Method for heat-treating ceramic substrate, method for producing ceramic metal bonded body, and method for producing ceramic circuit substrate

Abstract

Provided is a method capable of suppressing variation in insulation properties of a ceramic substrate. This method for heat-treating a ceramic substrate according to an embodiment involves: a step for preparing a ceramic substrate; and a step for heating the ceramic substrate within the range of 1720°C-1900°C while applying pressing force to the ceramic substrate.

Description

Method for heat treatment of ceramic substrate, method for manufacturing ceramic-metal bonded body, and method for manufacturing ceramic circuit board

[0001] The embodiments described below relate to a method for heat treating a ceramic substrate, a method for manufacturing a ceramic-metal bonded body, and a method for manufacturing a ceramic circuit board.

[0002] There are ceramic substrates such as silicon nitride substrates, aluminum nitride substrates, aluminum oxide substrates, and zirconium oxide substrates. Semiconductor elements are mounted on ceramic substrates. In recent years, with the increasing performance of semiconductor elements, ceramic substrates are required to have not only heat dissipation properties but also insulation properties.

[0003] For example, Japanese Patent No. 6293772 (Patent Document 1) discloses a silicon nitride substrate having a thermal conductivity of 50 W / m·K or more. Patent Document 1 shows that pores in the silicon nitride substrate can be eliminated or reduced in size by subjecting the silicon nitride substrate to an additional heat treatment. As a result, Patent Document 1 obtains a silicon nitride substrate having a dielectric strength voltage of 15 kV / mm or more and a variation of 20% or less from the average value of the dielectric strength voltage.

[0004] Patent No. 6293772

[0005] According to the technology described in Patent Document 1, variations in insulation properties within a single substrate can be suppressed. However, variations in insulation properties occur between substrates. In particular, when manufacturing large silicon nitride substrates with a size of 100 mm or more, large variations in insulation properties occur between substrates.

[0006] The embodiments are intended to address such issues and to provide a method capable of suppressing variations in the insulating properties of ceramic substrates.

[0007] The method for heat treating a ceramic substrate according to the embodiment includes the steps of preparing a ceramic substrate and heating the ceramic substrate at a temperature in the range of 1720° C. to 1900° C. while applying a pressure to the ceramic substrate.

[0008] 1 is a side view showing an example of a heat treatment method for a ceramic substrate according to an embodiment; 2 is a side view showing another example of a heat treatment method for a ceramic substrate according to an embodiment; 3 is a side view showing an example of a ceramic-metal bonded body according to an embodiment; 4 is a side view showing an example of a ceramic circuit substrate according to an embodiment; 5 is a flow chart showing an example of a treatment method according to an embodiment;

[0009] The method for heat treating a ceramic substrate according to the embodiment includes the steps of preparing a ceramic substrate and heating the ceramic substrate at a temperature in the range of 1720° C. to 1900° C. while applying a pressure to the ceramic substrate.

[0010] 1 and 2 are side views showing an example of a heat treatment method for a ceramic substrate according to an embodiment. In Fig. 1 and Fig. 2, reference numeral 1 denotes a ceramic substrate, reference numeral 2 denotes a setter plate, reference numeral 3 denotes the distance between the setter plates, and reference numeral 4 denotes a weight.

[0011] First, a step of preparing a ceramic substrate 1 is performed. The ceramic substrate 1 is a ceramic sintered body. That is, the ceramic substrate 1 is produced by subjecting a ceramic molded body to a sintering process. The ceramic substrate 1 may be prepared by subjecting a ceramic molded body to a sintering process, or may be prepared by purchasing a commercially available ceramic substrate.

[0012] Next, a step of heating the ceramic substrate 1 to a temperature in the range of 1720° C. to 1900° C. while applying a pressing force to the ceramic substrate 1 is performed. For example, the pressing force is applied by placing a weight on the ceramic substrate 1 or by pressing the ceramic substrate 1. A hot press may be used for pressing.

[0013] The heating temperature is within the range of 1720°C or higher and 1900°C or lower. The temperature range of 1720 to 1900°C corresponds to the sintering temperature for obtaining a ceramic sintered body. The heat treatment time is preferably within the range of 2 hours or higher and 10 hours or lower. The predetermined temperature within the range of 1720°C or higher and 1900°C or lower is maintained within the range of 2 hours or higher and 10 hours or lower.

[0014] By heat treating the ceramic substrate 1 at a high temperature, it is possible to reduce the size of relatively large pores in the grain boundary phase or to reduce the number of pores. The density of the ceramic substrate 1 after the heat treatment is higher than the density of the ceramic substrate 1 before the heat treatment. The heat treatment is performed until the density of the ceramic substrate 1 reaches 0.05 g / cm. ３ It is preferable that the heat treatment is carried out so that the density is at least as high as possible. The density can be measured by Archimedes' method. By increasing the density of the ceramic substrate 1 by heat treatment, the variation in the insulation properties between the ceramic substrates 1 can be reduced.

[0015] If the heat treatment temperature is less than 1720°C, the pore reduction effect will be insufficient. In particular, when the ceramic substrate 1 is large, 100 mm or larger, the amount of heat may be insufficient. For example, the center of the ceramic substrate 1 may not be heated sufficiently compared to the periphery, and the pores in the center may not be reduced sufficiently. The amount of heat may also be insufficient when multiple ceramic substrates 1 are stacked and heat treated simultaneously. For this reason, the heat treatment temperature is preferably in the range of 1720°C or higher and 1900°C or lower. Furthermore, the heat treatment temperature is preferably within the range of ±50°C of the sintering temperature in the sintering step when producing the ceramic substrate 1.

[0016] When applying a pressure to the ceramic substrate 1, it is preferable to sandwich the ceramic substrate 1 between setter plates 2 as shown in Fig. 1. The setter plates 2 are bottom plates having a high heat resistance. The setter plates 2 are preferably boron nitride (BN) plates.

[0017] The ceramic substrate 1 includes crystal grains and a grain boundary phase located between the crystal grains. During heat treatment, at least a portion of the grain boundary phase melts, forming a liquid phase. The molten grain boundary phase migrates between the crystal grains. In particular, a pressing force is applied to the ceramic substrate 1 during heat treatment, causing the molten grain boundary phase to migrate toward the surface of the ceramic substrate 1. If the surface of the ceramic substrate 1 is exposed, the migrated grain boundary phase seeps out onto the surface and spreads across the surface. As a result, a difference in the density of the grain boundary phase between the interior and surface of the ceramic substrate 1 after heat treatment may occur, potentially increasing pores within the interior. When the setter plate 2 is in contact with the surface of the ceramic substrate 1 during heat treatment, the migration of the molten grain boundary phase to the surface of the ceramic substrate 1 can be suppressed. In other words, the setter plate 2 acts as a lid to prevent the grain boundary phase from leaking out from between the crystal grains. Therefore, by sandwiching the ceramic substrate 1 between the setter plates 2, the grain boundary phase can be suppressed from seeping out onto the surface of the ceramic substrate 1 during heat treatment. That is, the grain boundary phase can be prevented from migrating to the surface of the ceramic substrate 1 and can be kept inside the ceramic substrate 1 .

[0018] In the heating step, it is preferable to stack a plurality of ceramic substrates 1 and heat the stack sandwiched between setter plates 2. By stacking a plurality of ceramic substrates 1, the number of ceramic substrates 1 that can be processed can be increased. As shown in Fig. 2, the setter plates 2 and the stacks may be stacked alternately. The number of stacks and setter plates 2 that can be stacked alternately is arbitrary.

[0019] The upper limit of the number of stacks is not particularly limited, but is preferably 21 or less. Here, each laminate and each setter plate 2 are counted as "1." That is, the total number of stacked laminates and the number of stacked setter plates 2 is preferably 21 or less. Since the setter plate 2 is placed at the bottom and the setter plate 2 on which the weight 4 is placed is placed at the top, when stacking laminates and setter plates 2 alternately, the number of stacks is odd. If the number of stacks exceeds 21, the risk of collapse during transport increases. Furthermore, if the number of stacks is large, the overall warping of the laminated ceramic substrate 1 may be significantly affected. As a result, even when a pressing force is applied by the weight, the warping of the ceramic substrate 1 may not be sufficiently reduced.

[0020] The vertical and horizontal dimensions of the setter plate 2 are preferably larger than the vertical and horizontal dimensions of each ceramic substrate 1. Furthermore, the vertical and horizontal dimensions of the setter plate 2 are preferably larger than the vertical and horizontal dimensions of each ceramic substrate 1 by 5 mm or more.

[0021] When the setter plate 2 is smaller than the ceramic substrate 1, the surface of the ceramic substrate 1 adjacent to the setter plate 2 has an area in contact with the setter plate 2 and an area not in contact with the setter plate 2. In this case, the grain boundary phase does not easily seep into the area in contact with the setter plate 2, but easily seeps into the area not in contact with the setter plate 2. As a result, the distribution of the grain boundary phase becomes uneven on the surface of the ceramic substrate 1. Furthermore, the density of the grain boundary phase inside the ceramic substrate 1 decreases by the amount that the grain boundary phase seeps into the area not in contact with the setter plate 2. When the setter plate 2 is larger than the ceramic substrate 1, the seepage of the grain boundary phase onto the surface of the ceramic substrate 1 can be suppressed.

[0022] The maximum height roughness Rz of the surface of the setter plate 2 is preferably 40 μm or less. While slight irregularities may exist on the surface of the setter plate 2, the maximum depth of the recesses is preferably 40 μm or less. The maximum height roughness Rz is measured in accordance with JIS-B-0601 (2013). The maximum height roughness Rz and maximum depth are measured using a three-dimensional shape measuring instrument. The three-dimensional shape measuring instrument used is a Keyence One-Shot 3D Shape Measuring Instrument VR Series or an instrument with equivalent or better performance. The entire surface of the setter plate 2 is measured, and the maximum height roughness Rz and maximum depth are measured. A smaller value for "maximum depth" indicates a shallower recess. The amount of warping of the setter is preferably within the range of 0 μm to 30 μm per 10 mm.

[0023] A weight 4 is preferably placed on the uppermost setter plate 2. The weight 4 may have various shapes, such as a plate or a disk. The weight 4 preferably has a high density and is heat resistant to temperatures of 1720 to 1900°C. The density of the weight 4 is preferably higher than the density of the setter plate 2. Examples of materials for the weight 4 include tungsten (including tungsten alloys), molybdenum (including molybdenum alloys), rhenium (including rhenium alloys), and niobium (including niobium alloys). The density of boron nitride that can be used for the setter plate 2 is 1.3 to 1.6 g / cm ３ In comparison, the density of tungsten is 19.3 g / cm ３ , the density of molybdenum is 10.3 g / cm ３ , the density of rhenium is 21.0 g / cm ３ , the density of niobium is 8.6 g / cm ３ The density of the weight 4 is 8 g / cm ３ It is preferable that this is equal to or greater than this.

[0024] It is preferable that the weight 4 does not protrude from the setter plate 2. "The weight 4 does not protrude from the setter plate 2" means that the entire weight 4 is located within the outer edge of the setter plate 2 when viewed from above. If the weight 4 protrudes from the setter plate 2, the weight of the weight 4 may not be evenly distributed on the setter plate 2. There is also the risk that the weight 4 may fall, which is dangerous. For this reason, it is preferable that the weight 4 does not protrude from the setter plate 2.

[0025] In the example shown in FIG. 2 , two weights 4 are used. The number of weights 4 is arbitrary. The number of weights 4 may be one or three or more. Multiple weights 4 may be arranged horizontally apart, or multiple large weights 4 may be stacked vertically. The number of weights 4 is preferably two or more. When multiple weights 4 are used, the weight of each weight 4 may be small. By reducing the weight of each weight 4, the burden of transporting the weights 4 onto the setter plate 2 can be reduced. As mentioned above, the weights 4 are made of a material with high density. If the weights 4 are excessively large and heavy, there is a risk of the weights 4 falling during the transport operation. By using multiple weights 4, each weight 4 can be made lighter, making it easier to transport.

[0026] For example, the weight of one weight 4 is preferably in the range of 0.3 kg to 5 kg. When the weight 4 is disk-shaped, it is preferable that the diameter is 20 mm to 140 mm and the thickness is 2 mm to 20 mm. When the weight 4 is square-shaped, it is preferable that the length of one side is 20 mm to 140 mm and the thickness is 2 mm to 20 mm. The "square shape" may be either a rectangle or a square. The total weight of the weights 4 placed on the setter plate 2 is preferably in the range of 0.9 kg to 20 kg. The weight of the setter plate 2 is not limited. For example, when a boron nitride plate is used for the setter plate 2, the weight of the setter plate 2 may be approximately 0.01 to 2 kg.

[0027] For example, it is preferable that a pressing force of 2 N or more is applied to the ceramic substrate 1 by the weight of the setter plate 2 and the weight 4. It is more preferable that the pressing force is 1 kN or more, and even more preferable that the pressing force is 10 kN or more.

[0028] The distance 3 between the setter plates 2 sandwiching the laminate is preferably 20 mm or less. The distance 3 is measured in the lamination direction (thickness direction of the laminate). A distance 3 of 20 mm or less indicates that the thickness of the laminate of the ceramic substrate 1 is 20 mm or less. By setting the distance 3 to 20 mm or less, not only can the insulation properties of the ceramic substrate 1 be improved, but the amount of warping of the ceramic substrate 1 can also be reduced. This is because setting the distance 3 to 20 mm or less allows a uniform pressing force to be applied to each ceramic substrate 1. For this reason, the distance 3 is preferably 20 mm or less, and more preferably 10 mm or less.

[0029] The greater the number of ceramic substrates 1 stacked, the longer the distance 3. When the ceramic substrates 1 are stacked such that the distance 3 exceeds 20 mm, shrinkage variation may occur between the center and periphery of each ceramic substrate 1. The occurrence of shrinkage variation affects the variation in insulation within each ceramic substrate 1 and the warping of the ceramic substrate 1. As mentioned above, the setter plate 2 acts as a lid and suppresses the seepage of the grain boundary phase. When the distance 3 exceeds 20 mm and the warping of the ceramic substrates 1 increases, gaps are likely to occur between the ceramic substrates 1. As a result, the effect of suppressing the seepage of the grain boundary phase at the contact surfaces between the ceramic substrates 1 may be insufficient. By setting the distance 3 to 20 mm or less, the flat setter plate 2 can apply a uniform pressing force to each ceramic substrate 1. As a result, the ceramic substrates 1 can be made as flat as the setter plate 2, suppressing warping of the ceramic substrate 1. For this reason, the distance 3 is preferably 20 mm or less.

[0030] The length of at least one side of the ceramic substrate 1 may be 100 mm or more. The heat treatment method according to the embodiment is also applicable to large ceramic substrates. The shape of the ceramic substrate 1 is, for example, a square or a rectangle. The shape of the ceramic substrate 1 may be a polygon such as a triangle or a pentagon, or a circle (including an ellipse). If the ceramic substrate 1 is circular, the diameter may be 100 mm or more. There is no particular upper limit to the length (or diameter) of one side of the ceramic substrate, but it is preferably 300 mm or less. If the length (or diameter) of one side of the ceramic substrate 1 exceeds 300 mm, a correspondingly larger setter plate 2 is required, which may reduce workability.

[0031] According to the heat treatment method described above, it is possible to suppress variations in insulation between ceramic substrates 1. For example, it is preferable that the variation in withstand voltage relative to the average value among 100 ceramic substrates 1 is 9 kV / cm or less. That is, it is preferable that both the difference obtained by subtracting the minimum withstand voltage from the average withstand voltage and the difference obtained by subtracting the average withstand voltage from the maximum withstand voltage are 9 kV / cm or less. Furthermore, being able to suppress variations in withstand voltage between ceramic substrates means that variations in withstand voltage at each location within each ceramic substrate can also be suppressed. For example, according to the heat treatment method of the embodiment, it is possible to suppress the variation in withstand voltage relative to the average value among 100 ceramic substrates 1 to 9 kV / cm or less.

[0032] The breakdown voltage refers to the dielectric breakdown voltage. The breakdown voltage is measured in accordance with the dielectric breakdown strength test of JIS-C-2141 (1992). JIS-C-2141 corresponds to IEC 672-2. The test is performed using the two-terminal method. Electrodes are placed on the front and back surfaces of the ceramic substrate 1, sandwiching the ceramic substrate 1 between the pair of electrodes. In this state, a 50 Hz or 60 Hz AC voltage is applied between the electrodes, and the voltage is increased from 0 V. The voltage increase rate is set to 200 V / s. The voltage at which the ceramic substrate 1 experiences dielectric breakdown is measured as the breakdown voltage (kV). The breakdown voltage may be measured in kV / cm or kV / mm, which is calculated by dividing the voltage by the thickness of the ceramic substrate. A breakdown voltage variation of ±9 kV / cm is converted to ±0.9 kV / mm. According to the heat treatment method of the embodiment, the average value of the breakdown voltage of 100 ceramic substrates 1 can be made 16 kV / mm or more, and even 17 kV / mm or more.

[0033] The amount of warpage per any 10 mm of the ceramic substrate 1 is preferably in the range of 0 μm or more and 30 μm or less. By using the heat treatment method according to the embodiment, not only can the variation in pressure resistance be suppressed but also the amount of warpage can be reduced. The amount of warpage is measured, for example, by a contact-type three-dimensional shape measurement method, a light-section method, an active stereo method, or white light interferometry. In any of these methods, the warpage in any diagonal direction of the ceramic substrate 1 is measured and converted into the amount of warpage per 10 mm. If the measuring instrument is compatible, area analysis may also be performed. White light interferometry is preferred as a method for measuring the amount of warpage. White light interferometry has a short measurement time and allows area analysis.

[0034] To apply a pressing force to the ceramic substrate 1, it is preferable to use a weight 4 as shown in FIG. 2. After using the weight 4, the ceramic substrate 1 is heated in a heating furnace. A method using the weight 4 can improve mass productivity. For example, methods of heating while applying a pressing force include hot pressing and HIP (hot isostatic pressing). In hot pressing, the laminated ceramic substrates 1 are likely to become misaligned during pressing. In HIP, the ceramic substrates 1 must be placed one by one in a rubber mold, which reduces mass productivity.

[0035] The ceramic substrate 1 is, for example, a silicon nitride substrate, an aluminum nitride substrate, an aluminum oxide substrate, or a zirconium oxide substrate. The thickness of the ceramic substrate 1 is preferably in the range of 0.2 mm to 3 mm. If the thickness of the ceramic substrate 1 is less than 0.2 mm, the insulating properties of the ceramic substrate 1 may be insufficient. If the thickness of the ceramic substrate 1 is greater than 3 mm, the insulating properties are good, but the ceramic substrate 1 may act as a thermal resistor, resulting in reduced heat dissipation. For this reason, the thickness of the ceramic substrate is preferably in the range of 0.2 mm to 3 mm, and more preferably in the range of 0.2 mm to 1 mm.

[0036] The ceramic substrate 1 preferably contains 1% by mass or more and 20% by mass or less of a grain boundary phase. The main component of the grain boundary phase is derived from a sintering aid. The use of a sintering aid improves sinterability and promotes densification of the ceramic substrate 1. As a result, a ceramic substrate 1 containing a grain boundary phase is obtained. Examples of the sintering aid include rare earth compounds, magnesium compounds, titanium compounds, hafnium compounds, zirconium compounds, aluminum compounds, tungsten compounds, and molybdenum compounds.

[0037] The ceramic substrate 1 is preferably a silicon nitride substrate. The thermal conductivity of the silicon nitride substrate is 40 W / m·K or more, and can be 80 W / m·K or more. The three-point bending strength of the silicon nitride substrate is 600 MPa or more, and can be 700 MPa or more.

[0038] The thermal conductivity of aluminum nitride substrates is 160 W / m·K or higher, and can even be 200 W / m·K or higher. The three-point bending strength of aluminum nitride substrates is approximately 300 to 450 MPa. The three-point bending strength of aluminum oxide substrates is approximately 300 to 450 MPa. Aluminum oxide substrates are less expensive than other substrates. The thermal conductivity of aluminum oxide substrates is approximately 20 to 30 W / m·K. The three-point bending strength of zirconium oxide substrates is high, approximately 550 MPa, but their thermal conductivity is approximately 30 to 50 W / m·K.

[0039] As described above, silicon nitride substrates have high strength. Therefore, the thickness of the silicon nitride substrate can be designed to be 0.2 mm or more and 0.5 mm or less. The thinner the substrate, the more likely variations in withstand voltage occur. This is because the thinner the substrate, the greater the impact of pores inside the substrate on the withstand voltage. According to the heat treatment method of the embodiment, variations in withstand voltage can be reduced even for thin substrates. From this perspective, the heat treatment method of the embodiment is particularly suitable for silicon nitride substrates.

[0040] The ceramic substrate 1 obtained by the heat treatment method according to the embodiment can be used for a ceramic-metal bonded body or a ceramic circuit board. A ceramic-metal bonded body is a member in which a metal part is bonded to a ceramic substrate 1. A ceramic circuit board is a member in which a part having a circuit shape is bonded to a ceramic substrate 1. In the ceramic-metal bonded body and the ceramic circuit board, it is preferable that a metal plate is bonded to the ceramic substrate.

[0041] The metal portion is a metal plate, a thin film, a metallized layer, or the like. The metal plate is, for example, a copper plate (including a copper alloy plate) or an aluminum plate (including an aluminum alloy plate). The metal plate is preferably bonded to the ceramic substrate 1 via a bonding layer. An active metal bonding method is preferably used to bond the metal plate. The thin film refers to a conductive film formed by a sputtering method or a plating method. The metallized layer is a conductive layer formed by applying a metal powder paste and firing it.

[0042] Fig. 3 is a side view showing an example of a ceramic-metal bonded body. In Fig. 3, reference numeral 1 denotes a ceramic substrate, reference numeral 10 denotes a ceramic-metal bonded body, reference numeral 11 denotes a metal plate, and reference numeral 12 denotes a bonding layer. The metal plate 11 is bonded to the ceramic substrate 1 via the bonding layer 12. In the example shown in Fig. 3, a metal plate 11 is bonded to each of both surfaces of the ceramic substrate 1. A plurality of metal plates 11 may be bonded to one surface of the ceramic substrate 1.

[0043] Fig. 4 is a side view showing an example of a ceramic circuit substrate. In Fig. 4, reference numeral 1 denotes a ceramic substrate, reference numeral 11 denotes a metal plate, reference numeral 12 denotes a bonding layer, and reference numeral 20 denotes a ceramic circuit substrate. After producing the ceramic-metal bonded body 10 shown in Fig. 3, a circuit shape is imparted to the metal plate 11, thereby obtaining the ceramic circuit substrate 20 shown in Fig. 4. The circuit shape is imparted, for example, by etching a portion of the metal plate. Alternatively, a metal plate having a circuit shape may be bonded to the ceramic substrate 1.

[0044] In the example shown in Fig. 4, a plurality of metal plates 11 in a circuit shape are provided on the surface of the ceramic substrate 1. The plurality of metal plates 11 are formed, for example, by etching a single metal plate. A single metal plate 11 is provided on the back surface of the ceramic substrate 1 as a heat sink. This is not limited to the example shown in Fig. 4, and the number of circuit-shaped metal plates 11 may be one or three or more. In addition to the front surface, a metal plate 11 having a circuit shape may also be provided on the back surface of the ceramic substrate 1.

[0045] The thickness of the metal plate 11 is preferably 0.3 mm or more, and more preferably 0.6 mm or more. The metal plate 11 can be thicker than a thin film and a metallized layer. By applying the metal plate 11 to the ceramic circuit substrate 20, the current-carrying capacity of the ceramic circuit substrate 20 can be improved. According to the embodiment, variations in the withstand voltage of the ceramic substrate 1 are suppressed. By using this ceramic substrate 1 for the ceramic circuit substrate 20, variations in the withstand voltage of the ceramic circuit substrate 20 can also be suppressed.

[0046] 5 is a flow diagram showing an example of a processing method according to the embodiment. The method shown in FIG. 5 includes a heat treatment method (step S10), a method for manufacturing a ceramic-metal bonded body (step S20), and a method for manufacturing a ceramic circuit board (step S30).

[0047] In the heat treatment method, first, the ceramic substrate 1 is prepared. In the example shown in FIG. 5, the ceramic substrate 1 is prepared by fabricating the ceramic substrate 1. Specifically, raw materials for the ceramic substrate 1 are mixed and shaped to produce a ceramic molded body (step S11). The ceramic molded body is sintered by maintaining it at a sintering temperature (step S12). In this way, the ceramic substrate 1 is fabricated. After sintering, the ceramic substrate 1 is removed from the sintering furnace and cooled (step S13). For example, the ceramic substrate 1 is cooled to room temperature.

[0048] A plurality of ceramic substrates 1 are stacked, and the stack is sandwiched between setter plates 2 (step S14). A weight 4 is placed on the setter plates 2 (step S15). Thereafter, the ceramic substrates 1 are heated within a range of 1720°C to 1900°C (step S16). After heating, the ceramic substrates 1 are cooled (step S17). The above steps complete the heat treatment method (step S10).

[0049] When producing the ceramic-metal bonded body 10, after the heat treatment method (step S10), a metal plate is bonded to the ceramic substrate 1 (step S21). The metal plate is preferably bonded by an active metal bonding method. By bonding the metal plate to the ceramic substrate 1, the ceramic-metal bonded body 10 is obtained.

[0050] When producing the ceramic circuit board 20, after the method for producing a ceramic-metal bonded body (step S20), a circuit shape is imparted to the metal plate 11 (step S31). By imparting the circuit shape to the metal plate 11, the ceramic circuit board 20 is obtained.

[0051] (Example) The ceramic substrates shown in Table 1 were prepared. The silicon nitride substrate was Si ３ N ４ In the examples and comparative examples, a ceramic substrate containing a grain boundary phase in the range of 1 mass % to 20 mass % was used.

[0052]

[0053] Next, multiple ceramic substrates were stacked and sandwiched between setter plates. Boron nitride plates were used for the setter plates. In Examples 1 to 7 and Comparative Example 2, the vertical and horizontal dimensions of the setter plates were 5 mm or more larger than the vertical and horizontal dimensions of the ceramic substrates. In Comparative Example 1, a setter plate measuring 90 mm in length and 90 mm in width was used. In Comparative Example 1, a setter plate smaller in vertical and horizontal dimensions than the ceramic substrates was used. In addition, in the Examples and Comparative Examples, setter plates with a maximum height roughness Rz of 40 μm or less and a maximum depth of surface recesses of 40 μm or less were used.

[0054] The ceramic substrate laminate and the setter plates were alternately stacked. When stacking, the setter plates were placed on the top and bottom. A number of weights were placed on the topmost setter plate. The weights were made of a tungsten alloy (density: approximately 17 g / cm ３ ) was used. Weights smaller in size in the vertical and horizontal directions than the setter plate were used. Pressing force was applied to the ceramic substrate by arranging the weights. The total weight of the weights placed on the setter plate was in the range of 0.9 kg to 20 kg.

[0055] In Example 1, 30 ceramic substrates were stacked. In Example 2 and Comparative Examples 1 and 2, 20 ceramic substrates were stacked. In Example 3, 15 ceramic substrates were stacked. In Example 4, 30 ceramic substrates were stacked. In Example 5, 10 ceramic substrates were stacked. In Example 6, 100 ceramic substrates were stacked. In Example 7, 23 ceramic substrates were stacked. Other conditions related to the arrangement, heat treatment conditions, etc. are as shown in Table 2.

[0056]

[0057] In Table 2, "number of alternating arrangements" indicates the number of laminates and setter plates arranged. Setter plates are arranged at the bottom and top, respectively, and laminates and setter plates are arranged alternately between them. For example, in Example 1, Example 5, Comparative Example 1, and Comparative Example 2, two laminates and one setter plate are arranged between the bottom and top setter plates, and the total number of setter plates and the number of laminates is "5." In Examples 2 and 4, three laminates and two setter plates are arranged between the bottom and top setter plates, and the total number of setter plates and the number of laminates is "7."

[0058] In the examples, the heat treatment temperature was set within a range of ±50°C from the sintering temperature for producing the ceramic substrate. The heat treatment time was set within a range of 2 hours to 10 hours. On the other hand, in Comparative Example 1, the heat treatment temperature was set to a temperature approximately 200°C lower than the sintering temperature for producing the ceramic substrate. In Comparative Example 2, the heat treatment temperature was set to a temperature approximately 150°C higher than the sintering temperature for producing the ceramic substrate.

[0059] The dielectric strength voltage of the ceramic substrates according to the examples and comparative examples was measured. One hundred ceramic substrates were extracted from each example and comparative example, and the dielectric strength voltage of each ceramic substrate was measured. The average value of the dielectric strength voltage, the minimum value of the dielectric strength voltage, and the maximum value of the dielectric strength voltage were investigated for each example and comparative example. One lot consisted of 100 ceramic substrates, and the dielectric strength voltage was measured for a total of three lots of ceramic substrates.

[0060] Furthermore, as Reference Example 1, the ceramic substrate according to Example 3 was prepared before being subjected to heat treatment. As Reference Example 2, the ceramic substrate according to Example 7 was prepared before being subjected to heat treatment. The dielectric strength voltage of the ceramic substrate according to the Reference Example was also measured in the same manner as the dielectric strength voltages of the ceramic substrates according to the Examples and Comparative Examples. The measurement results of the dielectric strength voltage are shown in Table 3.

[0061]

[0062] As can be seen from Table 3, in the Examples, the difference between the maximum and minimum values ​​in each lot was smaller than in the Comparative Examples. The variation in breakdown voltage in each lot was within ±0.9 kV / mm. In Comparative Examples 1 and 2, the maximum breakdown voltage in each lot was similar to that in the Examples. However, in each lot, the breakdown voltage of some ceramic substrates was low, and as a result, the difference between the maximum and minimum values ​​was also larger than in the Examples. Furthermore, in Comparative Examples 1 and 2, the minimum breakdown voltage was small, so the average breakdown voltage of the lot was also lower than in the Examples. In particular, some ceramic substrates had a minimum breakdown voltage that was more than 0.9 kV / mm lower than the average. From these results, it is believed that when the heat treatment temperature is low, the pores contained in the ceramic substrate are not sufficiently reduced, and the effect of improving breakdown voltage is small. On the other hand, when the heat treatment temperature is high, the grain boundary phase is excessively melted, increasing fluidity, making it difficult to achieve the effect of reducing pores contained in the ceramic substrate. For example, if the fluidity of the grain boundary phase is high, there is a possibility that the grain boundary phase will wet and spread through the small gap between the ceramic substrate and the setter plate even when the ceramic substrate is in contact with the setter plate. From the above results, it was found that the effect of suppressing variations in withstand voltage can be increased by setting the heat treatment temperature to 1720 to 1900°C.

[0063] Furthermore, the minimum value of withstand pressure decreased slightly when the distance between the setter plates exceeded 20 mm, as in Example 6. The minimum value of withstand pressure also decreased slightly when the number of stacked laminates and setter plates exceeded 20, as in Example 7. It can be seen that the effect of suppressing variations in withstand pressure was further improved by setting the distance between the setter plates to 20 mm or less and setting the number of laminates and setter plates to 20 or less, as in Examples 1 to 5.

[0064] Furthermore, the results of Reference Examples 1 and 2 show that when no heat treatment was performed, the withstand voltage of the ceramic substrate varied greatly.

[0065] Next, in each example and each comparative example, the presence or absence of an increase in density of the ceramic substrate due to heat treatment and the flatness of the ceramic substrate after heat treatment were examined. The increase in density due to heat treatment was confirmed by measuring and comparing the density of the ceramic substrate before and after heat treatment. ３ In cases where the density increased by more than 100 μm, the density was judged to be "present." For flatness, the amount of warpage per 10 mm was measured using white light interferometry. Ceramic substrates with an amount of warpage of 0 μm or more and 30 μm or less were judged to be "present" in flatness. The methods for measuring density and warpage were as described above. The measurement results are shown in Table 4.

[0066]

[0067] As can be seen from Table 4, in all examples, 0.05 g / cm ３ Density increases of 0.1 g / cm or more were observed in some ceramic substrates. ３ The above density increase was confirmed. In addition, in all Examples, flatness was also improved. In Examples 6 and 7, some of the preferable conditions were not satisfied, and therefore, about 2 to 5 ceramic substrates were confirmed to have a warpage of more than 30 μm per 10 mm.

[0068] On the other hand, no increase in density was confirmed in Comparative Example 1. From these results, it can be seen that the effect of improving density is small at heat treatment temperatures below 1720°C. It was also found that the effect of improving flatness is small. When the heat treatment temperature is high as in Comparative Example 2, grain growth of silicon nitride crystal particles occurs, and the effects of increasing density and improving flatness are not sufficiently obtained.

[0069] From the above, it has been found that the heat treatment method according to the embodiment can provide effects such as suppression of variations in breakdown voltage, improvement of density, and improvement of flatness.

[0070] Although several embodiments of the present invention have been described above, these embodiments are presented by way of example only and are not intended to limit the scope of the invention. These novel embodiments can be embodied in various other forms, and various omissions, substitutions, modifications, etc. can be made without departing from the spirit of the invention. Modifications of these embodiments are included within the scope and spirit of the invention, as well as within the scope of the invention and its equivalents as set forth in the claims. Furthermore, the above-described embodiments can be implemented in combination with each other.

[0071] REFERENCE SIGNS LIST 1 ceramic substrate 2 setter plate 3 distance between setter plates 4 weight 10 ceramic metal bonded body 11 metal plate 12 bonding layer 20 ceramic circuit board

Claims

1. A method for heat treating a ceramic substrate, comprising the steps of: preparing a ceramic substrate; and heating the ceramic substrate at a temperature in the range of 1720°C to 1900°C while applying a pressure to the ceramic substrate.

2. The method for heat treating a ceramic substrate according to claim 1, wherein in the heating step, the laminate of a plurality of the ceramic substrates is heated while sandwiched between a pair of setter plates.

3. The method for heat treating a ceramic substrate according to claim 2, wherein the vertical and horizontal dimensions of the setter plate are larger than the vertical and horizontal dimensions of each of the ceramic substrates.

4. A method for heat treating a ceramic substrate according to claim 2 or 3, wherein in the heating step, the pressing force is applied to the plurality of ceramic substrates by placing a weight on the setter plate located on the laminate, and the density of the weight is greater than the density of each of the setter plates.

5. The method for heat treating a ceramic substrate according to claim 4, wherein the weight is positioned so as not to protrude from the setter plate on which the weight is placed.

6. A method for heat treating a ceramic substrate according to any one of claims 2 to 5, wherein the distance between the pair of setter plates is 20 mm or less.

7. A heat treatment method for a ceramic substrate according to any one of claims 1 to 6, wherein the preparing step includes the steps of: sintering a compact to produce the ceramic substrate; and cooling the sintered ceramic substrate.

8. A method for heat treating a ceramic substrate according to any one of claims 1 to 7, wherein the ceramic substrate is a silicon nitride substrate.

9. A method for heat treating a ceramic substrate according to any one of claims 1 to 8, wherein the length of at least one side of the ceramic substrate is 100 mm or more.

10. A method for heat treating a ceramic substrate according to any one of claims 1 to 9, wherein the heating step is carried out on at least 100 of the ceramic substrates, and after the heating step, the variation in breakdown voltage of the 100 ceramic substrates is an average of ±9 (kV / cm).

11. A method for heat treating a ceramic substrate according to any one of claims 1 to 10, wherein after the heating step, the amount of warpage per any 10 mm of the ceramic substrate is within the range of 0 μm to 30 μm.

12. A method for producing a ceramic-metal bonded body, comprising carrying out the method for heat treating a ceramic substrate according to any one of claims 1 to 11, and bonding a metal plate to the heat-treated ceramic substrate.

13. A method for producing a ceramic circuit board, comprising carrying out the method for producing a ceramic-metal bonded body according to claim 12, and etching the metal plate of the produced ceramic-metal bonded body to provide a circuit shape.

Images