JPS6322043B2 - - Google Patents

Description

[Detailed description of the invention]

The present invention relates to a method of manufacturing a carbide film resistor by thinning a carbide film resistor material by sputtering. It is well known that this type of sputtering has been widely used as a method for manufacturing electronic components such as resistors and capacitors by thinning various materials such as conductors, dielectrics, and semiconductor materials. That's true. My theory is that the basic characteristics are determined by the resistor material, and various materials are selected as the target material depending on the purpose. In addition, the process of forming a film resistor is also
It has important factors in determining film quality, resistance characteristics, adhesion amount, specific resistance, etc. In sputtering, sputtering gas pressure, substrate temperature, sputtering power,
Sputtering time and gas atmosphere are the most important among various parameters during sputtering. Particular attention must be paid to the purity of the target material and the purity of the sputtering gas. Sputtering is a phenomenon in which ionized gas molecules are accelerated by an electric field and collide with a target electrode, releasing target molecules.The gas atmosphere used is usually
It is an argon gas atmosphere of 10 ^-1 to ^{10 -3} Torr.
The substrate temperature is usually maintained at an appropriate temperature in order to obtain good adhesion. Temperature required to remove water and organic matter from the substrate surface (ex100℃ or more),
A temperature at which the expansion coefficients of the substrate and the film are close to each other, a temperature at which the compound does not decompose, etc., are considered and selected. The sputtering power directly contributes proportionally to the amount of deposition per unit time, but beyond a certain point, the energy of the ions contributing to sputtering becomes saturated, so it is usually
~5kw. The sputtering time is determined depending on the desired film thickness. The above-mentioned high-purity argon gas is usually used as the gas atmosphere. This is because the material used for the target and an inert atmosphere are required to obtain the same product film material. Generally, in this way, evaporative gases and residual gases from the sputtering chamber walls and jigs, especially gases having active effects, are suppressed and avoided as impurities as much as possible. For example, when creating a silicon carbide thermistor (NTC) by sputtering, the conventional method uses a sintered body of silicon carbide as the target material, uses a high frequency power of 2.0Kw at an arbitrary substrate setting temperature, and sputtering gas pressure ×
The gas atmosphere was 99.9999% pure argon gas at 10 ^-2 Torr, and sputtering was carried out for 4 to 8 hours. With this method, the specific resistance was large, and it was difficult to suppress the effects of residual gas in the sputtering chamber, making it extremely difficult to stabilize the basic characteristics of thermistors, that is, the resistance value and thermistor constant. .
Furthermore, since the sputtering time dependence of the resistance value changes the resistance temperature characteristic depending on the sputtering time, it has been difficult to obtain a lower resistance value. That is, there was a drawback that the resistance value was not necessarily inversely proportional to the film thickness. Furthermore, the shape of the element (drawing) for a given resistance value is
Since it is determined by the relationship between the film thickness and specific resistance of the film, if the specific resistance is large, the practical resistance value range (50
In the case of temperature measurement: 1 to 1000 kΩ), it was difficult to obtain a miniaturized element. The same is true for other carbide film resistors, and when obtaining a low resistance value using the conventional method, the sputtering time is extremely long. As a result, a large amount of materials, energy, and working time are required, resulting in high costs. The present invention provides a novel method for manufacturing a carbide film resistor that eliminates these conventional drawbacks. The gist of the present invention is a method of sputtering a carbide resistor material by adding a small amount of impurity gas to a rare gas in a sputtering gas atmosphere, wherein the impurity gas is nitrogen, carbon dioxide, carbon monoxide, or a group thereof. A carbide film resistor characterized by using at least one kind of mixed gas selected from the above as an impurity gas, and the resistance value (or specific resistance) of the resulting film being smaller than that when no impurity gas is added. It's in the way the body is manufactured. Hereinafter, the present invention will be explained in detail with reference to Examples. Example 1 The configuration of the resistor element is as shown in the drawing, and the insulating substrate on which the resistor film is formed is an alumina substrate with a purity of 96%.
1t=0.65mm was used. Next, on the surface on which the resistor film is to be formed, an electrode 2 is formed using a conductive paste such as Ag.Au.Ag-Pd.Au-Pt. This 2-electrode pattern is configured in a geometric pattern, consisting of 2 electrodes with an effective width of 2.00 mm, and 1 electrode of the same size facing each other, and these 3 electrodes 2
However, the distance between adjacent electrodes was 0.30 mm, and the area between the electrodes 2 was 1.20 mm ² . A desired resistor film is formed on the surface of the substrate configured in this manner. In the following sputtering experiment, the substrate prepared above was used as a test piece. The sputtering device is a high-frequency two-pole type with a vacuum chamber.
A general-purpose type consisting of 350φ x 250hmm was used. The setting conditions for sputtering were high frequency power of 2.0Kw, sputtering time of 2.0Hrs, substrate temperature of 650°C, and sputtering pressure of ×10 ^-2 Orr. In advance, evacuation is performed to a sputter vacuum degree of ×10 ^-6 Torr, and then the amount of impurity gas is at least within the range of ~2.0 vol% with respect to the rare gas ×
A predetermined amount is introduced up to 10 ^-4 Torr. Next, as the target material, a sintered body of silicon carbide was selected as the resistor material. 99.9999 for noble gases
% argon gas and 99.9999% nitrogen gas were selected as the impure gas, and a constant amount was introduced while changing the partial pressure ratio, and sputtering was performed at a sputtering pressure × 10 ^-2 Torr. The resistance characteristics of the temperature-dependent silicon carbide film resistor thus obtained were measured in an oil bath at 50°C. The results are shown in the table below. Table No.1
The conventional method was to use only argon gas as the sputtering gas atmosphere (referred to as a blank). No.1
~8 is the effect when using nitrogen as an impure gas,
Nos. 9, 10, and 11 were created using the same method as No. 5, replacing the rare gas with 99.99% pure xenon, 99.99% pure neon, and 99.99% pure krypton gas. Next, using the same method as for silicon carbide described above, sputtering was carried out using boron carbide as the target material, that is, the carbide resistor material, and introducing a certain amount of a trace amount of nitrogen gas. The results are shown in Nos. 12 to 15 of the table. No. was carried out using the conventional method, using only argon gas as the sputtering gas atmosphere (referred to as a blank). Nos. 12 to 15 use argon, xenon, neon krypton gas as rare gases, and
It was created using the same method as No.5. As is clear from this, by adding a trace amount of nitrogen into a rare gas such as argon, xenon, neon, or krypton, a very small resistance value and specific resistance can be obtained. That is, the amount of nitrogen gas relative to the rare gas is at least 0 to
By selecting a range of 2.0vol%, the specific resistance can be reduced to 3200~
Characteristics can be controlled over a wide range of 3ohm-cm. For each sample prepared under the conditions No. 1 to 15 in the table, high temperature storage test (1000 hours left at 350℃), thermal shock resistance test (1 cycle of 15 minutes at room temperature to 15 minutes at 350℃ x 3000) (times). As a result, the resistance change rate was mostly within ±6%, as shown in the table.
No. There was no difference from the blank. Also, No. 8, 9, 10, 11, 12, 13,
Structural analysis was performed for each sample No. 14 and No. 15. As a result, X-ray diffraction showed a characteristic strong absorption peak of β-SiC at 2θ = 35.6°, and backscattered electron diffraction also revealed a β-SiC characteristic strong absorption peak at 2θ = 35.6°.
It was confirmed that it was SiC. It was also confirmed that when boron carbide was used, B ₄ C had a rhombohedral crystal structure. This had the same crystal structure as each sample numbered in the table. Furthermore, the presence of nitrides was not recognized in elemental analysis. There was also almost no difference in film thickness within the range of 2 μm±8%.

[Table] Example 2 Based on Example 1, an experiment was conducted by replacing nitrogen with carbon dioxide as the impure gas. Carbon dioxide has a purity of 99.99%, and other target materials,
The rare gas and the like are the same as in Example 1. The results are shown in the table. The numbers in the table are the same as the numbers in the table where samples were prepared using only argon gas using the conventional method. When carbon dioxide was used as the impure gas, its contribution to resistance characteristics was smaller than that of nitrogen in the example, but the pattern in which the concentration of carbon dioxide was dependent on resistance characteristics, etc. It was the same as nitrogen. That is, the amount of carbon dioxide relative to rare gas is
By selecting a range of at least 0 to 1.5 vol.%, the specific resistance can be controlled over a wide range of 3200 to 677 ohm-cm. A high temperature storage test and a thermal shock resistance test were conducted in the same manner as in Example 1 for each sample prepared using the method using carbon dioxide shown in the table. As a result, the resistance change rate was almost within ±6%, and there was almost no difference from the blank made by the conventional method. Also, Table-No. 1, 9, 10, 11, 12, 13, 14,
For each sample 15 and 16, the same analysis as in Example 1,
I did an analysis. As a result, even when carbon dioxide was used, structural analysis showed that it was β-SiC or B ₄ C having a rhombohedral crystal structure, and the presence of impurities was not recognized from elemental analysis. The film thickness was also within the range of 2 μm±7%, and there was almost no difference from the blank of the conventional method.

【table】

[Table] Example 3 Based on Example 1, an experiment was conducted by replacing nitrogen with carbon monoxide as the impure gas. Carbon monoxide uses 99.99% purity, and other target materials,
The rare gas and the like are the same as in Example 1. The results are shown in the table below. The No. of the table is
This is by the conventional method and is the same as the blank number in the table. When carbon monoxide was used as the impurity gas, its contribution to resistance characteristics was smaller than that of nitrogen in Example 1, but the pattern in which the concentration of carbon monoxide was dependent on resistance characteristics, etc. , was similar to nitrogen. Moreover, the pattern was closest to that of Example 2. That is, by selecting the amount of carbon monoxide relative to the rare gas in the range of at least 0 to 1.5 vol.%, the resistivity can be controlled over a wide range of 3200 to 749 ohm-cm. Next, for each sample prepared by the method using carbon monoxide shown in the table, in the same manner as in Example 1,
A high temperature storage test and a thermal shock resistance test were conducted. As a result, the resistance change rate was almost within ±6%, and Examples 1 and 2
obtained similar results. In addition, the same analysis as in Example 1 was performed for each sample No. 1.8 to 11 and No. 12 to 15 shown in the table.
As a result, even when carbon monoxide was used, structural analysis showed that it had a β-SiC or rhombohedral crystal structure.
It was B ₄ C, and elemental analysis did not show the presence of any impurities. The film thickness was also within the range of 2 μm±7%, and there was almost no difference from that of the conventional method.

[Table] Example 4 Based on Example 1, an experiment was conducted by replacing nitrogen as the impure gas with a gas mixed in a predetermined amount. The concentration of the mixed gas was adjusted in advance to 78.50% nitrogen, 21.45% oxygen, and 0.05% carbon dioxide. It had a composition similar to that of air. In addition, the target material, rare gas, etc. are as described in Example 1.
is the same as The results are shown in the table below. Table No.
, is based on the conventional method, and is No. in the table.
, is the same as the blank. impurity gas,
The resistance characteristics etc. when using the adjusted mixed gas are as follows.
A pattern relatively similar to that of nitrogen in Example 1 was shown.
In the case of mixed gases, various characteristic patterns are expected depending on the amount of adjustment. A high temperature storage test and a thermal shock resistance test were conducted in the same manner as in Example 1 for each sample prepared using the method using the mixed gas shown in the table. As a result, the resistance change rate is mostly within ±6%,
There was almost no difference from the blank of the conventional method. In addition, the same analysis as in Example 1 was performed for each sample No. 1, 6 to 9, and 10 to 13 shown in the table. As a result, even when mixed gas was used, structural analysis revealed that it had a β-SiC or rhombohedral crystal structure.
It was B ₄ C, and elemental analysis did not show the presence of any impurities. The film thickness was also within the range of 2 μm±7%, and there was almost no difference from that of the conventional method.

【table】

[Table] As can be seen from the above results, in the method of sputtering carbide resistor material, at least within the range where the carbide film obtained during sputtering or after sputtering has the same composition as the carbide resistor material, the sputtering gas atmosphere By adding trace amounts of nitrogen, carbon dioxide, carbon monoxide, or one or more impurity gases mixed at an arbitrarily selected concentration to the rare gas of It can be seen that a small amount can be obtained. This means that when obtaining the same resistance characteristics, a very compact resistance element can be made. In addition, the method of adding impure gas of the present invention allows sputtering to be performed in a short time, and controls such as resistance adjustment can be performed easily. The addition of impure gas had a great effect. Note that we have already applied for a method of manufacturing a carbide film resistor by adding a small amount of impurity gas, especially air or oxygen, to a rare gas in a sputtering atmosphere in Japanese Patent Application No. 151031/1982. The present invention is based on Japanese Patent Application No. 151031/1984 (1986) in that nitrogen, carbon dioxide, carbon monoxide, or a mixed gas of at least one selected from these groups is selected as the impure gas.
-22444 Publication) is obviously different. Furthermore, the conventional method called reactive sputtering is a method in which sputtering is performed by adding a large amount of impure gas such as oxygen or nitrogen to the sputtering atmosphere.
For example, using Si as a target, SiO ₂ , Si ₃ N ₄ films,
This is a method of forming. Therefore, the target material composition and the film composition are completely different. On the other hand, the method of the present invention only adds a small amount of impure gas.
Therefore, the target material composition and the film composition are the same, as shown in the examples. Therefore, it is clear that the so-called reactive sputtering method and the manufacturing method of the present invention are completely different.

[Brief explanation of drawings]

The drawing is a schematic diagram showing the structure of a carbide film resistance element obtained by the manufacturing method of the present invention. DESCRIPTION OF SYMBOLS 1... Alumina substrate, 2... Electrode, 3... Resistance film formation part.

Claims (1)

[Claims] 1. A method of sputtering a carbide resistor material by adding a small amount of impurity gas to a rare gas in a sputtering gas atmosphere, including nitrogen, carbon dioxide, carbon monoxide, or a gas selected from the group thereof. A method for manufacturing a carbide film resistor, characterized in that the resistance value of the film produced is smaller than that in the case where no impurity gas is added, using a mixed gas of at least one kind of the above as an impurity gas. 2. The method for manufacturing a carbide film resistor according to claim 1, wherein the carbide resistor material is silicon carbide or boron carbide. 3. The amount of impurity gas added to the rare gas in the sputtering gas atmosphere is such that the carbide film resistor obtained by sputtering is in a region where it does not undergo any reaction with the impurity gas, and the carbide film resistor has the same composition as the carbide resistor material. 2. The method of manufacturing a carbide film resistor according to claim 1, wherein: 4. The method for manufacturing a carbide film resistor according to claim 1, wherein the amount of nitrogen gas relative to the rare gas is in a range of at least 0 to 2.5 vol.%. 5. The method for manufacturing a carbide film resistor according to claim 1, wherein the amount of carbon dioxide relative to the rare gas is in a range of at least 0 to 1.5 vol.%. 6. The method of manufacturing a carbide film resistor according to claim 1, wherein the amount of carbon monoxide relative to the rare gas is in a range of at least 0 to 1.5 vol.%.