same as that used in the preliminary test and the like. In the evaluation of the continuous printability, during the continuous printing, wirings formed in the 3rd, 20th, 40th, 60th, and 100th printing operations were observed in optical microscope images. In the observation performed after each printing operation, 3 wirings were optionally selected from the printed 200 wirings, and the whole lengths of these were observed to check whether or not the wirings had disconnection. Then, when any one of the selected wirings had 2 or more disconnection portions, it was determined that printing blur was caused. A metal paste with which printing blur was not caused even in the 100th printing operation was evaluated to have good continuous printability, and a metal paste with which printing blur was caused before the 100th printing operation was evaluated to have poor continuous printability. The results of the printing drawability and the continuous printability obtained by the printing test are shown in Table 3.

TABLE-US-00003 TABLE 3 Printing drawability Continuous Mixing ratio (volume ratio) Ra/ Average printability Paste No. Solvent A Solvent B Solvent A Solvent B MPa.sup.1/2 deviation/?m Evaluation Evaluation P-1 Dihydroterpineol Diisobutyl adipate 0.85 0.15 0.17 11.63 good good P-2 Glycerol tributyrate 0.86 0.14 0.18 4.19 good good P-3 Acetylcitric acid triethyl ester 0.87 0.13 0.25 10.73 good good P-4 Glycerol triacetate 0.87 0.13 0.33 9.41 good good P-5 Dibutyl maleate 0.86 0.14 0.38 11.96 good good P-6 Dibutyl sebacate 0.85 0.15 0.42 11.96 good good P-7 Diethyl adipate 0.86 0.14 0.41 13.13 good good P-8 Dihydroterpineol 1.00 0.00 8.94 good poor P-9 Texanol 1.00 5.16 49.66 poor good (Nikko NG-120) * Ra: HSP distance of mixed solvent from dihydroterpineol.

[0095] It was confirmed, based on Table 3, that the mixed solvents used in the present embodiment had a HSP distance Ra from dihydroterpineol of 1.0 MPa.sup.1/2 or less, and that the metal pastes obtained therefrom exhibited extremely good printing drawability (P-1 to P-7). It is deemed that the metal paste using the single solvent of dihydroterpineol (P-8) was the best in the printing drawability. The metal paste using the single solvent of dihydroterpineol was, however, poor in the continuous printability. FIG. 3 illustrates photographs of printing blur found in the wirings formed with the paste P-8 (using the single solvent of dihydroterpineol). In using the metal paste containing only dihydroterpineol as the solvent, wiring disconnection started to be caused in a plurality of portions at the time of the 3rd printing operation, and in the 100th printing operation, the wiring was blurred and dotted with island-shaped metals. In using texanol, that is, the solvent of the conventional metal paste (P-9), although the continuous printability was favorable, the printing drawability was largely inferior. This result matches the result of the preliminary test.

[0096] Next, an evaluation test on adhesion of wirings obtained from the respective metal pastes was performed. As for the adhesion, some of the metal pastes, that is, a paste P-2 (solvent B: glycerol tributyrate), a paste P-3 (solvent B: acetylcitric acid triethyl ester), and the paste P-8 (single solvent of dihydroterpineol) were evaluated. In the evaluation of these three metal pastes, ones not containing the second additive of the polyacetal resin were produced for comparison.

[0097] In the evaluation test for the adhesion, a screen printing device similar to that described above was used to print the same wiring pattern (once). A firing treatment after the printing was performed at three types of firing temperatures of 60? C., 80? C., and 100? C., and metal wirings produced at the respective firing temperatures were evaluated for adhesion. The metal pastes containing no polyacetal resin as comparative examples were sintered only at the firing temperature of 100? C. The adhesion of the printed wirings was evaluated by a peel test. In the peel test, an adhesive tape (ASTM D3359 Cross Hatch Adhesion Test Tape) was adhered to a region having a wiring formed on a substrate, and was then peeled all at once. After peeling the adhesive tape, when all the wirings were not peeled, the adhesion was determined as good. When even one of the wirings was peeled, the adhesion was determined as poor. Evaluation results of the adhesion of the wirings are shown in Table 4.

TABLE-US-00004 TABLE 4 Firing Second temperature/ Paste No. Solvent A Solvent B additive ? C. Adhesion P-2 Dihydroterpineol Glycerol tributyrate used 60 good 80 good 100 good not used 100 poor P-3 Acetylcitric acid used 60 good triethyl ester 80 good 100 good not used 100 poor P-8 Dihydroterpineol used 60 good 80 good 100 good not used 100 poor

[0098] It was confirmed, based on Table 4, that the adhesion of the metal wiring is largely improved by adding the polyvinyl acetal resin to the metal paste as the second additive. It was also found that the adhesion of the metal wiring is favorable even if the firing temperature is comparative low (60? C.). The peel test of the present embodiment is a test method in which a metal film is forcedly peeled, and shows comparatively severe determination result on the adhesion. It was confirmed that the polyacetal resin used as the second additive can ensure adhesion of a wiring even under such a severe condition.

[0099] Second Embodiment: In the present embodiment, metal pastes containing mixed solvents using, as the solvent B, organic solvents having comparative large HSP distances Ra from dihydroterpineol (Ra: more than 3.0 MPa.sup.1/2) were produced/evaluated. Metal pastes obtained by adjusting the mixing ratio in the mixed solvent of the solvent B (Ra: 3.0 MPa.sup.1/2 or less) evaluated in First Embodiment were also studied. Organic solvents used in the present embodiment are shown in Table 5.

TABLE-US-00005 TABLE 5 Boiling point/ Ra/ SP/ Solvent No. Name of solvent ? C. MPa.sup.1/2 (cal/cm.sup.3).sup.1/2 A-1 Dihydroterpineol 208.9 8.94 A-2 Terpineol 219.0 1.10 9.27 B-2 Glycerol tributyrate 305.0 1.32 8.74 B-3 Acetylcitric acid 355.7 1.91 9.28 triethyl ester B-9 Nikko MARS 264.0 7.58 5.30 B-10 Texanol 248.0 5.16 9.27 (Nikko NG-120) * Ra is an HSP distance from dihydroterpineol.

[0100] Also in the present embodiment, the metal pastes were produced through the same steps and configurations as those employed in First Embodiment. Then, the printing test (for evaluation of the printing drawability and evaluation of the continuous printability) and the adhesion evaluation (peel test) were performed in the same manner as in First Embodiment. In the present embodiment, a PET substrate and a PEI substrate were used as substrates instead of the glass substrate used in First Embodiment, and the printing drawability and the like on the respective substrate materials were evaluated (only the adhesion was evaluated using a glass substrate). The dimension of each substrate and the printing pattern were the same as those of First Embodiment. Test results of the metal pastes of the present embodiment are shown in Table 6.

TABLE-US-00006 TABLE 6 Mixing ratio (volume ratio) Paste Solvent Solvent Ra/ Printing drawability Continuous No. Solvent A Solvent B A B MPa.sup.1/2 Glass PET PEI printability Adhesion P-10 Dihydroterpineol Nikko MARS 0.80 0.20 1.52 good good good good good P-11 Terpineol 0.80 0.20 2.32 good good good good good P-12 Dihydroterpineol Texanol 0.80 0.20 1.03 good good good good good P-13 Terpineol (Nikko NG-120) 0.50 0.50 2.77 poor good good good good P-14 0.20 0.80 4.18 poor poor poor good good P-15 Dihydroterpineol Glycerol tributyrate 0.86 0.14 0.18 good good good good good P-16 Dihydroterpineol Acetylcitric acid triethyl 0.87 0.13 0.25 good good good good good ester P-17 Dihydroterpineol 1.00 0.00 good good good poor good P-18 Terpineol 1.00 1.17 good good good poor good * Ra is an HSP distance of mixed solvent from dihydroterpineol. * Adhesion was evaluated on a wiring formed on a glass substrate at a firing temperature of 100? C.

[0101] It had been confirmed, based on the results of the preliminary test, that texanol is a solvent inferior in the printing drawability as the single solvent. Besides, it was predicted that Nikko MARS having a HSP Ra close to it was the same. It was, however, confirmed that these solvents can form a favorable mixed solvent when mixed as the solvent B with dihydroterpineol (solvent A). It is understood that when the HSP Ra of the mixed solvent adjusted by the mixing ratio of the solvent B is more than 3.0 MPa.sup.1/2, the printing drawability is reduced on all the materials of the base materials. When the Ra of the mixed solvent is 2.5 MPa.sup.1/2 or less, such a solvent is effective for both a glass substrate and a resin substrate, and even when it is 3.0 MPa.sup.1/2 or less, the drawability on a resin substrate is favorable. Glycerol tributyrate and acetylcitric acid triethyl ester, that is, the solvent B favorable from the viewpoint of the HSP Ra, do not largely increase the Ra of the mixed solvent even if the mixing ratio is increased, and exhibit favorable printing drawability and continuous printability. It is noted that the adhesion of a wiring was favorable in all owing to the addition of the polyacetal resin.

INDUSTRIAL APPLICABILITY

[0102] As described so far, a silver paste of the present invention is favorable in both drawability and continuous printability. The present invention enables a wiring pattern having a width of 100 ?m or less to be accurately formed. The present invention is favorable for formation of electrodes and wirings of electric/electronic devices such as an LED element, and a power semiconductor device.