JP5183717B2 - Electronic components - Google Patents

Description

  The present invention relates to an electronic component.

  The concave portion formed in the ceramic member is formed for use in dividing, and for mounting a circuit such as a semiconductor chip, particularly in a ceramic substrate for electronic components.

  A general method for forming these recesses is to form a recess by pressing a blade provided in a mold on a sheet-like ceramic green sheet. Recently, there is a problem that the shrinkage of the product is large and the dimensional accuracy is inferior, and in recent years when high dimensional accuracy is required, there are an increasing number of methods for thermally processing a sintered ceramic member such as a laser.

  The method of forming a recess by irradiating a sintered ceramic member with laser light is to melt the workpiece with the thermal energy of the laser light and process it into a desired shape. There has been a problem that the melted ceramic remains on the surface layer portion of the recess, and that the molten material lands on the periphery of the recess and the optical system of the laser device.

  Therefore, as a method of preventing the molten layer from remaining in the recesses formed by the laser processing, the assist gas or a gas different from the assist gas is sprayed on the surface to be processed at the same time as the laser beam irradiation (hereinafter not shown), A method is disclosed in which a molten layer does not remain in the surface portion of the recess by sucking ceramic vapor sublimated by laser light (Patent Document 1).

  Although there are many prior arts by the above method, none of the methods can satisfactorily remove the molten layer. Therefore, as described below, a number of prior arts have been disclosed regarding a method of removing a molten layer remaining in a recess processed with a laser beam.

  As one of the methods, there is a chemical etching method (not shown below). In this example, a through-hole is formed. However, the through-hole is formed in an alumina substrate with a laser beam and then immersed in a phosphoric acid solution to melt the through-hole. It is disclosed that the inner wall surface is roughened by etching the layer (Patent Document 2).

  As another method, there is a method of removing the molten layer by sandblasting (hereinafter not shown). In this case as well, a through hole is formed, but spherical glass beads are blown into the through hole processed with laser light. It is disclosed that the melted layer of the through hole can be removed without roughening the surface of the ceramic substrate (Patent Document 3).

Further, as a method having a different processing mechanism, as shown in FIG. 16, a ceramic member 101 is irradiated with a laser beam 121 oscillated by a CO ₂ laser through a condenser lens 122, and a temperature of 600 ° C. or less is applied to the irradiated surface. A method of cutting the ceramic member 101 by generating a crack 119 due to thermal stress is disclosed (Patent Document 4). Unlike the method described above, this method causes little thermal damage to the workpiece and does not generate a molten layer.

  As with the remaining of the molten layer on the concave surface, a major problem in laser processing is the landing of the melt around the processing site. Melting of the melt is a major problem regardless of whether the melt occurs on the workpiece or the optical system of the laser apparatus.

  As a method for solving the above-described problem, in Patent Document 1 described above (not shown), while irradiating a workpiece with laser light, a gas different from the assist gas is used as a waste material from the side that becomes the product of the workpiece. The welded material is prevented from adhering to the product side of the workpiece by spraying the sublimated vapor of the ceramic as the workpiece with the suction nozzle.

  However, not only can the welded material not be completely prevented by the above-described method, but the processing with laser light is not limited to the case where there is always a waste material on one side. For this reason, it is also disclosed that a part other than the laser light irradiation part (hereinafter not shown) is patterned with a protective layer such as a resin film to prevent the welded material from adhering to the workpiece (Patent Document 5).

Japanese Patent Application Laid-Open No. 8-141764 JP-A-1-112794 Japanese Patent Laid-Open No. 2004-22643 JP 2000-323441 A JP-A-3-252384

  However, the method in which gas is blown to the laser processing site described in Patent Document 1 to suck the vapor sublimated from the workpiece with a nozzle is difficult to completely suck even in cutting or hole processing. Furthermore, if the shape of the recess is a recess, the molten layer cannot remain on the surface of the recess. As shown in FIG. 17, when the recess 102 is formed as the dividing groove 103 in the ceramic member 101, the molten layer 116 remains with respect to the processing depth d <b> 2 of the recess 102 as shown in FIG. 17. The substantially concave portion 102 has a depth d1. Therefore, if the depth of the concave portion 102 is shallow, dividing the ceramic member 101 along the concave portion 102 may result in poor division of the ceramic member 101 or in order to avoid poor division. When 102 is formed deeply, problems such as cracking or cracking before the dividing step occur frequently.

  Furthermore, since the melted layer 116 of the recess 102 is formed by heating the ceramic with laser light and then cooling and solidifying it, innumerable microcracks 117 are generated due to shrinkage during cooling. As shown in FIG. 18, there is an electronic component 105 that forms an end face electrode 107 so as to be connected to the circuit pattern 106 after forming the circuit pattern 106 on the main surface of the ceramic member 101 and dividing it along the recess 102. If there is a microcrack 117 in the melted layer 116 on the surface of the divided recess 102, the melted layer 116 and the ceramic member 101 are likely to be peeled off by the microcrack 117, thereby reducing the adhesion strength of the end face electrode 107, and melting. Since the surface layer portion of the layer 116 is a smooth surface, the anchor effect is low, and there is a problem that the adhesion strength between the end face electrode 107 and the ceramic member 101 including the molten layer 116 is lowered.

  In order to solve these problems, Patent Document 2 discloses a method of removing the molten layer 116 by chemical etching. However, since the ceramic member 101 itself is immersed in an etching solution, it is necessary to protect a portion that is not to be roughened with a resist film. There is a problem that even if patterning is performed with a resist film, the etching solution does not roughen in the vicinity of the boundary of the resist film and the surface becomes rough, and the ceramic member 101 has, for example, a lattice-like delicate and numerous recesses 102. There is a problem that this method cannot be adopted in the formed one.

Further, in the method of forcibly removing the molten layer 116 by sandblasting in Patent Document 3, not only the molten layer 116 is removed but also the surface of the alumina substrate is ground, so that the substrate surface must be smooth. The ceramic member 101 was not a suitable method.

  Moreover, since the method of patent documents 1-3 mentioned above melts and sublimates the ceramic member 101 which is a to-be-processed object with a laser beam, as shown in FIG. 19, the ceramic of the periphery of the processed recessed part 102 is shown. It has been unavoidable that a ceramic weld 118 is attached to the member 101 and the condenser lens 122. Such contamination by the welded material 118 of the optical system reduces the laser processing capability, and the adhesion of the welded material 118 to the surface of the ceramic member 101 is a cause of poor formation of the circuit pattern 106 or damage to the mask for forming the circuit pattern. There was a problem of becoming.

  Here, the welded material 118 attached to the surface of the workpiece is generally removed by buffing or the like on the surface of the ceramic member 101. The process of removing the welded material 118 is a basic process after laser processing. The current situation is that it is a process. However, when the welded material 118 is removed by a physical process such as buffing, particle destruction occurs on the ceramic surface of the workpiece, and when this is used for the electronic component 105 to form the circuit pattern 106, the adhesion of the pattern is avoided. There was a problem that the strength decreased.

  On the other hand, Patent Document 4 is not based on the high-energy heat irradiation processing inherent to the laser, but is performed by running the crack 119 by a thermal stress of low-temperature heat irradiation at which the ceramic does not melt. However, in this method, as shown in the cross-sectional view of the recess 102 formed in the ceramic member 101 of FIGS. 20A, 20B, and 20C, a substantially V-shaped or U-shaped dividing groove of the recess 102 is provided. Processing of a desired shape and depth such as 103 and the semiconductor chip mounting recess 104 was impossible.

  Further, in the laser processing including the above-described conventional process for removing the molten layer 116, the main component ceramic and a small amount of various additives existing in the non-processed part are melted, and the generated molten layer 116 is removed. Therefore, the composition of the surface layer of the recess after laser processing is almost the same as the composition of the ceramic member other than the recess, and various additives are dispersed almost uniformly in the main component ceramic. there were.

  However, in the electronic component 105 in which the end face electrode 107 is formed in the ceramic member dividing recess 102, the adhesion between the conductive material constituting the end face electrode 107 and the main component ceramic constituting the ceramic member 101 is inferior in recent years. With the miniaturization of the electronic component 105 to be obtained, the contact area between the ceramic member 101 and the end face electrode 107 has been reduced, and the problem of adhesion has been exposed.

  Furthermore, in the conventional laser processing, as shown in the plan view of FIG. 21A, even if the dividing groove 103 closer to the continuous groove is formed, the wide portion D1 of the dividing groove 103 and the narrow width are narrowed. There is a portion D2, and the difference between the widths D1 and D2 is inevitable at least 10 μm or more. This is a reduction in the dimensional accuracy of the shape processing as it is, and after dividing along the dividing groove 103, the side surfaces are uneven, so that the ceramic members or jigs and the like are connected to the side surfaces. There is a problem that chipping of the ceramic member 101 is liable to occur when contact is made. Then, as shown in the cross-sectional view of FIG. 21 (b), the dividing groove 103 also has an up and down wave shape in the depth direction. For example, when the maximum depth d3 is processed to 100 μm or more, a shallow portion d4 and The difference d5 is inevitable at least 20 μm or more, and the problem of occurrence of poor division or cracks or cracks could not be solved.

Furthermore, although the ceramic member 101 including the semiconductor chip mounting recess 104 is illustrated in FIG. 22, it is possible to form the recess 102 in which the bottom surface 104 a of the recesses 102 and 104 is flat and the ceramic melt layer does not substantially exist. Such a method of manufacturing the recess 102 is impossible by laser processing. The method of manufacturing such a recess 102 is a method in which a ceramic green sheet is formed with a through-hole and a non-through-hole are bonded and fired. The method for forming the film is employed, and all of these methods have a problem of high cost.

Electronic component of the present invention, _Al 2 _O 3 from 91 to 99.6 wt%, _SiO 2, MgO, forming a conductor on at least one major surface of the ceramic substrate comprising a CaO as the additive component, before Si is present in the side surface layer portion of the ceramic substrate at 50 to 90% of the surface area of the side surface layer portion, an electrode containing Si is formed on the side surface layer portion, and the electrode and the conductor are formed. It is characterized by conduction.

Electronic component of the present invention, _Al 2 _O 3 from 91 to 99.6 wt%, _SiO 2, MgO, forming a conductor on at least one major surface of the ceramic substrate comprising a CaO as the additive component, before Si is present in the side surface layer portion of the ceramic substrate at 50 to 90% of the surface area of the side surface layer portion, an electrode containing Si is formed on the side surface layer portion, and the electrode and the conductor are formed. When the electrode is formed with the thick film paste on the side surface layer portion because of conduction, a part of the component constituting the vitreous in the thick film paste is a part of the additive component of the side surface layer portion of the ceramic substrate. Move to a portion where the additive does not substantially exist, and on the other hand, a part of the additive component constituting the ceramic substrate moves to the paste, whereby the additives mutually diffuse and the adhesion strength with the electrode is improved. Can A highly reliable electronic component can be obtained.

(A), (b) is a perspective view which shows an example of the ceramic member which is a part of structure of the electronic component of this embodiment. It is sectional drawing which shows an example of the electronic component of this embodiment. (A) is a concave surface layer part of a ceramic member which is a part of the configuration of the electronic component of the present embodiment, and (b) is a drawing-substituting photograph of the EPMA analysis result of each non-thermal irradiation part. It is sectional drawing at the time of using the ceramic member which is a part of structure of the electronic component of this embodiment as an electronic component of this embodiment. (A), (b), (c) is the schematic diagram of the crystal grain of a recessed part surface layer part. It is a top view of the ceramic member which is a part of structure of the electronic component of this embodiment. It is a top view which shows an example of the ceramic member which is a part of structure of the electronic component of this embodiment. It is a recessed part division | segmentation fracture surface of the ceramic member which is a part of structure of the electronic component of this embodiment. It is a fragmentary sectional view of the electronic component of this embodiment. It is a perspective view which shows an example of the manufacturing method of the ceramic member which is a part of structure of the electronic component of this embodiment. (A), (b) is recessed part sectional drawing of the ceramic member which is a part of structure of the electronic component of this embodiment. (A), (b) is a schematic diagram of a pulse waveform. (A) is a schematic diagram of ON-OFF of a pulse signal, (b) is a schematic diagram which shows the relationship between the rise of a pulse waveform, and temperature. (A) is a perspective view of a ceramic member, (b) is sectional drawing of a bending load measurement. It is the perspective view which formed the electrode in the recessed part surface layer part. It is a perspective view which shows an example of the conventional laser processing method. It is a perspective view of the conventional ceramic member. It is a fragmentary sectional view of the electronic component which used the conventional ceramic member. It is sectional drawing of the conventional laser processing. (A), (b), (c) is sectional drawing which shows an example of the ceramic member which has a recessed part. The conventional ceramic member is shown, (a) is a top view, (b) is sectional drawing. The perspective view of the conventional ceramic member is shown.

  Hereinafter, an embodiment of a ceramic member which is a part of the configuration of the electronic component of the present embodiment will be described.

  FIG. 1 is a perspective view showing an example of a ceramic member 1 which is a part of the configuration of the electronic component of the present embodiment. FIG. 1A is a view in which a recess 2 is formed as a dividing groove 3, and FIG. ) Shows the concave portion 2 formed as the semiconductor chip mounting concave portion 4. The recess 2 may be provided on both main surfaces 1a and 1a 'of the ceramic member 1.

  The ceramic member 1 which is a part of the configuration of the electronic component of the present embodiment includes a recess 2 on at least one main surface, and the ceramic member 1 includes a main component and an additive serving as a plurality of sintering aids. The concave portion 2 is formed by heat irradiation, and the molten layer 116 as shown in FIG. 17 does not substantially exist in the surface layer portion 2 a of the concave portion 2.

  In the conventional laser-processed heat-irradiated part, the main component ceramic and additive are melted by the high heat of the laser beam to cover the concave surface layer part, and this is used as the molten layer 116. Since the formed molten layer 116 of the ceramic does not exist in the surface layer portion 2a of the concave portion 2, when the concave portion 2 is used as the dividing groove 3, since the molten layer 116 does not exist, the splitting property is good, and the concave portion 2 Since the depth d can be reduced, it is possible to prevent the problem of cracking or cracking before the dividing step. This is because if the melted layer 116 is processed, the concave portion 2 had to be deeply formed in consideration of the thickness of the melted layer 116 in advance, and as described above, cracks occurred before the dividing step. Was.

  Further, when the recess 2 is used as the recess 4 for mounting a semiconductor chip, the molten layer 116 does not exist on the bottom surface 4a of the recess 2, so that a thin film metal layer 8 such as Ni is formed on the recesses 2 and 4 as shown in FIG. The metal layer 9 such as Au is deposited on the surface layer portion 4a, and the semiconductor chip 10 is mounted thereon. However, since the molten layer 116 does not exist on the bottom surface 4a of the recesses 2 and 4, the anchor effect And the adhesion strength of the thin film metal layer 8 to the ceramic member 1 can be improved.

  On the other hand, the ceramic member 1 that is a part of the configuration of the electronic component of the present embodiment includes a concave portion 2 that is composed of a main component and a plurality of additives and is formed on at least one main surface by heat irradiation. In the surface layer portion 2a of 2, the component (silica) having a melting point lower than that of the ceramic (alumina) as the main component is unevenly distributed in the surface layer portion 2a of the recess 2. FIG. 3A shows a drawing-substituting photograph of EPMA analysis (Electron Probe Micro Analysis) of the surface layer portion 2a of the concave portion 2 of the ceramic member 1 processed by the heat irradiation, and FIG. The drawing substitute photograph of the EPMA analysis of the irradiation part is shown. Here, as an example, the ceramic that is the main component of the ceramic member 1 that is a part of the configuration of the electronic component of the present embodiment is alumina, and calcia, magnesia, and silica as an additive having a lower melting point than ceramic. Used. In each of the EPMA analysis methods, the analysis site was vapor-deposited with gold, the analysis area was set to 67 μm in both length and width, the mapping conditions were K spectral line, and the distribution state of Si element was analyzed with spectral crystal TAP. Here, the black portions in the photographs of FIGS. 3A and 3B represent portions where Si is not substantially present in the above method.

That is, as shown in FIG. 3B, silica, which is an additive having a melting point lower than that of alumina, is dispersed throughout the non-heat-irradiated portion, but as shown in FIG. It can be seen that the silica is unevenly distributed in the portion 2a.

  As described above, a part of the component of the additive forming the glassy material is unevenly distributed in the surface layer portion 2a of the concave portion 2 without being uniformly distributed. Is formed, a part of the additive contained in the paste moves to a portion where the additive forming the vitreous of the ceramic member 1 is not substantially present, while the ceramic as a main component constituting the ceramic member 1 Since the additive (silica) having a lower melting point moves to the paste and diffuses, there is an effect that the adhesion strength of the conductor to the ceramic member 1 is improved.

  Further, in the ceramic member 1 which is a part of the configuration of the electronic component of the present embodiment, it is preferable that the welded material 118 as shown in FIG. Here, the welded material 118 is a material in which the ceramics and additives as the main components described above are melted by high heat from a laser beam or the like and are adhered to the periphery of the processed part. This eliminates the need to perform the step of removing the welded material 118 by buffing or the like, and causes poor adhesion strength when a circuit pattern is formed on the particle fracture surface due to crystal particle destruction of the surface of the ceramic member 1 by the removal step. In addition, it is possible to reduce the occurrence of problems such as defective circuit patterns and breakage of the printing mask due to the deposit 118 remaining.

Further, in the ceramic member 1 which is a part of the configuration of the electronic component of the present embodiment, an additive having a melting point lower than that of the ceramic as the main component is present in an area of the surface layer portion 2a of 80% or less of the concave portion 2. Is preferred. This is because, as shown in FIG. 4, in the small electronic component in which the concave portion 2 formed in the ceramic member 1 is divided and the end face electrode 7 is formed on the side surface 1 b including the concave portion 2 and connected to the circuit pattern 6. The adhesion strength between the side surface 1b and the end surface electrode 7 can be increased. For example, an electronic component of the so-called 0402 mm type has a length of 0.4 mm, a width of 0.2 mm, a thickness of 0.1 mm, and an end face electrode 7 formed on a side surface having a width of 0.2 mm. Part 2a
If the molten layer 116 and the microcracks 117 are present, the adhesion strength of the end face electrode 7 is greatly reduced. Further, even if the molten layer 116 is removed by the above-described method, the adhesion strength cannot be satisfied because the electrode formation area is small, but as described above, it is included in the paste for forming the end face electrode 7. A part of the additive and the additive in the ceramic member 1 move to each other and mutually diffuse, so that the additive of the component lower than the melting point of the ceramic is uniformly dispersed, and the end face electrode 7. Therefore, the adhesion strength between the ceramic member 1 and the terminal electrode 7 is easily improved. And, since the component having a melting point lower than that of the ceramic is unevenly distributed in the surface layer portion 2a of the concave portion 2, the adhesion strength is improved by 80% or less in the area of the surface layer portion 2a of the concave portion 2 as compared with the case of not unevenly distributing. This is a condition in which a component having a melting point lower than that of the main component ceramic is present.

  Here, when an additive having a melting point lower than that of the ceramic as the main component occupies an area exceeding 80% of the surface layer portion 2a of the concave portion 2, it is difficult to improve the adhesion strength of the conductor, When the above value is less than 50%, the ceramic as the main component is partially melted, and the features of the present embodiment are impaired. A more preferable range of the ratio in which the additive having a lower melting point than the ceramic as the main component exists in the area of the surface layer portion 2a of the recess 2 is 50% or more and 70% or less.

Further, as shown in the schematic diagram of FIG. 5A, the surface layer portion 2 a of the recess 2 is preferably made of a grain interface of ceramic crystal particles 13 that substantially form the main component. FIG. 5 (b) shows a schematic view of the surface layer portion 102a of the conventional laser processed recess 102. The surface layer portion 102a is a molten glassy material 14 and the molten glassy material 14 is solidified immediately after laser processing. It is covered with microcracks 117 due to shrinkage. When a conductor is formed on the surface layer portion 102a, the adhesion strength is reduced in both thick and thin conductor formation. When the molten layer 116 is removed by a physical method to solve this problem, as shown in FIG.
Although the molten layer 116 does not exist, even the ceramic crystal grains 13 under the molten layer 116 are broken, so that the surface layer portion 102a of the recess 102 becomes an intragranular fracture surface.

  However, when a conductor is formed on the intragranular fracture surface with a thick film or a thin film, the surface of the particle is a smooth surface, so the anchor effect is low and the adhesion strength of the conductor may be reduced. On the other hand, in the surface layer portion 2a of the concave portion 2 of the ceramic member 1 which is a part of the configuration of the electronic component of the present embodiment, the ceramic crystal particles 13 are the grain interfaces, and the molten layer 116 covers the surface. Therefore, the anchor effect is large even when the conductor is formed by either a thick film or a thin film, and therefore, a high adhesion strength of the conductor can be obtained.

Moreover, as shown in FIG. 6, it is preferable that the dispersion | variation in the distance D between the recessed lines 2 and the main surface 1a of the ceramic member 1 is 4 micrometers or less. Conventionally, when a concave portion having a depth of 100 μm or more is formed in a ceramic member, it must be processed by a powerful CO ₂ laser. In this case, at least the variation in the distance D between the boundary lines is 50 μm or more. There was a problem of becoming. And when this recessed part is used as a dividing groove, there are at least 25 μm or more unevenness on the side surface after the division, and when the side surface of the ceramic member 1 and the jig or the like come into contact, the ceramic member 1 is likely to be chipped, It may be difficult to form a conductor on the side surface, and sufficient adhesion strength may not be obtained.

  On the other hand, when the variation in the distance D between the boundary lines is 4 μm or less, the unevenness on the side surface after the division can be reduced to 2 μm or less even when the concave portion 2 is used as the dividing groove 3. Even if it comes in contact with a jig or the like, it is possible to prevent the ceramic member 1 from being chipped, and to suppress poor printing of conductor formation. Furthermore, high dimensional accuracy can be ensured even if the fine lattice-shaped dividing grooves 3 as shown in FIG. 7 are formed.

  Here, the variation in the distance D between the boundary lines is that if the length of the concave portion 2 is 1 mm or more in the laser processing distance, the laser light dots are included in about 10 dots in the conventional laser. Since the variation is not greatly different from the case of the investigation, the maximum value D1 and the minimum value D2 of the distance D between the boundary lines of the opposing main surface of the recess 2 at a predetermined length of 1 mm or more are measured, and the difference is measured between the boundary lines. It is assumed that the distance D varies. The measurement method may be a factory microscope for conventional laser processed products, but the measurement of the product of this embodiment may be within the measurement error range of the factory microscope due to small variations, so as to increase the measurement accuracy. Gold for surface analysis was deposited on the main surface side including the concave portion 2 of the ceramic member 1, and an SEM analysis photograph was taken to measure the distance on the photograph. In addition, although the substantially linear recessed part 2 is shown in Fig.6 (a), you may form the meandering recessed part 2 like FIG.6 (b).

  Further, as shown in FIG. 8, the variation d5 in the depth d of the recess 2 is preferably 8 μm or less. Conventionally, even when the concave portion 2 having a depth of 100 μm or more is formed in the ceramic member so as to be a continuous groove for division, the concave portion 2 becomes a triangular wave groove and the variation d5 is at least 30 μm or more. Is difficult to form, and if the depth d of 20% or more of the thickness t of the ceramic member 1 is not processed, defective division is likely to occur.

  On the other hand, since the variation d5 of the depth d is as small as 8 μm or less, when the recess 2 is used as a dividing groove, if the thickness t of the ceramic member 1 is 15% or more and 20% or less, there is a problem in the dividing property. In addition, since it is not necessary to increase the depth d more than necessary, it is possible to prevent the occurrence of cracks and cracks in the handling and heat treatment steps before the dividing step.

Here, the variation d5 of the depth d of the recess 2 is a laser processed product, as in the variation of the distance D between the boundary lines, and a highly accurate value can be obtained by confirming a length of 1 mm or more. . The difference between the maximum depth d3 and the minimum depth d4 of the concave portion 2 at the predetermined distance is the variation d5 of the depth d. Generally, after applying the red flaw detection liquid to the concave portion 2, it is divided along the concave portion 2 and is divided by a factory microscope. The depth of the colored portion is measured. In the product of the present embodiment, the variation d5 of the depth d is small, so measurement by a factory microscope may fall within the measurement error range. Therefore, after dividing along the recess 2, gold for surface analysis was deposited on the divided side surface to take a SEM analysis photograph, and the variation d5 of the depth d was obtained from the photograph.

In addition, the ceramic member 1 which is a part of the configuration of the electronic component of the present embodiment contains Al ₂ O ₃ 91 to 99.6% by mass of alumina as a main component and SiO ₂ , MgO, and CaO as additives. It is preferable. Here, the range of Al ₂ O ₃ can secure the necessary characteristics as a ceramic member for electronic parts, and the above additives are desirable as a binder. And, in the surface layer portion 2a of the concave portion 2 of the ceramic member 1 which is a part of the configuration of the electronic component of the present embodiment, the alumina constituting the main component is not substantially melted, and thus has a grain boundary crystal plane, Silica having a melting point lower than that of alumina is not uniformly dispersed in the surface layer portion 2a, but there is a region where the silica is present and a region where the silica is not substantially present, and the silica is unevenly distributed in the surface layer portion 2a ( Is unevenly distributed). Therefore, when a conductor made of a thick film, a thin film, or the like is formed on the surface layer portion 2a of the recess 2, the grain boundary crystal plane is combined with the mutual diffusion action of the additive in the ceramic member 1 and the additive in the conductor. It is possible to improve the adhesion strength of the conductor due to the high anchor effect due to.

  Here, in addition to the above alumina, aluminum nitride, silicon nitride, ferrite, or alumina zirconia-based ceramics may be used as the ceramic constituting the main component of the ceramic member 1 that is a part of the configuration of the electronic component of the present embodiment.

  Further, the ceramic member 1 which is a part of the configuration of the electronic component of the present embodiment has the ceramic member 1 as a plate-like body, and the concave portion 2 serves as a groove for dividing the ceramic member 1. Even if a large number of fine dividing grooves 3 are formed, the dimensional accuracy is high. Furthermore, since the variation in the depth d of the recess 2 is small, even if the depth d is shallow, the splitting property is good and the generation of cracks and cracks can be prevented.

  As shown in FIG. 4, the electronic component 5 of the present embodiment is obtained by dividing the concave portion 2 of the ceramic member 1 formed as a groove to be divided along the concave portion 2. A circuit pattern 6 is formed on the main surface a, and a conductor is formed so as to connect the circuit pattern 6 and the recess 2, and the electrode 7 is provided on the end surface 1 b obtained by dividing the ceramic member 1 along the recess 2. Forming. The electrode 7 is usually formed by dipping a thick film paste on a ceramic member and firing it, but there is no molten layer in the surface layer portion 2a of the recess 2, and the crystal grains of the main component ceramic are grain boundary crystal planes. In addition, since the additive of the component having a melting point lower than that of the ceramic is unevenly distributed, the thick film paste can be easily penetrated into the ceramic member 1 and constitutes the vitreous contained in the thick film paste. When a part of the agent and an additive having a melting point lower than that of the ceramic as the main component in the ceramic member 1 move to each other, the adhesion strength of the conductor can be improved by the high anchor effect and the mutual diffusion action of the additive.

  Moreover, the electronic component 5 of this embodiment uses the ceramic member 1 which is a part of the structure of the electronic component of this embodiment described above, and as shown in FIG. The thin film metal layer 8 described above is deposited, at least one metal plating layer 9 is formed on the thin film metal layer 8, and the semiconductor chip 10 is mounted on the metal plating layer 9. Since there is no molten layer 116 on the bottom surface 4a of the recess 4 for mounting a semiconductor chip, the ceramic crystal particles constituting the main component are grain interfaces, and the additive having a component lower than the melting point of the ceramic is unevenly distributed. When the thin film metal layer 8 is formed on the bottom surface 4a of the semiconductor chip mounting recess 4 by vapor deposition, the thin film metal layer 8 having high adhesion strength can be formed by the high anchor effect and the mutual diffusion action of the additive. Accordingly, when the metal layer 9 is formed thereon by plating or the like and the semiconductor chip 10 is mounted, the electronic component 5 having high adhesion strength and high reliability can be obtained.

  Next, as shown in FIG. 9, in the electronic component 5 of the present embodiment, the conductor 11 is formed on at least one main surface 16a in the ceramic substrate 16 composed of the main component and a plurality of additive components, and the addition A component having a melting point lower than that of the main component of the agent is unevenly distributed in the side surface layer portion 16b of the ceramic substrate 16, and the electrode 17 is formed on the side surface layer portion 16c including the side surface layer portion 16b. The body 11 is electrically connected. As described above, since a part of the additive component is unevenly distributed in the side surface layer portion 16b, when the electrode 17 is formed with the thick film paste in that portion, one of the components constituting the glassy material in the thick film paste. Part of the side surface layer portion 16b of the ceramic substrate 16 moves to a portion where a part of the additive component does not substantially exist, while a part of the additive component constituting the ceramic substrate 16 is transferred to the paste. By moving, the additives diffuse to each other, the adhesion strength with the electrode 17 can be improved, and the highly reliable electronic component 5 can be obtained.

  Below, the manufacturing method of the ceramic member 1 which is a part of structure of the electronic component of this embodiment is demonstrated.

  As shown in FIG. 10, a method of manufacturing a ceramic member 1 in which a recess 2 is formed on at least one main surface 1 a of the ceramic member 1 by heat irradiation 20, wherein the ceramic member 1 is formed in a portion where the recess 2 is formed. It is a manufacturing method which forms the said recessed part 2 by performing the heat irradiation 20 of temperature lower than melting | fusing point of a main component and higher than melting | fusing point which at least 1 type of component has.

  More specifically, when the main component ceramic is alumina and the additive is silica, magnesia or calcia, the melting point of alumina is about 2050 ° C., the melting point of silica is about 1730 ° C., and the melting point of magnesia is about 2800 ° C. , Calcia has a melting point of about 2570 ° C. In the conventional laser processing, since it is processed by heat irradiation at about 3500 ° C., all of the main component alumina, the additive silica, magnesia, and calcia melt, and partly sublimates, so that the melt is formed. The molten material is scattered around to form a welded material, and the melted material is cooled and solidified in the processed portion to form a molten layer, and microcracks are generated in the molten layer by the cooling.

  However, according to the method for manufacturing a ceramic member that is a part of the configuration of the electronic component of the present embodiment, the melting point of the additive is less than the melting point of the ceramic (alumina) and lower than the melting point of alumina. In order to irradiate the processed portion of the ceramic member 1 with the heat irradiation 20 that is equal to or higher than the melting point of silica, since the alumina does not substantially melt, the above-described molten layer or welded material is not formed, and therefore microcracks are generated. While being able to suppress, the recessed part 2 of a desired shape can be formed.

  FIG. 11A shows a cross-sectional shape after processing the recess 2 processed by the method of the present embodiment, and the recess 2 is closed by the altered layer 18. The generation mechanism of the deteriorated layer 18 is that the additive is only melted by heat irradiation at a temperature higher than the melting point of one or more of the additives constituting the vitreous without melting the main component ceramic. The deteriorated layer 18 having weak adhesion is formed by moving and partially sublimating. That is, the alteration layer 18 has a distribution of additive components having a melting point higher than the melting point of the ceramic among the ceramics constituting the main components and the vitreous additive, but the entire ceramic member 1 is not changed. The component having a lower melting point than the ceramic of the additive component constituting the glassy material, which was uniformly distributed, was moved by the heat irradiation described above, and the distribution was uneven in the ceramic member 1. Therefore, the composition is different from that of the molten layer 116 and is also defined as a deteriorated layer because it is different from the matrix of the ceramic member 1 before heat irradiation.

  Since the portion of the above-mentioned deteriorated layer 18 is exposed to the heat irradiation 20 at a high speed, the unheated portion and a clear boundary line are formed by the cracks 19, so that the deteriorated layer 18 is loosely clogged in the recess 2. It has become. Therefore, the altered layer 18 can be easily removed by a physical method such as ultrasonic vibration, and FIG. 11B shows a layer obtained by removing the altered layer 18 from the recess 2.

  Further, the moving speed s of the heat irradiation is preferably 10 m / min or more. This utilizes the characteristics of ceramics that are inferior in thermal conductivity. By moving the heat irradiation part at high speed, a crack 19 is generated due to the difference in thermal stress between the heat irradiation part and the other part. If the speed s is less than 10 m / min, it is difficult to form a crack accurately. When the heat irradiation is performed by the laser beam 21, the processing start generally enters from the dummy part and proceeds to the processing of the product part, but the width of the dummy part is large and the processing table moving speed s is a predetermined speed. Therefore, when the allowance is taken into consideration, a more preferable value of the moving speed is 13 m / min or more. Here, the upper limit value of the moving speed s is actually limited by the capability (moving speed, accuracy) of the processing table, and in the manufacturing method of the present embodiment, the desired recess 2 is obtained even at 100 m / min. It has been confirmed that can be formed.

  Moreover, it is preferable to use the laser beam 21 as the heat used for the heat irradiation 20 of the manufacturing method of the ceramic member 1 which is a part of the structure of the electronic component of this embodiment. The heat irradiation 20 by the laser beam 21 can be performed with high dimensional accuracy, and therefore the recess 2 having a desired shape can be formed.

The laser light 21 is preferably a CO ₂ laser. This is because the heat irradiation 20 is performed with the laser beam 21 by the CO ₂ laser, so that even a thick ceramic member 1 can be processed with a stable depth at a high speed.

  Next, a manufacturing method in which the concave portion of the ceramic member 1 which is a part of the configuration of the electronic component of the present embodiment is produced with laser light will be described in detail.

  A pulse waveform 24 generally used for ceramic processing is shown in FIG. Since ceramic processing requires high energy, a pulse waveform 24 that rises quickly after laser oscillation has been used. Specifically, after laser oscillation, 20 to 60 μsec. Thus, a desired shape is processed by melting and sublimating the ceramic at a temperature of about 3500 ° C., reaching the maximum temperature range 24a. Therefore, at this time, a problem peculiar to laser processing that a ceramic melt layer and a welded product are generated has occurred. In contrast, the method of the present embodiment uses a laser waveform 24 that rises slowly as shown in FIG. 12B. Specifically, the time until the maximum temperature 24a is reached after laser oscillation is obtained. 90-200 μsec. It is said.

Further, as shown in the ON-OFF signal diagram of FIG. 13A, one pulse of the oscillation signal 25 of the normal laser is set to 200 μsec. In comparison with the manufacturing method of this embodiment, the manufacturing method of the present embodiment has a total oscillation time of about 60 μsec. For one pulse of the normal signal 25, for example, the ON signal 26 is two short pulses. And about 30% of the normal laser oscillation signal 25. By oscillating the pulse waveform 24 having a slow rise as described above with a short ON signal 26, the pulse waveform 24 having a slow rise within the short ON signal 26 is oscillated as shown in the schematic diagram of FIG. Heat irradiation with the maximum temperature 24a controlled can be performed. That is, it can be set to process at a temperature lower than the melting point a of the main component ceramic and higher than one or more melting points b of the additive components. In addition, the pulse period is 200 μsec. A pulse waveform approximating CW (Continuity Wave) continuous oscillation is obtained by the following, and the recess 2 processed by this is the recess 2 when viewed in plan, as described above with reference to FIGS.
It is possible to perform processing while suppressing the variation D in the distance between the boundary lines and the variation in the depth d.

  A ceramic member which is a part of the configuration of the electronic component of the present embodiment was created by the following method.

The main component ceramic is composed of 96% by mass of Al ₂ O ₃ alumina, and the balance is an additive and impurities. The additive is 2.0% by mass of SiO ₂ , 0.2% by mass of CaO and 1% of MgO. A ceramic green sheet was formed by adding 0.0 mass%, the ceramic green sheet was cut into a predetermined shape, and then fired at a temperature of about 1600 ° C. to produce a ceramic sintered body having a thickness of 0.635 mm.

  Next, as shown in FIG. 14 (a), a continuous dividing groove 3 having a processed width D of the recess 2 of 30 μm and a depth d of 100 μm is formed on the main surface 1′a of the ceramic sintered body 1 ′. did. The recess 2 was formed with a radius of curvature R at the tip of 5 μm.

Conditions of thermal irradiation, using a slowly rising pulse waveform 24 as shown in FIG. 12 (b) by a CO ₂ laser oscillator, as shown in FIG. 13, 30 .mu.sec time ON signal 26. And the pulse period is 100 μsec. (Frequency 10 kHz), moving speed s of laser light 21
Was 13 m / min. The number of samples is 5 sheets and sample numbers 1-5. Here, the temperature of heat irradiation which is an example of the present embodiment is about 1850 ° C. on the irradiation surface of the laser light 21.

  In addition, as a comparative example, the ceramic sintered body 1 ′ uses the same member as that of the present embodiment described above, and the dividing groove having the processing width D of the recess 102 of 90 μm, the depth d of 100 μm, and the tip curvature radius R of 5 μm. 103 was produced. Here, the depth d and the radius of curvature R of the tip indicate the processing depth with the laser beam 21 and indicate the processing depth d including the molten layer 116 deposited in the concave portions 2 and 102 after processing.

Heat irradiation conditions at this time, with rising early pulse waveform 24 as shown in FIG. 12 (a) by a CO ₂ laser oscillator, as shown in FIG. 13, 400 .mu.sec time ON signal 26. And the pulse period is 1050 μsec. The recess 102 was processed at a frequency of 952 Hz and a moving speed s of the laser light 21 of 8 m / min. The number of samples is 5 sheets and sample numbers 6-10. Here, the temperature on the irradiation surface of the laser beam 21 of the comparative example is about 3400 ° C.

  The presence or absence of the molten layer 116 of the surface layer portions 2a and 102a having a length H of 12 mm in the longitudinal direction of the recesses 2 and 102 processed under the above-described conditions was visually confirmed. Next, similarly, the SEM analysis is performed on the difference between the maximum value and the minimum value of the width D when the length H of the recesses 2 and 102 is 12 mm, that is, the variation in the distance between the boundary lines of the main surfaces 1a and 101a and the recesses 2 and 102. (Scanning electron microscope analysis) was performed and calculated from SEM photographs. In addition, the depth d of the recesses 2 and 102 is divided along the recesses 2 and 102, and then the fracture surface is vapor-deposited and an SEM analysis photograph is taken. When the length H of the recesses 1 and 102 is 12 mm, the maximum The difference between the depth and the minimum depth was calculated from the above photograph and used as the variation in depth d.

  Further, the bending load of the concave portions 2 and 102 is set such that the span L is 28 mm around the concave portions 2 and 102 in the ceramic members 1 and 101 before the division of the SEM analysis as shown in FIG. A pressing load P was applied from the back side of Nos. 2 and 102, and the bending load when divided was measured. The width H of the sample was 12 mm and the pressing speed was 5 mm / min. The above measurement results are shown in Table 1.

  As can be seen from Table 1, in sample numbers 6 to 10 which are comparative examples, the surface layer portion 102a of the recess 102 has a molten layer 116, and although not shown in the table, Since the welded material 118 was adhered, the molten layer 116 and the welded material 118 were cooled after the laser processing, and microcracks were generated at the periphery of the surface layer portion 102a and the recess 102.

  On the other hand, in sample numbers 1 to 5 which are examples of the ceramic member 1 which is a part of the configuration of the electronic component of the present embodiment, the molten layer 116 does not substantially exist in the surface layer portion 2a of the recess 2. In addition, although not shown in the table, the welded material 118 did not adhere to the periphery of the recess 2, so that the recess 2 substantially free of microcracks could be formed.

  Further, the variation in the width D of the recesses 2 and 102, that is, the variation in the distance between the boundary lines, the average value of the sample numbers 6 to 10 in the comparative example was 52 μm, and the variation in the depth was 35 μm, both of which were large values. On the other hand, the average value of the boundary line variation of sample numbers 1 to 5, which is an example of the ceramic member 1 which is a part of the configuration of the electronic component of the present embodiment, is 2.4 μm, and the average value of the variation in depth. Was 5.8 μm, so when divided by the concave portion 2, the surface layer portion 2a of the concave portion 2 had few surface irregularities and the surface state was good, and the variation in depth was small. There was not.

  Furthermore, the bending load of the concave portions 2 and 102 is 6.4N as the average value of the sample numbers 6 to 10 of the comparative example and the variation thereof is large, but the ceramic member which is a part of the configuration of the electronic component of the present embodiment Since the average value of Sample Nos. 1 to 5 as an example of Example 1 was a low value of 5.0 N and the variation was small, the splitting property in the concave portion 2 was improved.

  From these, the ceramic member 1 which is a part of the configuration of the electronic component of the present embodiment has no adhesion of the melted layer 116 or the welded material 118, which has been a defect of the conventional laser processing, and has a high-accuracy dimension. Since the position-accurate recess 2 can be formed and the depth d thereof can be produced with less variation that cannot be obtained by conventional laser processing, it is suitable as the dividing groove 3 or the semiconductor chip mounting recess 4.

  Next, using the ceramic member 1 described above, a sample in which the moving speed s of heat irradiation was changed was created, and the generation of the molten layer 116 and the welded material 118 was confirmed.

The processing conditions of the ceramic member 1 which is a part of the configuration of the electronic component of the present embodiment are the same as in Example 1, and the processing width D of the recess 2 is 30 μm on the main surface 1′a of the ceramic sintered body 1 ′. the split groove 3 depth d are continuous in 100 [mu] m, using a slowly rising pulse waveform 24 Do shown in FIG. 12 (b) by a CO ₂ laser oscillator, as shown in FIG. 13, the time oN signal 26 3
0 μsec. And the pulse period is 100 μsec. Irradiation was performed at a frequency of 10 kHz, and the moving speed s of the laser beam 21 was 8, 9, 10, 13, 20 m / min, and the sample numbers were 11 to 15, respectively.

  Here, the temperature on the irradiation surface of the laser beam 21 is about 2100 to 2200 ° C. for sample numbers 11 and 12, while it is about 1770 to 1950 ° C. for sample numbers 13 to 15.

  Then, the presence / absence of the melted layer 116 in the recess 2 of the processed ceramic member 1 and the presence / absence of the adhesion of the welded material 118 to the peripheral edge of the recess 2 were visually confirmed. The results are shown in Table 2.

  As can be seen from Table 2, in Sample No. 11 where the moving speed s of the laser beam 21 is 8 m / min, the melted layer 116 exists in the recess 2, and the welded material 118 adheres to the periphery of the recess 2. It was. Sample No. 12 having a moving speed s of 9 m / min had the melted layer 116 in the concave portion 2, but the welded material 118 did not adhere to the periphery of the concave portion 2. In Sample Nos. 13 to 15 having the moving speed s of 10 m / min or more, there was no existence of the melted layer 116 in the concave portion 2 and adhesion of the welded material 118 to the peripheral edge of the concave portion 2.

  If the moving speed s of the heat irradiation is made slower than this, the rise is slow as described above, and even in the method of processing with the pulse waveform 24 that is cut in the middle of the rise, the main component ceramic starts to melt, It can be seen that a deposit is generated. Therefore, the preferable moving speed s of heat irradiation is 10 m / min or more, and more preferably 13 m / min or more in consideration of the error range.

  Using the same ceramic sintered bodies 1 ′ and 101 ′ as those used in Example 1, the recesses 2 and 102 were formed under the following conditions.

  The ceramic member 1 which is a part of the configuration of the electronic component of the present embodiment is a continuous division in which the processing width D of the recess 2 is 30 μm and the depth d is 100 μm on the main surface 1a ′ of the ceramic sintered body 1 ′. The concave portion 2 was produced in the groove 3 at a moving speed s of the laser beam 21 of 13 m / min.

The conditions of the heat irradiation were such that the pulse waveform 24 with a slow rise of CO ₂ laser oscillation as shown in FIG. 12B was used, and the sample number 16 was set to 38 μsec. And the pulse period is 125 μsec. (Frequency 8000 Hz) Sample No. 17 has a time of 33 μsec. And the pulse period is 111 μsec. (Frequency 9000 Hz) Sample No. 18 has a time of 30 μsec. And the pulse period is 100 μsec. (Frequency: 10 kHz) Sample No. 19 sets the time of the ON signal 26 to 25 μsec. And the pulse period is 87 μsec. (Frequency 11.5 kHz) Sample No. 20 has the above ON signal 26 time of 23 μsec. And the pulse period is 77 μsec. (
Processing was performed at a frequency of 13 kHz. Here, as for the temperature on the irradiation surface of the laser beam 21, the sample number 16 is about 2030 ° C, while the sample numbers 17 to 20 are about 1770 ° C to 1950 ° C.

Further, as a comparative example, using the same ceramic sintered body 1 ′, a dividing groove 103 having a processing width D of the recess 102 of 90 μm and a depth d of 100 μm is formed on the main surface of the ceramic sintered body 1 ′. The moving speed s of the laser beam 21 was 8 m / min. The heat irradiation conditions at this time were as follows: a pulse waveform 24 having a fast rise as shown in FIG. 12A by CO ₂ laser oscillation, and the ON signal 26 time being 400 μsec. And the pulse period is 1050 μsec. Sample No. 21 was processed at a frequency of 952 Hz and the temperature on the irradiated surface was about 3400 ° C.

  Then, EPMA analysis was performed on the surface layer portions 2a and 102a of the recesses 2 and 102 formed in each sample, thereby confirming the dispersion state and the ratio of the additive. Here, the case where silica is present in an area of 95% or less with respect to the area of the surface layer portions 2a and 102a of the recesses 2 and 102 is defined as uneven distribution. In addition, the ratio of the area of the silica which exists in the surface and cross section of the ceramic member 1 which does not add heat irradiation etc. is about 97 to 99% in a certain arbitrarily defined area.

  Next, the variation in the distance between the boundary lines with the main surfaces 1a and 101a and the variation in the depth over the width 12 mm of the recesses 2 and 102 were measured in the same manner as in Example 1.

  Next, after dividing along the concave portions 2 and 102, as shown in FIG. 15, electrodes 17 were formed on the surface layer portions 2a and 102a of the divided concave portions 2 and 102 for testing. The paste of the electrode 17 was applied using a commercially available copper electrode paste so as to have a film thickness of 50 μm, and baked at 850 ° C. in a nitrogen reduction furnace. The paste contains silica as an additive.

  Then, an aluminum cylinder having a diameter of 1 mm was adhered to the surface of the electrode 17 with a resin adhesive, and the peel strength between the electrode 17 and the ceramic members 1 and 101 was measured by pulling in a direction perpendicular to the surface of the electrode 17. . The pulling speed was 1 mm / min. The above results are shown in Table 3.

  As can be seen from Table 3, silica is uniformly dispersed in the surface layer portion 102a of the concave portion 102 of the sample number 21, which is a comparative example, and the dispersion corresponds to 99% of the area of the surface layer portion 102a. In Sample No. 16, the silica distribution ratio was 90%, and the degree of uneven distribution was small. Thereby, in sample number 21, since the action of mutual diffusion between the silica and the silica in the electrode 17 was small, the peel strength of the electrode with respect to the ceramic member 1 was lowered, and the adhesion strength between the ceramic member 1 and the electrode 17 was remarkably high. Declined.

On the other hand, as for the sample numbers 17-20 which are the examples of the ceramic member 1 which is a part of the structure of the electronic component of this embodiment, the said silica is unevenly distributed by the surface layer part 102a, and the ratio of the distribution is It occupied 50 to 80% in the area of the surface layer part 102a. Thereby, since the adhesion strength between the ceramic member 1 and the electrode 17 was improved by the mutual diffusion of the silica and the silica in the electrode 17, the peel strength of the electrode 17 was increased.

  On the other hand, Sample No. 16 had a large variation in the distance between the boundary lines of 4.3 μm and the depth variation of 8.4 μm, so that printing failure occurred at the time of electrode formation, and the peel strength of the electrode slightly increased.

  However, the variation in the distance between the boundary lines of sample numbers 17 to 20 was 4.0 μm or less, and the depth variation was 8.0 μm or less. The strength was further improved.

  Also, electrodes were formed in the recesses 2 and 102, and the test results of the peel strength showed that the sample number 21 of the comparative example was 30 MPa and the sample number 16 was 39 MPa, and part of the configuration of the electronic component of this embodiment Sample numbers 17 to 20 which are examples of the ceramic member 1 are 45 to 54 MPa, and if the surface layer portion of the concave portion is unevenly distributed, for example, if the end face electrode is formed after dividing the concave portion, the electrode adhesion It can be seen that the strength can be improved. Further, although not in the examples, when the concave portion of the ceramic member, which is a part of the configuration of the electronic component of the present embodiment, is used as the concave portion for mounting the semiconductor chip, there is no molten layer in the surface layer portion of the concave portion, and the periphery thereof Since the welded material is not attached, when a thin metal layer is deposited as the base layer, a high anchor effect can be obtained, and there is no possibility that the delicate circuit pattern on the periphery is disconnected by the welded material.

DESCRIPTION OF SYMBOLS 1,101: Ceramic member 2,102: Concave part 2a, 102a: Surface layer part 3,103: Dividing groove | channel 4,104: Concave part for mounting a semiconductor chip 5,105: Electronic component 6,106: Circuit pattern 7,107: End surface Electrode 8: Thin film metal layer 9: Metal layer 10: Semiconductor chip 11: Conductor 13: Crystal particle 14: Glassy 15: Additive 16: Ceramic substrate 17: Electrode 18: Altered layer 19: Crack 20: Thermal irradiation 21: Laser beam 22: Condensing lens 23: Processing table 24: Pulse waveform 25: Normal laser oscillation signal 26: ON signal 116: Molten layer 117: Micro crack 118: Welded material 119: Crack

Claims (1)

Al ₂ O ₃ 91~99.6 wt%, _SiO 2, MgO, forming a conductor on at least one major surface of the ceramic substrate comprising a CaO, side surface portion of the front Symbol ceramic substrate as an additive component Si is present in 50 to 90% of the surface area of the side surface layer portion, an electrode containing Si is formed on the side surface layer portion, and the electrode and the conductor are electrically connected. Electronic components.