JP2011176186A - Multilayer ceramic capacitor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a multilayer ceramic capacitor which has high dielectric constant and is excellent in DC bias characteristics and DC aging characteristics while satisfying X5R characteristics in EIA standards. <P>SOLUTION: A dielectric layer 5 is made of a dielectric porcelain containing barium titanate as a main component and having calcium, magnesium and a rare earth element, wherein main crystal particles comprise crystal particle having a core shell structure, the crystal particles each have a first crystal particle having calcium concentration of <0.3 atom% and a second crystal particle having calcium concentration of ≥0.3 atom% and each have an average particle size of 0.15-0.4 μm, and the crystal structure of a core portion 9a is a tetragonal system and the crystal structure of a shell portion 9b surrounding the core portion 9a is a cubic system, and the average thickness of the shell portion 9b which is calculated based on X-ray diffraction information is 5-15 nm. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a multilayer ceramic capacitor that is made of crystal grains mainly composed of barium titanate and can be thinned.

  Conventionally, barium titanate has been used as a dielectric material for multilayer ceramic capacitors because of its high relative dielectric constant, and inexpensive base metals (such as Ni) have been used for the internal electrode layers of multilayer ceramic capacitors. It is used. When the dielectric layer mainly composed of barium titanate and the internal electrode layer are fired at the same time, it is necessary to reduce the oxygen partial pressure (for example, 0.03 Pa or less at 1300 ° C.) in order not to oxidize Ni. In this case, there is a problem that the dielectric layer is reduced and the insulating property is lowered, so that practical characteristics cannot be obtained.

Therefore, in the case of the X5R characteristic (or JIS standard B characteristic) of the EIA standard, as a dielectric material, for example, barium titanate is a main component, and oxides of rare earth elements, Mn, V, Cr, Mo, Fe, Reduction-resistant dielectric ceramic compositions to which acceptor-type and donor-type element compounds such as Ni, Cu, and Co are added are used. A dielectric crystal particle (for example, BaTiO ₃ ) in which a plurality of additive components are solid-dissolved in barium titanate has a tetragonal core part (usually pure BaTiO ₃ ) and a core part. And a shell portion in which the added component is dissolved (see, for example, Patent Document 1). Here, the X5R characteristic of the EIA standard indicates that the rate of change of the electrostatic capacity is within ± 15% with respect to 25 ° C. in the temperature range of −55 to 85 ° C. Means that the rate of change in capacitance is within ± 10% when 20 ° C. is used as a reference in the temperature range of −25 to 85 ° C.

  Furthermore, in recent years, with the trend toward smaller and higher density electronic circuits, multilayer ceramic capacitors are also required to be smaller and larger in capacity while satisfying the X5R characteristics of the EIA standard. In order to reduce the number of dielectric layers, the number of dielectric layers is increasing and the thickness is being reduced.

However, when the dielectric layer is thinned, the electric field strength per layer increases, and the multilayer ceramic capacitor has a problem that the capacitance decreases when a DC voltage is applied (DC bias characteristics). Here, the DC bias characteristic is ((C ₁ − −) when C _{0 is} the capacitance when no DC voltage is applied to the multilayer ceramic capacitor, and C ₁ is the capacitance when a DC voltage is applied. C ₀ ) / C ₀ ) × 100 (%)).

  For this reason, in the multilayer ceramic capacitor, the effective capacitance when a DC voltage is applied together with an AC voltage has been emphasized, and improvements for improving the DC bias characteristics of the multilayer ceramic capacitor have been made for many years. (For example, Patent Documents 2 to 5).

JP 2001-230150 A JP 2001-316176 A JP 2006-111468 A JP 2008-162818 A JP 2009-238885 A

  However, the problems associated with the thinning of the dielectric layer of the multilayer ceramic capacitor are not limited to the above-described deterioration of the DC bias characteristics accompanying the increase of the electric field strength per layer, but the multilayer ceramic capacitor is also subject to direct current. When voltage is continuously applied, there is also a problem that the capacitance gradually decreases (DC aging characteristics). Here, the DC aging characteristic is a rate of change (decrease) in capacitance with time when a constant DC voltage is continuously applied to the multilayer ceramic capacitor.

  Accordingly, an object of the present invention is to provide a multilayer ceramic capacitor having a high dielectric constant and excellent in DC bias characteristics and DC aging characteristics while satisfying the X5R characteristics of the EIA standard.

The multilayer ceramic capacitor of the present invention is provided on a capacitor body in which a plurality of dielectric layers and a plurality of internal electrode layers made of nickel as a conductive material are alternately stacked, and on the end face of the capacitor body where the internal electrode layers are exposed. A multilayer ceramic capacitor having a plurality of external electrodes, wherein the dielectric layer is composed mainly of barium titanate, calcium, magnesium, and at least one rare earth selected from yttrium, dysprosium, holmium, terbium and ytterbium. Element (RE), and the total amount of barium and calcium is 100 mol, the magnesium is 0.5 to 1.2 mol in terms of MgO, and the rare earth element (RE) is 0 in terms of RE ₂ O _3. 0.5 to 1.0 mol, and the main crystal particles are mainly composed of the barium titanate. A core-shell structure in which the calcium, the magnesium, and the rare earth element (RE) are solid-solved, the first crystal particles having a calcium concentration of less than 0.3 atomic%, and the calcium concentration of 0.3 atomic% or more The crystal grains have an average particle size of 0.15 to 0.4 μm, the core portion has a tetragonal crystal structure, and a shell portion surrounding the core portion. The crystal structure is cubic and the average thickness of the shell portion calculated based on the X-ray diffraction information is 5 to 15 nm, and the area of the first crystal particle seen on the polished surface of the dielectric ceramic is S1 is characterized in that it is composed of a dielectric ceramic having S2 / (S1 + S2) of 0.4 to 0.8, where S2 is the area of the second crystal grain.

  In the multilayer ceramic capacitor, it is desirable that the dielectric ceramic does not contain vanadium.

  In the multilayer ceramic capacitor, it is desirable that the dielectric ceramic does not contain manganese.

  According to the present invention, it is possible to obtain a multilayer ceramic capacitor having a high dielectric constant and excellent DC bias characteristics and DC aging characteristics while satisfying the X5R characteristics of the EIA standard.

(A) is a schematic sectional drawing which shows an example of the multilayer ceramic capacitor of this invention, (b) is an enlarged view of an inside. FIG. 2 is a schematic cross-sectional view showing the internal structure of crystal grains mainly composed of barium titanate in a dielectric ceramic that is a dielectric layer constituting the multilayer ceramic capacitor of the present invention.

  The multilayer ceramic capacitor of the present invention will be described in detail based on the schematic sectional view of FIG. FIG. 1A is a schematic sectional view showing an example of the multilayer ceramic capacitor of the present invention, and FIG. 1B is an enlarged view of the inside. FIG. 2 is a schematic cross-sectional view showing the internal structure of crystal grains mainly composed of barium titanate in a dielectric ceramic that is a dielectric layer constituting the multilayer ceramic capacitor of the present invention.

  In the multilayer ceramic capacitor of this embodiment, external electrodes 3 are formed at both ends of the capacitor body 1. The external electrode 3 is formed, for example, by baking Cu or an alloy paste of Cu and Ni.

  The capacitor body 1 is configured by alternately laminating dielectric layers 5 made of dielectric ceramics and internal electrode layers 7. In FIG. 1, the laminated state of the dielectric layer 5 and the internal electrode layer 7 is shown in a simplified manner, but the multilayer ceramic capacitor of this embodiment has several hundreds of dielectric layers 5 and internal electrode layers 7. It is a laminate.

  The dielectric layer 5 made of dielectric porcelain is composed of crystal grains 9 and grain boundary phases 11, and the thickness is preferably 3 μm or less, particularly 2 μm or less, thereby reducing the size and increasing the capacity of the multilayer ceramic capacitor. It becomes possible to do. If the thickness of the dielectric layer 5 is 0.5 μm or more, it becomes possible to stabilize the temperature characteristics of the capacitance.

  The internal electrode layer 7 is preferably made of nickel (Ni) in that the manufacturing cost can be suppressed even when the internal electrode layer 7 is made highly laminated, and simultaneous firing with the dielectric layer 5 can be achieved.

  In the multilayer ceramic capacitor of this embodiment, the dielectric ceramic constituting the dielectric layer 5 is mainly composed of barium titanate, and at least one selected from calcium, magnesium, yttrium, dysprosium, holmium, terbium and ytterbium. Rare earth element (RE).

  Further, this dielectric porcelain is composed of 0.5 to 1.2 mol of magnesium with respect to 100 mol of barium and calcium in total, and at least one rare earth element selected from yttrium, dysprosium, holmium, terbium and ytterbium. In an amount of 0.5 to 1.0 mol.

  Further, the crystal grains 9 as the main crystal grains constituting the dielectric ceramic have an average particle diameter of 0.15 to 0.4 μm.

  The crystal particles 9 constituting the dielectric ceramic are composed of first crystal particles 9A having a low calcium concentration and second crystal particles 9B having a higher calcium concentration than the first crystal particles 9A. The first crystal particle 9A is a crystal particle mainly composed of barium titanate having a calcium concentration lower than 0.3 atomic%, and the second crystal particle 9B has a calcium concentration of 0. 3. Crystal grains mainly composed of 3 atomic% or more of barium titanate, and the area of the first crystal grains seen on the polished surface of the dielectric ceramic is S1, and the area of the second crystal grains is S2. S2 / (S1 + S2) is 0.4 to 0.8.

  Further, the first crystal particle 9A and the second crystal particle 9B constituting the dielectric porcelain each have a core part 9a having a tetragonal crystal structure, and surrounding the core part 9a, magnesium and rare earth elements (RE). The additive component is solid-solved, the crystal structure is composed of a cubic shell portion 9b, and the average thickness of the shell portion 9b calculated based on X-ray diffraction information is 5 to 15 nm.

  A dielectric ceramic that constitutes the multilayer ceramic capacitor has the above-described composition and includes the first crystal particles 9A and the second crystal particles 9B at a specific ratio, and the first crystal particles 9A and the second crystals When the crystal structure of the particle 9B is a core-shell structure including a tetragonal core portion 9a and a cubic shell portion 9b, and the average thickness t of the shell portion 9b is within a specific range, a multilayer ceramic capacitor is formed. The dielectric layer 5 has a relative dielectric constant of 3650 or more at room temperature, and the capacitance temperature characteristic is an EIA standard X5R characteristic (in the temperature range of −55 to + 85 ° C. The capacitance change rate is within ± 15%) and DC bias characteristics (with respect to the capacitance when no DC voltage is applied to the multilayer ceramic capacitor) DC aging characteristics (rate of change in capacitance with time when a constant DC voltage is continuously applied to the multilayer ceramic capacitor) is within -55%. Can be a multilayer ceramic capacitor having excellent dielectric properties of -20% or less. The second crystal particles 9 are in a state in which calcium is substantially uniformly distributed and dissolved in the entire crystal particles. Further, the first crystal particle 1A includes one having a calcium concentration of zero.

  Here, the DC bias characteristic larger than −55% and the DC aging characteristic larger than −20% mean that the change in capacitance becomes larger on the negative side.

  That is, the content of each element contained in the dielectric ceramic serving as the dielectric layer 5 in the multilayer ceramic capacitor of this embodiment is such that the magnesium content is 0.5 mol in terms of MgO with respect to 100 mol of the total amount of barium and calcium. If less, no shell part is formed in the crystal grain 9 and the core shell structure is not formed. Therefore, the temperature characteristic of the capacitance is increased, and the X5R characteristic of the EIA standard is not satisfied, and the DC bias characteristic is reduced. It becomes larger than −55%, and the DC aging characteristic becomes larger than −20%.

  On the other hand, when the content of magnesium with respect to 100 mol of the total amount of barium and calcium is more than 1.2 mol in terms of MgO, the relative dielectric constant at room temperature decreases from 3650 and the DC aging characteristic becomes larger than −20%. .

Next, when the content of the rare earth element (RE) with respect to 100 mol of the total amount of barium and calcium is less than 0.5 mol in terms of RE ₂ O ₃ , the shell portion is not formed in the crystal particles 9 in this case as well. As a result, the core-shell structure is not provided, the capacitance temperature characteristic does not satisfy the X5R characteristic of the EIA standard, the DC bias characteristic becomes larger than −55%, and the DC aging characteristic becomes larger than −20%.

On the other hand, when the content of the rare earth element (RE) with respect to 100 mol of the total amount of barium and calcium is more than 1.0 mol in terms of RE ₂ O ₃ , the thickness of the shell portion 9b becomes thicker than 15 nm. As the dielectric constant falls below 3650, the DC aging characteristics become greater than -20%.

The multilayer ceramic capacitor of this embodiment may contain vanadium together with magnesium and rare earth elements (RE) in a range of 0.1 mol or less in terms of V ₂ O _{5 with} respect to 100 mol of the total amount of barium and calcium. Although it is good, it is preferable that it does not contain vanadium. When the dielectric porcelain does not contain vanadium, the thickness of the shell portion 9b in the crystal particle 9 is reduced to 8 nm or less, and the relative dielectric constant of the dielectric layer 5 at room temperature can be increased to 3800 or more. The aging characteristic can be set to -12.28% or less.

Further, in the multilayer ceramic capacitor of this embodiment, the dielectric ceramic serving as the dielectric layer 5 is 0.2 mol in terms of MnO with respect to 100 mol of the total amount of barium and calcium together with magnesium and rare earth elements (RE). Although it may be contained in the following range, it is preferable that it does not contain manganese. In the multilayer ceramic capacitor of the present embodiment, when the dielectric ceramic does not contain manganese, the thickness t of the shell portion 9b is 5.5 to 6.8 nm, and thereby the DC aging characteristic is −9.34%. It can be:

  In the multilayer ceramic capacitor of this embodiment, it is important that the average thickness of the shell portion 9b of the crystal particle 9 having the core-shell structure is 5 to 15 nm, and when the average thickness of the shell portion 9b is thinner than 5 nm. Although the relative dielectric constant at room temperature is 3650 or more, the temperature change rate of the capacitance is larger than ± 15%, which does not satisfy the X5R characteristic of the EIA standard. On the other hand, when the average thickness of the shell portion 9b is greater than 15 nm, the DC aging characteristics exceed −20%.

  In the multilayer ceramic capacitor of this embodiment, it is important that the average particle size of the crystal particles 9 constituting the dielectric ceramic that is the dielectric layer 5 is 0.15 to 0.4 μm. When the average particle diameter of the crystal particles 9 is smaller than 0.15 μm, the relative dielectric constant at room temperature becomes lower than 3650. In this case, the average particle diameter of the crystal particles 9 is small, and the average of the shell portion 9b is diffused by the diffusion of the additive element. When the thickness is thicker than 15 nm, the DC aging characteristic exceeds −20%. On the other hand, when the average grain size of the crystal grains 9 is larger than 0.4 μm, the DC bias characteristic becomes larger than −55%.

  Further, in the multilayer ceramic capacitor of this embodiment, when the area of the first crystal particle 9A seen on the polished surface of the dielectric ceramic is S1, and the area of the second crystal particle 9B is S2, S2 / (S1 + S2 ) Is 0.4 to 0.8. In this case, when S2 / (S1 + S2) is smaller than 0.4 or S2 / (S1 + S2) is larger than 0.8, the relative dielectric constant at room temperature is lower than 3650.

  In addition, the multilayer ceramic capacitor of this embodiment has a glass component and other additive components in an amount of 4% by mass or less in the dielectric ceramic as an aid for improving the sinterability as long as desired dielectric characteristics can be maintained. You may make it contain in a ratio.

  Here, the calcium concentration in the crystal particles (hereinafter sometimes referred to as Ca concentration) is measured using a transmission electron microscope apparatus provided with an energy dispersive analyzer (EDS). In this case, with respect to the crystal particles projected on the cross section of the polished dielectric porcelain, an arbitrary location in the vicinity of the center portion of the crystal particles is analyzed using EDS, and the main component element detected from the crystal particles ( The total content of Ba, Ti, Ca) and each additive element (V, Mg, Mn, RE) is taken as 100%, and the content of Ca contained therein is determined. And it classify | categorizes into the 1st crystal particle 9A in which Ca concentration is lower than 0.3 atomic%, and the 2nd crystal particle 9B in which Ca concentration is 0.3 atomic% or more. The ratio between the first crystal particle 1A having a Ca concentration lower than 0.3 atomic% and the second crystal particle 1B having a Ca concentration of 0.3 atomic% or more is the ratio of the cross section of the crystal particle reflected in the transmission electron micrograph. Obtained from the area ratio. In this case, in the area of the photograph taken with the transmission electron microscope, the area of the first crystal particle 9A is S1, the area of the second crystal particle 9B is S2, and the ratio is obtained from the S2 / (S1 + S2) ratio. The number of crystal grains 9 to be evaluated is about 100. Further, the vicinity of the center portion of the crystal grains refers to a region on the inner side of a depth of 1/3 of the diameter from the grain boundary of the crystal grains, which is seen on the polished surface of the sample obtained by polishing the dielectric ceramic.

  Further, in the present invention, a method for measuring the average thickness t of the shell portion 9b of the crystal particle 9 is based on an evaluation method using an X-ray diffraction method disclosed in JP-A-2006-137647, as follows. It can be obtained from the equation.

The crystal structure of the target crystal grain 9 is compared with the reflection of a pure tetragonal crystal (h k l) or cubic crystal (h ′ k ′ l ′) in the measured X-ray diffraction pattern. It is assumed that the reflection of the tetragonal crystal (h k l) and the cubic crystal (h ′ k ′ l ′) is included from the identification of the peak position. In this case, the target diffraction data is tetragonal (h k l) and cubic (h ′ k ′ l ′) reflections. Then, parameters such as peak intensity (= integrated intensity), peak top 2θ position, half-value width, peak shape function, and the like are obtained from the diffraction peak.

Next, the peak half-value width (B) of the cubic crystal (h ′ k ′ l ′) obtained from the diffraction peak
Is substituted into equation (2) to determine the true half-value width (β), and then the true half-value width (β) and the Bragg angle (θ) obtained from the diffraction data are expressed in equation (1). The thickness (d _obs ) of the shell portion is determined by substitution, and this is set as the average thickness t of the shell portion. At that time, peak separation is performed as necessary. In this case, the conditions for peak separation are as follows: background function: 0th order polynomial, synchrotron radiation: Kα1, profile function: the psedo-Voigt function, half width: different half width for all reflections, object of profile: object, Data resolution: Sharp (minimum half-value width: about 0.1 °). Commercially available software can be used for peak separation, and the tools for peak separation are not particularly limited.

  Moreover, the average particle diameter of the crystal particle 9 (the first crystal particle 9A and the second crystal particle 9B) is measured by the following procedure. First, the fracture surface of the sample which is the capacitor body 1 after firing is polished. Thereafter, a photograph of the internal structure is taken of the polished sample using a scanning electron microscope, a circle containing 50 to 100 crystal particles 9 is drawn on the photograph, and the crystal particles 9 in and around the circle are drawn. select. Next, the contour of each crystal particle 9 is image-processed to determine the area of each crystal particle 9, the diameter when replaced with a circle having the same area is calculated, and the average value is determined.

In addition, the composition of the dielectric ceramic is such that a solution obtained by dissolving a multilayer ceramic capacitor in an acid is ICP.
It is determined using (Inductively Coupled Plasma) analysis and atomic absorption analysis. In this case, the amount of oxygen is determined using the valence of each element as the valence shown in the periodic table.

  Next, a method for manufacturing the multilayer ceramic capacitor of this embodiment will be described.

  First, a ceramic slurry is prepared by using a ball mill or the like together with a dielectric powder, an organic resin such as polyvinyl butyral resin, and a solvent such as toluene and alcohol. A ceramic green sheet is formed on the substrate. The thickness of the ceramic green sheet is preferably 1 to 5 μm from the viewpoint of reducing the thickness of the dielectric layer 5 to increase the capacity and maintaining high insulation.

As the main powder used in the method of manufacturing the multilayer ceramic capacitor of this embodiment, barium titanate powder (Ba / Ti molar ratio is 1.001 to 1.009, hereinafter referred to as BT powder) not containing calcium Barium titanate powder (hereinafter referred to as BCT powder) in which calcium is dissolved is used. The BCT powder represented by (Ba _1-x Ca _x ) TiO ₃ (X = 0.01 to 0.2) and having a Ba + Ca / Ti molar ratio of 1.001 to 1.009 is preferably used. . The average particle size of the BT powder and the BCT powder is preferably 0.06 to 0.2 μm. This facilitates the thinning of the dielectric layer 5, and from the mixed powder of BT powder and BCT powder, has a high dielectric constant under the firing conditions described later, satisfies the X5R characteristics of the EIA standard, Crystal grains 9 suitable for DC aging characteristics can be obtained.

  The dielectric powder used in manufacturing the multilayer ceramic capacitor of this embodiment is mainly composed of the above-mentioned mixed powder of BT powder and BCT powder, and contains all the components of magnesium, rare earth elements and sintering aid. Use quantitatively coated one.

  In this case, the dielectric powder to be used is prepared as follows. First, BT powder having a purity of 99.9% or more, a Ba / Ti molar ratio of 1.001 to 1.009, and an average particle size of 0.1 to 0.2 μm, and a purity of 99.9% or more , (Ba + Ca) / Ti molar ratio is from 1.001 to 1.009, and a mixed powder suspension with BCT powder having an average particle size of from 0.1 to 0.2 μm is used as a pH adjuster. Using water, the pH is in the range of 6-8, and at least one selected from lithium aqueous solution, silica sol, barium carbonate aqueous solution, magnesium hydroxide aqueous solution, calcium hydroxide aqueous solution and yttrium, dysprosium, holmium, terbium and ytterbium. An aqueous solution of seed rare earth elements is added and mixed in this order to prepare a ceramic slurry. Note that the purity of these raw material reagents is preferably 99.5% or more for the purpose of suppressing the mixing of impurities into the obtained dielectric ceramic and obtaining high dielectric properties.

  Next, the ceramic slurry is put into a spray dryer equipped with a four-fluid nozzle, droplets having a diameter of 10 μm or less are generated from the four-fluid nozzle, and dried at a temperature of about 200 ° C. A precursor is prepared, and then the dielectric powder precursor is prepared by heat treatment at a temperature higher than the temperature of the drying treatment.

The composition is 0.5 to 1.2 mol in terms of Mg contained in the magnesium hydroxide aqueous solution as MgO when the total amount of BT powder and BCT powder is 100 mol, yttrium, dysprosium. Of rare earth elements (RE) contained in an aqueous solution of at least one rare earth element (RE) selected from holmium, terbium and ytterbium in an amount of 0.5 to 1.0 mol in terms of RE ₂ O ₃ Add to make composition.

Moreover, when using a sintering auxiliary agent, the addition amount is prepared so that it may become 0.5-2 mass parts with respect to 100 mass parts of total amounts of BT powder and BCT powder. Thereby, the sinterability of the dielectric ceramic can be further enhanced. As the sintering aid, it is preferable to use a low melting point glass containing lithium. For example, the composition is Li ₂ O = 1-15 mol%, SiO ₂ = 40-60 mol%, BaO = 15-35 mol%. , And CaO = 5 to 25 mol% is preferable.

  Next, a rectangular internal electrode pattern is printed and formed on the main surface of the obtained ceramic green sheet. The conductor paste used as the internal electrode pattern is prepared by mixing Ni or an alloy powder thereof as a main component metal, mixing ceramic powder as a co-material with this, and adding an organic binder, a solvent and a dispersant. Further, in order to eliminate the step due to the internal electrode pattern on the ceramic green sheet, it is preferable to form the ceramic pattern with substantially the same thickness as the internal electrode pattern around the internal electrode pattern. In this case, it is preferable to use the dielectric powder used for the ceramic green sheet as the ceramic component constituting the ceramic pattern in that the firing shrinkage in the simultaneous firing is the same.

  Next, a desired number of ceramic green sheets with internal electrode patterns are stacked, and a plurality of ceramic green sheets without internal electrode patterns are stacked on top and bottom of the ceramic green sheets so that the upper and lower layers have the same number. Form the body. The internal electrode patterns in the temporary laminate are shifted by half patterns in the longitudinal direction. By such a laminating method, the internal electrode pattern can be formed so as to be alternately exposed on the end face of the cut laminate.

  The multilayer ceramic capacitor of the present invention has a method of laminating after the internal electrode pattern is formed in advance on the main surface of the ceramic green sheet. Is printed, dried, and the ceramic green sheet without the internal electrode pattern printed thereon is temporarily adhered to the printed and dried internal electrode pattern, and the adhesion of the ceramic green sheet and the printing of the internal electrode pattern are sequentially performed. It can also be formed by a construction method.

  Next, the temporary laminate is pressed under conditions of higher temperature and higher pressure than the temperature and pressure at the time of temporary lamination to form a laminate in which the ceramic green sheet and the internal electrode pattern are firmly adhered.

  Next, the capacitor body molded body in which the end portions of the internal electrode patterns are exposed is formed by cutting the laminate into a lattice shape.

  Next, the capacitor body 1 is formed by firing the capacitor body molded body in a predetermined atmosphere under temperature conditions. In some cases, the ridge line portion of the capacitor body 1 may be chamfered, and barrel polishing may be performed to expose the internal electrode layer 7 exposed from the opposite end surface of the capacitor body 1.

  Next, the obtained capacitor body molded body is degreased and fired. Firing is preferably performed in a hydrogen-nitrogen atmosphere at a temperature increase rate of 1500 to 4000 ° C./h, a maximum temperature of 1030 to 1200 ° C., a holding time of 0.1 to 4 hours. Then, the capacitor body 1 is obtained by performing reoxidation treatment in the temperature range of 900 to 1100 ° C. By performing the firing under such conditions, the average particle diameter of the crystal particles 9 constituting the dielectric layer 5 is in the range of 0.15 to 0.4 μm, and the crystal structure of the crystal particles 9 is a tetragonal core part. Capacitor body 1 can be obtained in which 9a and a cubic shell portion 9b that surrounds the core portion and in which at least magnesium and a rare earth element additive are in solid solution are formed, and the thickness of shell portion 9b is 5 to 15 nm.

As described above, in the multilayer ceramic capacitor of this embodiment, in order to produce the dielectric layer 5, the main component is barium titanate, and at least all the components of calcium, magnesium, rare earth elements, and sintering aid are contained therein. The obtained raw capacitor body molded body is fired under firing conditions with a high rate of temperature rise using a predetermined amount coated, whereby crystal grains 9 having a small average thickness of the shell portion 9b can be obtained. In contrast, even when barium titanate is used as a main component and at least a predetermined amount of calcium, magnesium, rare earth elements and sintering aids are coated, the rate of temperature increase is 1500 ° C./h. Under slower conditions, the thickness of the shell portion 9b becomes large, and even if barium titanate is the main component and at least a predetermined amount of magnesium and rare earth elements are coated on this, the glass powder is used as a sintering aid. In any case where the addition method is used, the DC aging characteristic is increased.

  Next, an external electrode paste is applied to the opposing ends of the capacitor body 1 and baked to form the external electrodes 3. In some cases, a plating film is formed on the surface of the external electrode 3 in order to improve mountability. Thus, the multilayer ceramic capacitor of the present invention can be obtained.

First, as a raw material powder, a BT powder having a purity of 99.9%, an average particle diameter of 0.06 to 2.0 μm, and a Ba / Ti molar ratio of 1.005 (Ba _0.97 Ca _{0. 0). 03} ) BCT powder represented by TiO ₃ , having a purity of 99.9%, an average particle size of 0.06 to 2.0 μm, and a Ba + Ca / Ti molar ratio of 1.005 was prepared at the ratio shown in Table 1. Mixed.

  Next, ammonia water was used as a pH adjuster in the suspension of the mixed powder of BT powder and BCT powder so that the pH was in the range of 6-8. Next, it is added at least one rare earth selected from lithium aqueous solution, silica sol, barium carbonate aqueous solution, magnesium hydroxide aqueous solution, calcium hydroxide aqueous solution, ammonium vanadate aqueous solution, manganese acetate aqueous solution and yttrium, dysprosium, holmium, terbium and ytterbium. An aqueous solution of the elements was added and mixed in this order to prepare a ceramic slurry.

Next, the ceramic slurry is put into a spray dryer equipped with a four-fluid nozzle, droplets having a diameter of 10 μm or less are generated from the four-fluid nozzle, and dried at a temperature of about 200 ° C. A precursor is prepared, and then the dielectric powder precursor is heat-treated at 400 ° C. to cover the surface of the BT powder with a predetermined amount of all components of vanadium, magnesium, rare earth element, manganese and sintering aid. A dielectric powder was prepared. The sintering aid was adjusted to have a composition of SiO ₂ = 55, BaO = 20, CaO = 15, Li ₂ O = 10 (mol%), and the amount of the sintering aid added was BT. It adjusted so that it might become 1 mass part with respect to 100 mass parts of powder. As a comparative example, a sample was prepared by adding glass powder as a sintering aid to a mixed powder of BT powder and BCT powder coated with a predetermined amount of vanadium, magnesium, rare earth element and manganese (Sample No. 1). 36).

  Next, the obtained dielectric powder is put into a mixed solvent of polyvinyl butyral resin, toluene and alcohol, and wet-mixed using a zirconia ball having a diameter of 1 mm to prepare a ceramic slurry. A 2 μm ceramic green sheet was prepared.

  Next, a plurality of conductive pastes containing Ni as a main component were formed on the upper surface of the ceramic green sheet so as to form a rectangular internal electrode pattern. The conductor paste for forming the internal electrode pattern was obtained by adding BT powder to 100 parts by mass of Ni powder having an average particle size of 0.3 μm.

Next, 200 ceramic green sheets on which internal electrode patterns were printed were laminated, and 20 ceramic green sheets on which the internal electrode patterns were not printed were laminated on the upper and lower surfaces, respectively, using a press machine at a temperature of 60 ° C. and pressure A laminated body was prepared by closely adhering under conditions of 10 ⁷ Pa and time 10 minutes, and then the laminated body was cut into a predetermined size to form a capacitor body molded body.

  Next, the molded body of the capacitor body was treated to remove the binder in the air, and then fired at 1100 in hydrogen-nitrogen to produce a capacitor body. In this firing, firing was performed using a roller hearth kiln under the conditions of the heating rate shown in Table 1.

The produced capacitor body was subsequently reoxidized at 1000 ° C. for 4 hours in a nitrogen atmosphere. The size of the capacitor body was 2.05 × 1.28 × 1.28 mm ³ , the thickness of the dielectric layer was 2.0 μm, and the effective area of one internal electrode layer was 1.78 mm ² . The effective area is the area of the overlapping portion of the internal electrode layers that are alternately formed in the stacking direction so as to be exposed at different end faces of the capacitor body.

  Next, after barrel-polishing the capacitor body, an external electrode paste containing Cu powder and glass was applied to both ends of the capacitor body and baked at 850 ° C. to form external electrodes. Thereafter, using an electrolytic barrel machine, Ni plating and Sn plating were sequentially performed on the surface of the external electrode to produce a multilayer ceramic capacitor.

  Next, the following evaluation was performed on these multilayer ceramic capacitors. The dielectric constant at room temperature (25 ° C.) was measured by using an LCR meter (manufactured by Hewlett-Packard Co.) with a capacitance of 25 ° C., a frequency of 1.0 kHz, and an AC voltage of 1.0 V / μm. And the effective area of the internal electrode layer. Moreover, the temperature characteristic of the capacitance was measured in a temperature range of −55 to 85 ° C.

  The DC bias characteristics are C1, electrostatic capacity when a direct current (DC) voltage is not applied under conditions of a temperature of 25 ° C., a frequency of 1.0 kHz, and an AC voltage of 1.0 V / μm, a temperature of 25 ° C., a frequency of 1.0 kHz, It was determined from ((C2−C1) / C1) × 100 (%), where C2 is the capacitance when a direct current (DC) voltage is applied under the condition of AC voltage of 0.2 V / μm.

  DC aging characteristics are as follows: a direct current (DC) voltage of 4.5 V / μm is applied, the electrostatic capacity is measured when 1 hour and 10 hours have elapsed, and the electrostatic capacity after 10 hours is defined as C3. And ((C3-C2) / C1) × 100 (%).

  The average particle size of the crystal particles constituting the dielectric layer is determined by polishing the fracture surface of the sample that is the capacitor body after firing, and then taking a picture of the internal structure using a scanning electron microscope. Draw a circle that contains 30 circles, select the crystal particles that fall within and around the circle, image the outline of each crystal particle, determine the area of each particle, and replace it with a circle with the same area Was calculated from the average value.

The Ca concentration in the crystal particles was measured using a transmission electron microscope apparatus equipped with an energy dispersive analyzer (EDS). In this case, with respect to the crystal particles projected on the cross section of the polished dielectric porcelain, an arbitrary location in the vicinity of the center portion of the crystal particles is analyzed using EDS, and the main component element detected from the crystal particles ( The total amount of each element (Ba, Ti, Ca) and additives (V, Mg, Mn, RE) was 100%, and the content of Ca contained therein was determined. And it classified into the 1st crystal particle whose Ca concentration is lower than 0.3 atomic%, and the 2nd crystal particle whose Ca concentration is 0.3 atomic% or more. The ratio of the first crystal particles having a Ca concentration lower than 0.3 atomic% and the second crystal particles having a Ca concentration of 0.3 atomic% or more is the ratio of the cross section of the crystal particles shown in the transmission electron micrograph. It calculated | required from the area ratio. In this case, in the area of the photograph of the photographed transmission electron microscope, the area of the first crystal particle was S1, the area of the second crystal particle was S2, and the ratio was obtained from the S2 / (S1 + S2) ratio. The number of crystal grains 9 to be evaluated was about 100. Further, as the vicinity of the center portion of the crystal grain, an area inside the depth of 1/3 of the diameter from the grain boundary of the crystal grain found on the polished surface of the sample whose cross section was polished with the dielectric ceramic was analyzed. Regarding the dielectric ceramic after firing, none of the samples contains Ca (peak intensity below the noise level in the analysis) crystal grains and crystal grains having an average Ca concentration of 0.8 atomic%. It was confirmed.

The average shell thickness of the crystal grains was determined using the above-described Equation 1 and Equation 2. At this time, X'PertPro manufactured by Panallytical was used as the X-ray diffractometer. At this time, the crystal structure of the target crystal particle is that the selected X-ray diffraction pattern is reflected in pure tetragonal (h k l) or cubic (h ′ k ′ l ′) in the measured X-ray diffraction pattern. In comparison, the peak was identified, and the one containing tetragonal (h k l) and cubic (h ′ k ′ l ′) reflections was selected.

The diffraction peaks of tetragonal (002) and (200) and cubic (200) were measured. The size of the beam was 0.5 mm in the vertical direction and 5 mm in the horizontal direction. The wavelength was 1.54982 mm. In the measurement of the dielectric ceramic, the step width was 0.02 °, and the counting time per point was 5.0 seconds. Further, the number of repetitions was 10, and the integration for 10 times was defined as the diffraction intensity. In the evaluation, tetragonal (002), (200) and cubic (200) were peak-separated from the diffraction peak using peak separation software under the following conditions. The conditions for peak separation are as follows: background function: 0th order polynomial, synchrotron radiation: Kα1, profile function: the psedo-Voigt function, half width: different half width for all reflections, profile subjectivity: subject, data resolution: Sharp (minimum half width: about 0.1 °) and analysis range: 44 ° <2θ <47 °.

  Moreover, the composition analysis of the sample which is the obtained sintered body was performed by ICP (Inductively Coupled Plasma) analysis and atomic absorption analysis. In this case, the obtained dielectric porcelain mixed with boric acid and sodium carbonate and dissolved in hydrochloric acid is first subjected to qualitative analysis of the elements contained in the dielectric porcelain by atomic absorption spectrometry, and then specified. The diluted standard solution for each element was used as a standard sample and quantified by ICP emission spectroscopic analysis. Further, the amount of oxygen was determined using the valence of each element as the valence shown in the periodic table. The composition of the dielectric layer constituting the obtained multilayer ceramic capacitor matched the composition shown in Table 1.

  The composition and firing conditions are shown in Table 1, and the average particle diameter and dielectric characteristics of the dielectric grains (dielectric constant, capacitance temperature characteristics, DC bias characteristics, DC aging) in the obtained multilayer ceramic capacitor The characteristic results are shown in Table 2, respectively.

As is clear from the results in Tables 1 and 2, the sample No. 2 to 4, 9, 10, 12, 13, 16, 17, 19 to 21, 23 to 30, and 32 to 34, the relative dielectric constant at room temperature (25 ° C.) is 3650 or more, and the capacitance temperature characteristic is X5R In the measurement of DC bias characteristics satisfying (temperature change rate of relative permittivity with respect to 25 ° C. is ± 15% at −55 to 85 ° C.), a DC voltage is measured. When the unapplied capacitance value is used as a reference, the decrease rate of the capacitance value when applied under the condition of 4.5 V / μm is 52.83% (−52.83%: sample No. 25) or less. Furthermore, in the DC aging characteristics in which a DC voltage is applied under the condition of 4.5 V / μm, the rate of decrease in capacitance until the lapse of 0.1 to 10 hours is 18.23% (−18.23%). : Sample No. 32) or less.

  In addition, in the sample (sample No. 12) in which the dielectric ceramic serving as the dielectric layer does not contain vanadium, the relative dielectric constant at room temperature is 3800, and the electrostatic capacity is maintained until 0.1 to 10 hours in the DC aging characteristics. The rate of decrease in capacity was 12.28% (-12.28%: sample No. 13), which was higher in dielectric constant and superior in DC aging characteristics than the sample containing vanadium (sample No. 13). It was.

  In particular, in the samples (samples No. 19 and 27 to 30) in which the dielectric ceramic serving as the dielectric layer does not contain manganese, the capacitance until 0.1 to 10 hours elapses in the DC aging characteristics under the same conditions. The rate of decrease was 10.04% (−9.34%: Sample No. 28) or less.

  On the other hand, sample no. 1, 5-8, 11, 14, 15, 18, 22, 31 and 35-38 have a relative dielectric constant of 3650 or more at room temperature (25 ° C.) and a temperature characteristic of capacitance of X5R (based on 25 ° C.) The temperature change rate of the relative permittivity is within ± 15% at −55 to 85 ° C.), and in the DC bias measurement in which a direct current of 4.5 V / μm is applied, no direct current voltage is applied. In terms of the DC aging characteristics in which the decrease rate of the capacitance value when applied with 4.5 V / μm is within 55% and a DC voltage is applied under the condition of 4.5 V / μm when the value is used as a reference, 0.1 to 10 The capacitance reduction rate until the passage of time did not satisfy any of the characteristics within 20%.

DESCRIPTION OF SYMBOLS 1 Capacitor body 3 External electrode 5 Dielectric layer 7 Internal electrode layer 9 Crystal grain 9A 1st crystal grain 9B 2nd crystal grain 9a Core part 9b Shell part 11 Grain boundary phase

Claims (3)

A laminate having a capacitor body in which a plurality of dielectric layers and a plurality of internal electrode layers made of nickel as a conductive material are alternately laminated, and an external electrode provided on an end surface of the capacitor body where the internal electrode layer is exposed A ceramic capacitor, wherein the dielectric layer has barium titanate as a main component, calcium, magnesium, and at least one rare earth element (RE) selected from yttrium, dysprosium, holmium, terbium and ytterbium. In addition, 0.5 to 1.2 moles of the magnesium in terms of MgO and 0.5 to 1.0 mole of the rare earth elements (RE) in terms of RE ₂ O ₃ with respect to 100 moles of the total amount of barium and calcium In addition, the main crystal particles are mainly composed of the barium titanate, the calcium, and the magnesium. The first crystal particles having a core-shell structure in which the rare earth element (RE) is solid-solved and having a calcium concentration of less than 0.3 atomic% and a calcium concentration of 0.3 atomic% or more The crystal particles have an average particle size of 0.15 to 0.4 μm, the crystal structure of the core part is tetragonal, and the crystal structure of the shell part surrounding the core part is cubic. And the average thickness of the shell portion calculated based on the X-ray diffraction information is 5 to 15 nm, and the area of the first crystal particle seen on the polished surface of the dielectric ceramic is S1, and the second crystal A multilayer ceramic capacitor comprising a dielectric ceramic having S2 / (S1 + S2) of 0.4 to 0.8 when the area of the particles is S2.   The multilayer ceramic capacitor according to claim 1, wherein the dielectric ceramic does not contain vanadium. 2. The multilayer ceramic capacitor according to claim 1, wherein the dielectric ceramic does not contain manganese.

Images