JP3499588B2 - Method of manufacturing thick film electrical component on hybrid circuit substrate - Google Patents

Description

Description: BACKGROUND OF THE INVENTION The present invention relates to improvements in the field of hybrid circuit manufacturing. More specifically, the present invention relates to an improved method for manufacturing a hybrid circuit, wherein an Ag / Pd conductor is formed on a hybrid substrate. 2. Description of the Related Art With the trend toward higher density and higher integration of semiconductors such as ICs and LSIs, demands for higher performance and higher reliability of hybrid modules are increasing. Conventionally, Ag, as a conductor of a hybrid circuit,
Noble metals such as Pt, Au, Ag / Pd have been used.
These noble metal conductors are usually fired in an air atmosphere. Since there is only 21% oxygen in the air, it is necessary to allow sufficient firing time for materials that require burnout of the binder such as a dielectric. In the case of forming a multi-layer thick film, if the firing of the dielectric is insufficient, the leakage current becomes excessive, which causes a product failure. The inventor of the present invention has made it possible to reduce the leakage current of the Ag / Pd conductor used in the hybrid circuit and to increase the firing time of the dielectric to increase the productivity of the multilayer thick film. In order to keep the length as short as possible, the present inventors have conducted intensive studies and have found that by performing firing under specific conditions, the leakage current is reduced and other properties are not significantly deteriorated, leading to the present invention. The present invention relates to (a) applying an Ad / Pd conductor paste on a substrate, (b) firing the paste in a belt furnace, and removing oxygen in a burnout area of the firing. The oxygen concentration in the area is maintained above 40% by introduction, in the firing the object is kept at a temperature above 850 ° C. for at least 10 minutes, and (c) then an insulating layer is applied thereon. The present invention also relates to a method for producing a thick film electric component on a hybrid circuit substrate, comprising treating the insulator layer within the range described in the above (b). It is essential that the oxygen concentration in the burnout be 40% or more, but preferably 60% or more.
Most preferably, it is at least 75%. In firing,
If the object is not held at a temperature of 850 ° C. or higher for 10 minutes, the leakage current increases. In the present invention, the insulator layer (dielectric layer) is preferably applied twice after firing the conductor. The thickness of the insulator layer is 5
Since about 0 μm is required, coating is performed twice or more. After the application of the insulator layer, the insulator layer is fired within the range of firing conditions. That is, in firing the insulator layer, the oxygen concentration in the burnout is 40% or more, and in firing, the object is held at a temperature of 850 ° C. or more for 10 minutes or more. EXAMPLES Examples and Comparative Examples Materials used in the experiments were as follows. Substrate: 96% alumina, 1 ″ × 1 ″ or 2 ″ × 2 ″ Ag / Pd Conductor: DuPont 6134 Dielectric: DuPont 5704 Resistor: DuPont 1921 (100 ohm) DuPont 1939 (10 kohm) DuPont 1959 ( (1 Megohm) 3-1 Dielectric and Conductor For the preparation of the test sample, an Ag / Pd conductor was screen-printed on an alumina substrate (dimensions 2 ″ × 2 ″) using a semi-automatic printing machine. Fifty samples were prepared for each experiment. The substrate was left at room temperature for 5-10 minutes, and then dried in an infrared drying belt conveyor oven at 150 ° C. for 15 minutes. Next, the sample was heated in a heating zone having a length of 1 foot for 8 hours.
Calcination was performed in various atmospheres in a Lindberg 6 inch wide belt furnace. The furnace also includes an inlet and outlet purge chamber, and an atmosphere separator with a powered exhaust system between the burnout zone and the firing zone. The binder burnout compartment has a rake assembly that includes multiple alternating ports for incoming oxygen and outgoing volatiles, and the exhaust is powered by a venturi. The furnace is equipped to operate in nitrogen, air or oxygen. The main flow is regulated by a mass flow controller. Using this method, a total flow rate of 500 SCFH was reached. Dispensed as follows: inlet purge--80, outlet purge--40, burnout lake--220, separator--30, firing zone--70, and cooling zone--60. A schematic diagram of the muffle and piping of the furnace is shown in FIG. The temperature distribution in the furnace was examined using a strike thermocouple, and is shown in FIG. In FIG. 3, each temperature distribution is as follows. Temperature distribution 1: The belt speed is 3 inches / minute.
The maximum temperature is 852 ° C, and the maximum temperature zone exceeding 850 ° C is 6 minutes. Temperature distribution 2: The belt speed is 3 inches / minute.
The maximum temperature is 860 ° C, and the maximum temperature zone exceeding 850 ° C is 10 minutes. Temperature distribution 3: The belt speed is 6 inches / minute.
The maximum temperature is 870 ° C, and the maximum temperature zone exceeding 850 ° C is 10 minutes. After firing the conductor, an insulator layer was screen-printed on the conductor and dried at 150 ° C. A second insulator layer was screen printed over the first insulator layer and dried to a total thickness of about 2 mils. Next, change the conductor to Lindb
Both insulator layers were simultaneously fired in the same atmosphere as when firing at 150 ° C. in an erg furnace. The oxygen concentration distribution in the furnace was monitored using the strike pipe and the oxygen analyzer, and is shown in FIG. The two-layer printing / dried insulator layer was simultaneously fired under eight kinds of atmosphere conditions. Experiment 1 The furnace was operated using air as the atmosphere. The temperature distribution is shown as temperature distribution 1 in FIG. Experiment 2 The furnace was operated in an atmosphere doped with oxygen. Oxygen was introduced from the dopant gas inlet into the burnout and firing zones to a maximum oxygen concentration of 66%. The oxygen concentration distribution in the furnace is shown as the O ₂ dopant in FIG. The temperature distribution is shown as temperature distribution 1 in FIG. Experiment 3 The furnace was operated with oxygen doped nitrogen up to an oxygen concentration of 83%. Oxygen was introduced only from the dopant gas inlet to the burnout zone. The oxygen concentration distribution in the furnace is shown as O ₂ dopant + N ₂ cooling in FIG. The temperature distribution is shown as temperature distribution 1 in FIG. Experiment 4 The furnace was operated with oxygen as atmosphere and the inlet and outlet areas were purged with air. The oxygen concentration distribution in the furnace is shown as the O ₂ atmosphere in FIG. The maximum oxygen concentration is 90%, and the temperature distribution is shown as temperature distribution 1 in FIG. Experiment 5 The furnace was operated in an air atmosphere. The temperature distribution is shown as temperature distribution 2 in FIG. Experiment 6 The furnace was operated in an oxygen atmosphere and the inlet and outlet areas were purged with air. The oxygen temperature distribution in the furnace is shown as the O ₂ atmosphere in FIG. The maximum oxygen concentration is 90%, and the temperature distribution is shown as temperature distribution 2 in FIG. Experiment 7 The furnace was operated in an air atmosphere. The temperature distribution is shown as temperature distribution 3 in FIG. Experiment 8 The furnace was operated in an air atmosphere and oxygen was introduced only from the dopant gas inlet into the burnout zone. The oxygen concentration distribution in the furnace is shown as O ₂ dopant + air cooling in FIG. The maximum oxygen concentration is 75%, and the temperature distribution is shown as temperature distribution 3 in FIG. 3-2 Resistor To prepare a test sample, a semi-automatic printing machine was used to deposit Ag / P on an alumina substrate (dimension 1 ″ × 1 ″ or 2 ″ × 2 ″).
The d conductor was screen printed. Ten samples were prepared for each experiment. Next, these substrates were dried in an infrared drying belt conveyor oven, and Lindb described in Section 3-1 was used.
Fired in an erg furnace. By setting the temperatures of the individual zones, an in-furnace temperature distribution suitable for the generally favorable operation of heating, holding and cooling was obtained. FIG. 5 shows the temperature distribution. The maximum temperature is 870 ° C, and the time exceeding 850 ° C is 1
0 minutes. This temperature distribution was employed during the resistor experiment. After preparation of the conductor, the resistor is screen-printed,
It was dried and fired in the same atmosphere as when the conductor was fired in the Lindberg furnace. The resistor sample was fired under five types of atmosphere conditions. FIG. 6 shows the oxygen concentration distribution in the furnace. Experiment 9 The furnace was operated using an air atmosphere. Experiment 10 The furnace was operated in a nitrogen atmosphere and oxygen was introduced only from the dopant gas inlet to the burnout zone. The oxygen concentration distribution in the furnace is shown as O ₂ dopant + N ₂ cooling in FIG. The maximum oxygen concentration was 83%. Experiment 11 The furnace was operated in a nitrogen atmosphere. Experiment 12 The furnace was operated in an air atmosphere and oxygen was introduced only from the dopant gas inlet to the burnout zone. The oxygen concentration distribution in the furnace is shown as O ₂ dopant + air cooling in FIG. The maximum oxygen concentration was 75%. Experiment 13 The furnace was operated with oxygen, but the inlet and outlet compartments were
Purged with air. The oxygen concentration distribution in the furnace is shown as the O ₂ atmosphere in FIG. The maximum oxygen concentration was 90%. Test 4-1 Conductor To prepare a test sample, a conductor for an adhesion test was screen-printed on a dielectric. The substrate was then dried in an infrared drying belt conveyor oven at 150 ° C. for 15 minutes. In addition, the samples were fired in the same eight atmospheres as when firing the two-layer print / dry insulator layer simultaneously in Experiments 1-8. The adhesive force , expressed as tensile force in kg, is 63 Sn
By immersion soldering in a / 37Pb solder alloy, 8
Measured by pulling a 20 gauge copper wire soldered to a 0x80 mil pad. Table 1 shows the test results. [Table 1] 4-2. Tests were performed on measurements of important functional properties that may be affected by the dielectric atmosphere, namely leakage current, dielectric constant and breakdown voltage. A surface conductor was screen printed on the dielectric for relative permittivity and breakdown voltage tests. Next, the test sample was dried and fired under the same atmosphere conditions as when the dielectric was fired simultaneously. Leakage current : Ag / Pd on the test sample was used for one electrode, and platinum foil was used for the other electrode.
The leakage current was measured using an electrolytic cell immersed in an aCl solution. Leakage current was measured by applying a potential of 10 volts between the electrodes. The test method is shown in FIG. Dielectric constant K Hewlett Packard type 4192A LCR
The relative dielectric constant K was measured using a meter. All measurements of capacitance were performed at room temperature and at a frequency of 10 KHz. The relative permittivity was calculated using the following equation. K = (0.95 CT) / (EA) where C = measured capacitance (Farad) T = dielectric thickness (inches) E = dielectric constant of free space, 2.249 × 10 Farad / inch a = measured DC breakdown voltage using 4045 type hypot tester correlation coefficient breakdown voltage the Associated Research of surface area 0.95 = edge effect of the capacitor plates.
Next, the sample was immersed in silicone oil to suppress arcing at the time of destruction. The voltage was increased at a rate of about 100 volts / second until failure occurred. 4-3 Resistor A test for measuring the resistivity and TCR of the resistor was performed. The resistance is measured using a Keithley multimeter in an environmental chamber at temperatures of −55, 25 and 125 ° C. and using these values, the coefficient of change C
V, HTCR (high temperature TCR) and CTCR (low temperature TC)
R) was calculated. 5. Test results of adhesion test results and Discussion 5-1 conductor Ag / Pd conductors are shown in Table 1. Kg
Adhesive force, expressed in units of tensile force, was measured by pulling a wire soldered by immersion soldering. Under the conditions of temperature distributions 2 and 3, the adhesion was slightly improved. The highest temperature zones of temperature distributions 2 and 3 (> 850 ° C.) are longer than temperature distribution 1. The results of experiments 5 to 8 show that adhesion is not related to oxygen concentration. 5-2 Dielectric Table 2 shows the test results of the two-layer adhered insulator (derivative) layer. [Table 2] Experiments 6 and 8 show significantly better leakage current values.
Low leakage current indicates good airtightness, a desirable feature. Experiment 5 is a 68 minute cycle in an air atmosphere. Experiment 6 is a 68 minute cycle in an oxygen atmosphere.
The leakage current value of 17 μA / cm ² in Experiment 6 was 58 in Experiment 5
Much better than μA / cm ² . Experiment 7 is a 34 minute cycle in an air atmosphere. Experiment 8 are introduced and 34 minute cycle to burnout zone of O ₂ dopant. In both experiments, the highest temperature zone (> 850 ° C.) was 10 minutes. In Experiment 8, the leakage current value was 5.8 μA /
cm ² , surprisingly improved. These results show that oxygen is effective for air fired silver / palladium dielectrics. 5-3 Resistors Tables 3 to 5 compare the average values of resistivity and TCR obtained under various atmospheric conditions. [Table 3] [Table 4] [Table 5] There is a clear trend in the data as a function of atmosphere composition. The value for air atmosphere were all better than the case where the introduction / N ₂ cooling or N ₂ atmosphere to burn out the O _2. In the case of introducing O _{2 into} the burnout / air cooling or in a O ₂ atmosphere, the resistivity and the TCR value of the resistor processed in the O ₂ atmosphere are close to those of the resistor processed in the air atmosphere. However, compared to the air atmosphere,
No favorable effect was observed for the resistor under these conditions. The resistor DuPont 1921 (107
Ohms) air introduction / air cooling to burnout of O _2, and in the case of an atmosphere of O ₂ values of resistivity of less than 10% of the design value is obtained. However, the CV value was best in an air atmosphere. Except for the N ₂ atmosphere, the HTCR value was less than 50 under these atmosphere conditions. In the case of CTCR, only the value in the case of the air atmosphere was less than 50. The resistor DuPont 1934 (10.
(4 kOhm) The resistivity in the O ₂ atmosphere was close to the value in the air atmosphere, but lower than in the atmosphere of O ₂ introduction to burnout / air cooling. Introducing O ₂ into burnout / air cooling or C processed in O ₂ atmosphere
The V value was better than the value obtained when processing was performed in an air atmosphere. Both TCR values are smallest in an air atmosphere. The resistor DuPont 1959 (1.0
The resistivity values in an air atmosphere ( 2 megohms) were closer to the design values than those made under other atmosphere conditions. Both TCR value when introducing / air cooling of O ₂ to burnout was close to the value of having a CV value close to the value in the case of an air atmosphere, also capable generally acceptable.

[Brief description of the drawings] FIG. 1 is a schematic diagram of a muffle and an atmosphere pipe in a furnace. FIG. 2 is a schematic diagram of a test airtight electrolytic cell. FIG. 3 is a graph showing a temperature distribution. FIG. 4 is a graph showing an oxygen concentration distribution. FIG. 5 is a graph showing a temperature distribution. FIG. 6 is a graph showing an oxygen concentration distribution.

──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Shigeki Hayashi 2-13-309, Daisennakacho, Sakai City, Osaka (72) Inventor Mark Jay Kirshner, New Jersey, USA 07960, Morristown, Fairmount Avenue 26 ( 72) Inventor Rudolf Wolf, New Jersey, USA 07060, North Plainfield, Harley Avenue 11 (72) Inventor Satish es Tamhanker, New Jersey, USA 07076, Scotch Plains, Argonquin Drive 2011 (56) References JP-A-60-3149 (JP, A) JP-A-2-126700 (JP, A) JP-A-1-235291 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) H05K 3/10-3/38

Claims (1)

(57) [Claim 1] (a) applying an Ag / Pd conductor paste on a substrate (b) baking of the paste is performed in a belt furnace, and in a burnout area of the baking, Oxygen is introduced to keep the oxygen concentration in the area above 40%, in the firing the object is kept at a temperature above 850 ° C. for at least 10 minutes, and (c) then an insulating layer is placed thereon. A method of manufacturing a thick film electrical component on a hybrid circuit substrate, comprising applying and treating the insulator layer within the conditions described in (b) above.