JP6121764B2 - Air electrode used in solid oxide fuel cell - Google Patents

Description

  The present invention relates to an air electrode used in a solid oxide fuel cell.

  Solid oxide fuel cells (hereinafter, sometimes referred to as SOFC) are attracting attention as next-generation high-efficiency fuel cells. The SOFC has an advantage that a power generation efficiency equal to or higher than that of the PEFC can be obtained while keeping the system cost low as compared with a polymer electrolyte fuel cell (hereinafter referred to as PEFC) which is widely used at present. In particular, it is required to improve the performance of the air electrode in order to realize low temperature operation, high efficiency and low cost of the SOFC.

In this regard, 113-based perovskite materials having a composition represented by (La, Sr) (Co, Fe) O ₃ (hereinafter referred to as LSCF), which is the current air electrode material, have a high coefficient of thermal expansion. There is a problem that cracks are likely to occur during firing.

On the other hand, as an air electrode material having a relatively low coefficient of thermal expansion, a 214 series perovskite material represented by a general formula A ₂ BO ₄ (where A is La or the like, B is Ni or the like) is As a known SOFC air electrode, the relationship between the baking temperature on the cell and the reaction resistance has been reported.

For example, Non-Patent Document 1 (Lihua Lu et al., “Electrochemical performance of La ₂ NiO _{4 + δ} -La _0.6 Sr _0.4 Co _0.2 Fe _0.8 O _3-δ composite cathodes for intermediate temperature solid oxide fuel cells”, Materials Research Bulletin 45 (2010) 1135-1140) include La ₂ NiO _{4 + δ} (referred to as LN in the document) and (La, Sr) (Co, Fe) O _3-δ (referred to as LSCF in the document). The composite air electrode material was manufactured at a baking temperature of 1050 ° C. to 1150 ° C., and the reaction resistance was measured. In this document, an LN electrode composed only of a 214-based material has a high reaction resistance of 0.63 Ωcm ² when baked at 1050 ° C., whereas it is 0.21 Ωcm ² and 0 when baked at 1100 ° C. and 1150 ° C. It has been reported that reaction resistance as low as .22Ωcm ² was obtained.

Non-Patent Document 2 (Yue Cao et al., “Preparation and characterization of Nd _2-χ Sr _χ CoO _{4 + δ} cathodes for intermediate-temperature solid oxide fuel cell”, International Journal of Hydrogen Energy 35 (2010) 5594-5600) Discloses that the reaction resistance was measured by changing the baking temperature in the range of 1000 to 1200 ° C. in the Nd _2−x Sr _x CoO _{4 + δ} air electrode material. This document reports that when the baking temperature is 1000 ° C., the sinterability of the particles is poor, so the reaction resistance is high, and when the baking temperature is 1200 ° C., the particle size increases and the performance is lower than the baking at 1100 ° C. Has been. That is, the results show that the baking resistance of 1100 ° C. has the lowest reaction resistance.

Non-Patent Document 3 (C. Lalannea et al., “Neodymium-deficient nickelate oxide Nd _1.95 NiO _{4 + δ} as cathode material for anode-supported intermediate temperature solid oxide fuel cells”, Journal of Power Sources 185 (2008) 1218-1224 ) Discloses that the reaction resistance was measured by changing the baking temperature in the range of 1000 to 1200 ° C. in the Nd _1.95 NiO _{4 + δ} air electrode material. This document reports that the reaction resistance at the baking temperature of 1100 ° C. was the lowest at 0.183 Ωcm ² .

  Thus, the present situation is that 214-type perovskite materials have poor sintering properties and high reaction resistance even at a baking temperature equivalent to that of LSCF, and have not fully exhibited their original performance.

LihuaLu et al., "Electrochemical performance of La2NiO4 + δ-La0.6Sr0.4Co0.2Fe0.8O3-δ composite cathodes for intermediate temperature solid oxide fuel cells", Materials Research Bulletin 45 (2010) 1135-1140 YueCao et al., `` Preparation and characterization of Nd2-χSrχCoO4 + δ cathodes for intermediate-temperature solid oxide fuel cell '', International Journal of Hydrogen Energy 35 (2010) 5594-5600 C. Lalannea et al., `` Neodymium-deficient nickelate oxide Nd1.95NiO4 + δ as cathode material for anode-supported intermediate temperature solid oxide fuel cells '', Journal of Power Sources 185 (2008) 1218-1224

  In the air electrode containing a 214-type perovskite material having a specific composition having a relatively low coefficient of thermal expansion, the present inventors have recently made a certain microstructure that contributes greatly to reducing the reaction resistance in a solid oxide fuel cell. I got the knowledge that.

  Accordingly, an object of the present invention is to provide an air electrode that can exhibit a reduced reaction resistance in a solid oxide fuel cell while having a relatively low coefficient of thermal expansion.

According to one aspect of the present invention, there is provided an air electrode used by bonding a solid electrolyte layer and / or a reaction preventing layer in a solid oxide fuel cell,
The air electrode has a general formula: A ₂ BO _{4 ± δ} (wherein A is at least one rare earth element selected from La, Pr and Nd, and B is from Ni, Cu, and Fe, Ga and Al). A layered perovskite oxide having a composition represented by: at least one element selected, wherein δ represents an oxygen excess or oxygen deficiency but may be 0),
Provided is an air electrode having a microstructure in which an average length of a conductive path at a bonding interface formed when bonded to a solid electrolyte layer and / or a reaction preventing layer is 0.5 μm or more.

  According to another aspect of the present invention, an air electrode for a solid oxide fuel cell comprising the air electrode of the present invention and a solid electrolyte layer and / or a reaction preventing layer provided on one surface side of the air electrode. A complex is provided.

It is a figure for demonstrating the length of the electroconductive path in the joining interface of an air electrode and a reaction prevention layer. It is a figure for demonstrating the relationship between the conductive path in the joining interface of an air electrode and a reaction prevention layer, and the length of one visual field of a microstructure image. It is an example of the cross-sectional image binarized in order to calculate the porosity of an air electrode. Black portions in the figure correspond to pores. It is a figure for demonstrating specification of the particle diameter at the time of calculating | requiring the average particle diameter of an air electrode by the intercept method. It is a figure for demonstrating the neck diameter of the particle | grains which comprise an air electrode. 5 is a schematic view of a coin cell sample for reaction resistance measurement manufactured in Example 3. FIG. 6 is a schematic diagram of a reaction resistance measurement system used in Example 3. FIG. It is a graph which shows the relationship between the baking temperature to the coin cell of the air electrode in Example 3, and reaction resistance. 10 is a graph showing the relationship between the average length of conductive paths and the reaction resistance at the air electrode bonding interface in Example 4. 6 is a graph showing the relationship between the average ratio of the length of the conductive path at the bonding interface of the air electrode in Example 4 and the reaction resistance. 10 is a graph showing the relationship between the average porosity of the air electrode and the reaction resistance in Example 4. 10 is a graph showing the relationship between the average particle diameter of the air electrode and reaction resistance in Example 4. 10 is a graph showing the relationship between the average neck diameter of air electrode and reaction resistance in Example 4.

Air electrode present invention relates to an air electrode to be used by bonding the solid electrolyte layer and / or the reaction-preventing layer in the solid oxide fuel cell (SOFC). The air electrode has a general formula: A ₂ BO _{4 ± δ} (where A is at least one rare earth element selected from La, Pr, and Nd, and B is selected from Ni, Cu, and Fe, Ga, and Al) A layered perovskite oxide having a composition represented by: δ represents an oxygen excess or oxygen deficiency, but may be 0). The air electrode has a microstructure in which the average length of the conductive path at the bonding interface formed when bonded to the solid electrolyte layer and / or the reaction preventing layer is 0.5 μm or more. According to the present knowledge of the present inventors, such a unique microstructure can contribute to a reduction in reaction resistance in a solid oxide fuel cell.

  In particular, the layered perovskite having the composition of the above general formula is a 214-based perovskite material whose thermal expansion coefficient is relatively lower than that of LSCF (114-based perovskite material), which is the current air electrode material, and cracks are generated when the SOFC cell is co-fired. It has the advantage that it does not easily occur. On the other hand, the present condition is that this material has a poor sintering property and a high reaction resistance even at a baking temperature equivalent to that of LSCF, and the original performance is not fully exhibited. In this regard, according to the present invention, since a structure that can contribute to the reduction of the reaction resistance has been specified, it is possible to realize both the reaction resistance and the thermal expansion coefficient, which are problems in the air electrode for the solid oxide fuel cell. Thus, the performance of the solid oxide fuel cell can be improved. In addition, according to the calculations by the present inventors, when the reaction resistance is reduced by 30%, the cell resistance of the solid oxide fuel cell is reduced by 10% (this means that the cell output is increased by 10%), Accordingly, the material cost of the cell is reduced by 10%, so that the cost of the solid oxide fuel cell can be reduced.

The air electrode includes a layered perovskite oxide, and the layered perovskite oxide is represented by a general formula: A ₂ BO _{4 ± δ} . In the above general formula, A is at least one rare earth element selected from La, Pr and Nd, preferably Nd. B is Ni, Cu, and at least one element selected from Fe, Ga, and Al, and preferably a combination of Ni, Cu, and Ga. δ represents an oxygen excess amount or an oxygen deficiency amount, but may be 0. The basic composition represented by the above general formula includes both a stoichiometric composition and a non-stoichiometric composition. However, even if the latest apparatus is used, oxygen excess amount or oxygen deficiency (δ) is analyzed and In light of the fact that cannot be quantified, it may be abbreviated as A ₂ BO ₄ conventionally. In any case, it is considered that there is no problem if 0 ≦ δ <0.4. The air electrode of the present invention is characterized by a unique microstructure. As a premise that the microstructure contributes to a reduction in reaction resistance, the layered perovskite oxide having a composition of A ₂ BO _{4 ± δ} is as described above. It is necessary to be composed of specific elements carefully selected.

  In the air electrode of the present invention, the average length of the conductive path at the bonding interface formed when bonded to the solid electrolyte layer and / or the reaction preventing layer is 0.5 μm or more, preferably 0.5 to 2.0 μm. More preferably, it has a microstructure that is 0.7 to 1.5 μm. As shown in FIG. 1, the “average length of the conductive path at the bonding interface” means the direction parallel to the interface at the location where the air electrode particles are connected to the solid electrolyte layer and / or the reaction preventing layer. Refers to the average length of This length can be measured with a personal computer monitor for a scanning electron microscope. In this case, a length measurement function of SEM software, image analysis software (for example, Adobe Photoshop), or the like can be preferably used. In order to improve the objectivity of the numerical value, it is preferable to measure the average length of the conductive path in one visual field, perform this for a total of five visual fields, and adopt the average value for the five visual fields as the average length of the conductive path. However, the air electrode of the present invention may have any microstructure that can realize the average length of the conductive path within the above range when bonded to the solid electrolyte layer and / or the reaction preventing layer. Needless to say, it may be in a single form without the layer and / or the reaction preventing layer.

  In addition, when the bonding interface is observed with a microscope, the average ratio of the length of the conductive path in the length of one visual field of the obtained microstructure image is preferably 30% or more, more preferably 30 to 60%. More preferably, it is 35 to 57%. Within such a range, the reaction resistance is reduced. “The average ratio of the length of the conductive path to the length of one visual field of the microstructure image” is the total value of the length of the conductive path obtained by the above-described method for one visual field as shown in FIG. Is obtained by dividing the total length of the reaction preventing layer surface in the visual field (that is, the horizontal length of one visual field). This length can be measured with a personal computer monitor for a scanning electron microscope. In this case, a length measurement function of SEM software, image analysis software (for example, Adobe Photoshop), or the like can be preferably used. In order to increase the objectivity of the numerical value, the average ratio of the length of the conductive path is obtained for one field of view, and this is performed for a total of five fields of view, and the average value for the five fields of view is adopted as the average ratio of the length of the conductive path. Is preferred.

The air electrode preferably has an average porosity of 35% or less, more preferably 20 to 35%, and still more preferably 25 to 32%. Within such a range, the reaction resistance is reduced. The porosity can be determined, for example, by the following procedure:
-Obtaining an electron microscope (SEM) image of a cross section in the thickness direction of the air electrode;
-Identifying pores in this image (field of view),
-Obtaining the area of the cathode in this image,
-Obtaining the total area of the pores in this image; and-calculating [total pore area / total area of the air electrode] in this image,
Can be obtained. SEM, FE-SEM, etc. can be used for acquisition of a cross-sectional image, and image analysis software (for example, Adobe Photoshop) etc. can be used for quantification of the area of the pores thereafter. In order to improve the objectivity of the numerical values, the cross-sectional image is binarized into black and white as shown in FIG. 3 using image editing software, and the ratio of the black portion of the obtained binary image is read. It is preferable to adopt the average value of the values for the 10 visual fields that have been read and read as the average porosity.

  The particles constituting the air electrode preferably have an average particle diameter of 0.8 μm or more, more preferably 0.8 to 2.0 μm, more preferably 0.9 to 0, when measured by the intercept method. 1.6 μm. Within such a range, the reaction resistance is reduced. As shown in FIG. 4, the intercept method is a method in which the length of particles passing through a straight line drawn on a microstructure image is measured and an average value is obtained. The straight line is drawn, the length between the boundaries of the particles on the straight line is obtained, and the particle diameter is calculated based on the scale value of the microstructure image. For this operation, image analysis software (for example, Adobe Photoshop) or the like can be preferably used. In order to increase the objectivity of the numerical values, all the particle sizes obtained in one field of view are arranged in order of size, and the particle size data of a total of 50 points of the upper 25 points and the lower 25 points are collected in light of their average values, It is preferable to collect this data in a total of three fields and adopt the average value of the particle diameters for the three fields as the average particle diameter.

  According to a preferred embodiment of the present invention, the particles constituting the air electrode have an average neck diameter of 0.35 μm or more, preferably 0.35 to 1.0 μm, more preferably 0.5 to 0.9 μm. It is. Within such a range, the reaction resistance is reduced. “Neck” refers to the contact surface between the particles, and “neck diameter” refers to the shortest distance of the contact surface between the particles, as clearly shown in FIG. The neck diameter can be measured with a personal computer monitor for a scanning electron microscope. In this case, a length measurement function of SEM software, image analysis software (for example, Adobe Photoshop), or the like can be preferably used. “Average neck diameter” is defined as the average value of 30 neck diameters per field of view thus measured. In order to increase the objectivity of the numerical values, all neck diameters obtained in one field of view are arranged in order of size, and the neck diameter data of a total of 30 points of the top 15 points and the bottom 15 points is collected in light of their average value, It is preferable that this data collection is performed for a total of three fields of view, and an average value of neck diameters for the three fields of view is adopted as the average neck diameter.

  The air electrode preferably has a thermal expansion coefficient of 12.5 to 13.5 ppm / K, more preferably 12.5 to 13.4 ppm / K. Such a thermal expansion coefficient is lower than the thermal expansion coefficient of LSCF, which is the current air electrode material, and leads to a reduction in the probability of occurrence of cracks during cofiring of SOFC cells.

The air electrode can be manufactured as a composite structure further provided with a solid electrolyte layer and / or a reaction preventing layer on one surface side thereof, preferably an air electrode-reaction preventing layer composite or an air electrode-solid electrolyte composite. Can be manufactured. When the reaction preventing layer is configured as a fuel cell, it is interposed between the air electrode and the solid electrolyte, and a high resistance layer is formed by an interfacial reaction between the air electrode and the solid electrolyte. It is a layer for preventing, and the battery resistance can be improved by reducing the internal resistance of the cell. However, when a high resistance layer is not formed between the air electrode and the solid electrolyte by an interface reaction, the air electrode and the solid electrolyte layer may be directly joined without providing a reaction preventing layer. A preferable reaction preventing layer is made of a material containing cerium, more preferably ceria and a ceria-based material containing a rare earth metal oxide solid-dissolved in ceria. The concentration of the rare earth metal in the ceria-based material is preferably 5 to 20 mol%. Examples of such ceria-based materials include GDC ((Ce, Gd) O ₂ : Gadolinium-doped ceria), SDC ((Ce, Sm) O ₂ : samarium-doped ceria), and the like. Particularly preferably, the reaction preventing layer is composed of gadolinium-doped ceria (GDC), and when the composition is expressed as Gd _α Ce _1-α O ₂ , the preferable doping amount α of Gd is 0.05 to 0. .20. The reaction preventing layer may contain an additive in addition to the ceria-based material. The shape and size of the reaction preventing layer may be appropriately determined according to the design of the applied fuel cell, but when applied to a typical SOFC, it is formed in a layered or plate shape with a thickness of 30 μm or less. It is preferable. On the other hand, the material of the solid electrolyte layer can be various solid electrolytes generally used for fuel cells, but is preferably a solid electrolyte containing zirconium, more preferably zirconia (ZrO ₂ ) as a main component. It is a material that contains. The solid electrolyte may contain additives such as Y ₂ O ₃ and / or Sc ₂ O ₃ in addition to zirconia, and these additives can function as a stabilizer. The addition amount of the additive in the solid electrolyte layer is about 3 to 20 mol%. Particularly preferred solid electrolytes include yttria stabilized zirconia such as 3YSZ, 8YSZ and 10YSZ, and zirconia based materials such as ScSZ (scandia stabilized zirconia). The shape and size of the reaction preventing layer may be appropriately determined according to the design of the applied fuel cell, but when applied to a typical SOFC, it is formed in a layered or plate shape with a thickness of 30 μm or less. It is preferable.

Production Method The air electrode of the present invention comprises a raw material powder mixture containing each constituent element at a blending ratio giving a composition represented by the general formula: A ₂ BO _{4 ± δ} , and a predetermined base material (typically It can be produced by applying and baking on a solid electrolyte layer and / or a reaction preventing layer. The production method is not particularly limited as long as the desired microstructure can be finally realized, but the slurry containing the raw material powder is dried, calcined, pulverized and dried to obtain a precursor powder. It is preferable to apply a paste containing powder onto a substrate and bake it under a predetermined temperature condition (preferably 1050 to 1150 ° C.). The microstructure of the present invention can be controlled by appropriately adjusting the temperature conditions during baking as shown in the examples described later.

  The present invention is more specifically described by the following examples.

Example 1 : Preparation of powder sample As 214 type layered perovskite oxide material for constituting an air electrode, powder samples having the following three compositions similar to Example 1 were prepared.
_{- Nd 2 Ni 0.75 Cu 0.2 Ga} 0.05 O 4 ( hereinafter, abbreviated as NNCG)
_{- Nd 2 Ni 0.75 Cu 0.2 Al} 0.05 O 4 ( hereinafter, abbreviated as NNCA)
-Nd ₂ Ni _0.75 Cu _0.2 Fe _0.05 O ₄ (hereinafter abbreviated as NNCF)

Specifically, each raw material powder of Nd ₂ O ₃ , NiO, CuO, Ga ₂ O ₃ , Al ₂ O ₃ and Fe ₂ O ₃ is put into a 500 ml pot containing about 1 kg of zirconia cobblestone (diameter 3 mm), A total of 50 g was weighed and put in such a manner that each composition ratio of NNCG, NNCA and NNCF was obtained. In this pot, isopropyl alcohol (IPA) as a solvent is put in an amount (specifically, about 100 ml) to which the cobblestone is immersed, the lid is closed, and the slurry is mixed for 4 hours at a rotation speed of 100 rpm using a pot mill carrier. Got. The slurry was dried with an N ₂ dryer in an N ₂ atmosphere at 110 ° C. for 16 hours, and sieved with 100 mesh to obtain a powder. This powder is put in a crucible, the crucible is put in a sheath to form a double sheath, and is heated in an electric furnace in an air atmosphere at 1300 ° C. for 20 hours at a heating rate of 200 ° C./h and a cooling rate of 300 ° C./h. Calcination was performed under conditions. The obtained calcined powder was pulverized in a mortar. In a 500 ml pot containing about 1 kg of zirconia cobblestone (diameter 5 mm), the above pulverized powder and isopropyl alcohol (IPA) as a solvent are placed so that the cobblestone is immersed (about 100 ml), and the pot mill rotates at a rotational speed of 120 rpm. By grinding for a time, a slurry was obtained. The slurry was dried with an N ₂ dryer in an N ₂ atmosphere at 110 ° C. for 16 hours, and sieved with 100 mesh to obtain NNCG, NNCA, and NNCF powder samples.

Example 2 : Measurement of coefficient of thermal expansion Using each powder sample of NNCG, NNCA, and NNCF obtained in Example 1, a square bar of each cathode material having a sample length of 20 mm was prepared. Using this square bar and alumina as a reference, the coefficient of thermal expansion was measured by raising and lowering the temperature in the atmosphere in the temperature range of 40 to 1000 ° C. For comparison, the thermal expansion coefficient of (La _0.6 Sr _0.4 ) (Co _0.2 Fe _0.8 ) O ₃ (hereinafter abbreviated as LSCF) was also measured. Table 1 shows the coefficient of thermal expansion of each material at 1000 ° C. when the temperature is lowered.

  From the results shown in Table 1, the thermal expansion coefficient of the 214 series material is 12.9 to 13.1 ppm / K, which is lower than that of 15.1 ppm / K of the current air electrode material LSCF. It can be seen that it has a coefficient. Lowering the thermal expansion coefficient leads to a reduction in the probability of occurrence of cracks during cofiring of SOFC cells.

Example 3 : Measurement of reaction resistance An air electrode material was prepared using the powder sample obtained in Example 1, and the change in reaction resistance in relation to the baking temperature on the coin cell was examined. Specifically, it is as follows.

(1) Preparation of air electrode Using the powder samples of NNCG, NNCA and NNCF obtained in Example 1, a paste for air electrode printing was prepared as follows. That is, 0.7 g of PVA resin and 1.6 g of BCA (Butyl carbitol acetate) as a solvent were added to the evaporating dish. The evaporating dish was covered with aluminum foil so that BCA would not evaporate, and was heated in an air atmosphere at 120 ° C. for 30 minutes in a dryer in order to dissolve the resin in BCA. After allowing to cool, 10 g of a powder sample was placed in an evaporating dish and mixed with a pending knife. Then, it mixed until the viscosity of the paste became constant using tri-roll, and the paste sample for air electrodes was obtained.

(2) Production of coin cell A coin cell having a diameter of 12 mm and a thickness of 1 mm was produced as follows. That is, as shown in FIG. 6, in order to form a coin cell 10, an anode 12 (nickel oxide-yttria stabilized zirconia (NiO-YSZ)), an active layer 14 (nickel oxide-yttria stabilized zirconia (NiO-YSZ)). )), An air electrode paste sample is 6 mm in diameter on the reaction preventing layer 18 of the laminate 20 composed of the electrolyte 16 (yttria stabilized zirconia (YSZ)) and the reaction preventing layer 18 (gadolinium-doped ceria (GDC)). The air electrode 22 was formed by printing in a circle and baking in an air atmosphere for 1 hour. This baking was performed by appropriately selecting any one of the temperatures of 1000 ° C., 1050 ° C., 1100 ° C. and 1150 ° C. so that the microstructure of the air electrode 22 can be controlled. The baking temperature is 1000 ° C.). NiO paste was applied to the surface on the fuel electrode 12 side, and Ag paste was applied to the surface on the air electrode 22 side to form electrode terminals 24 and 26, respectively.

(3) Production of reaction resistance measurement system A reaction resistance measurement diameter 30 as shown in FIG. 7 was produced using the coin cell sample 10 thus obtained. Specifically, the coin cell sample 10 was set in an alumina holder (not shown), and the alumina tube 34 of the measuring apparatus was sandwiched and fixed. Further, each gap was sealed with glass (PYREX) so that hydrogen gas did not leak from the alumina holder 32. Further, quartz tubes 36 for supplying air or hydrogen are arranged above and below the coin cell sample 10, and at that time, in order to improve current collection efficiency, a Pt mesh 38 is provided between the quartz cells above and below the coin cell sample. I caught it. Thus, a reaction resistance measuring system 30 in which a coin cell sample was incorporated was obtained.

(4) Measurement of reaction resistance In the obtained measurement system 30, the reaction resistance of the entire cell was measured while supplying synthetic air from the air electrode side and hydrogen from the fuel electrode side. This measurement was performed using a potentio / galvanostat (Sorlartron 1287 type) or FRA (Sorlatron 1260 type) under the following conditions by current-voltage measurement or AC impedance measurement.
・ Measurement temperature: 750 ℃
-Current value range during current-voltage measurement: 0-0.53 A / cm ²
AC impedance measurement conditions: OCV (no load), 0.3 A / cm ²
・ Frequency range of AC impedance measurement: 10 ^{6 to} 0.1 Hz

The reaction resistance was calculated from the obtained measured value based on the following formula.
Reaction resistance (Ω / cm ² ) = ((measured voltage−OCV) −R _ohm I) / J
That is, the voltage value actually applied to the cell was calculated by subtracting the open circuit voltage (OCV) from the voltage obtained by the current-voltage measurement. From this voltage value, the voltage drop (overvoltage) of the cell due to the reaction resistance was calculated by removing the ohmic resistance (R _ohm ) of the sample obtained from AC impedance measurement (at the time of OCV). The reaction resistance was determined by dividing this voltage drop by the current density (J).

The value of the reaction resistance at the point of 0.3 A / cm ² , which is the operating current of the SOFC device, is shown in FIG. 8 in relation to the baking temperature. From the results shown in FIG. 8, it can be seen that the reaction resistance is reduced by increasing the baking temperature.

Example 4 : Observation and evaluation of microstructure The microstructure of the air electrode in the coin cell produced in Example 3 was observed and evaluated. Specifically, it is as follows.

(1) Observation of microstructure After the coin cell produced in Example 3 was resin-embedded, the sample cross section was polished by an ion milling method. Ion milling was performed under the following conditions using an ion milling device (IM4000, manufactured by Hitachi High-Tech Co., Ltd.).
・ Acceleration voltage: 5 kV
・ Processing time: 1.5 hours ・ Atmosphere: Ar
However, when calculating the particle diameter, which will be described later, since the grain boundary is difficult to distinguish on the polished surface, the fracture surface was used instead of the polished surface.

The microstructure of the polished surface was observed using a scanning electron microscope (SEM) (JSM-6610LV Scanning Electron Microscope, manufactured by JEOL Ltd.) under the following conditions.
・ Acceleration voltage: 5 kV (other than particle diameter observation) or 20 kV (particle diameter observation)
・ Working distance: 9-10mm
・ Spot size: 30

(2) Evaluation of microstructure Based on the microstructure image of the air electrode observed by SEM, the average length of the conductive path at the bonding interface, the average ratio of the length of the conductive path at the bonding interface, the average porosity, and the average particle Evaluation was made from five viewpoints of diameter and average neck diameter, and the relationship between the evaluation results and reaction resistance was examined.

<Average length of conductive path at bonding interface>
Based on the microstructure image of the bonding interface between the air electrode and the reaction preventing layer, the length of the air electrode particles connected to the reaction preventing layer is measured in the direction parallel to the interface to conduct electricity at the bonding interface. The average length of the path was determined. The microstructure image used was about 8000 to 10,000 times, and image analysis software (Adobe Photoshop) was used for length measurement. Specifically, the average length of the conductive path was measured in one field of view, and this was performed for a total of five fields of view, and the average value for the five fields of view was taken as the average length of the conductive path. FIG. 9 shows the relationship between the average length of the conductive path at the bonding interface and the reaction resistance. The results shown in FIG. 9 indicate that the reaction resistance is reduced when the average length of the conductive path at the bonding interface is 0.5 μm or more.

<Average ratio of length of conductive path at bonding interface>
Based on the microstructure image of the bonding interface between the air electrode and the reaction preventing layer, the total value for one field of the length of the conductive path obtained by the above method is calculated as the total length of the surface of the reaction preventing layer in that field (ie The average ratio of the length of the conductive path at the bonding interface was determined by dividing by the (length in the lateral direction). The microstructure image used was about 8000 to 10,000 times, and image analysis software (Adobe Photoshop) was used for length measurement. Specifically, the average ratio of the length of the conductive path in one field of view was obtained, and this was performed for a total of five fields of view, and the average value for the five fields of view was taken as the average ratio of the length of the conductive path. FIG. 10 shows the relationship between the average ratio of the length of the conductive path at the bonding interface and the reaction resistance. The results shown in FIG. 10 indicate that the reaction resistance is reduced when the microstructure has a conductive path length average ratio of 30% or more at the bonding interface.

<Average porosity>
The porosity of the air electrode was obtained by image processing of the microstructure image of the air electrode. Specifically, first, a coin cell sample in which a cross section in the thickness direction of the air electrode was obtained by embedding the air electrode sample in a resin and performing ion polishing treatment was prepared. Next, an electron microscope (SEM) image of a cross section in the thickness direction of the air electrode was acquired at a magnification of 3000 times (n = 10 fields of view). Read grayscale SEM image with image analysis software (Adobe Photoshop), and adjust the threshold of histogram appropriately by the procedure of [Image] → [Tonal Correction] → [Turn Tone] to create a monochrome binary image did. The ratio occupied by the black portion of the binary image was read, and this reading was performed for 10 fields of view. The average value of the values read for 10 fields of view was defined as the average porosity. FIG. 11 shows the relationship between the average porosity and the reaction resistance. The results shown in FIG. 11 indicate that the reaction resistance is reduced when the microstructure has an average porosity of 35% or less.

<Average particle size>
Based on the microstructure image of the air electrode, the average particle diameter of the air electrode was determined by the intercept method. That is, six straight lines were arbitrarily drawn on the microstructure image, the length between the boundaries of the particles on the straight line was obtained, and the particle diameter was calculated based on the scale value of the microstructure image. The magnification of the microstructure image used was about 8000 to 10,000 times, and image analysis software (Adobe Photoshop) was used for length measurement. Specifically, all the particle sizes obtained in one field of view are arranged in order of size, and the particle size data of a total of 50 points of the upper 25 points and the lower 25 points are collected in light of the average value thereof. The measurement was performed for a total of three fields of view, and the average value of the particle diameters for the three fields of view was calculated as the average particle size. FIG. 12 shows the relationship between the average particle diameter and the reaction resistance. The results shown in FIG. 12 indicate that the reaction resistance decreases when the microstructure has an average particle diameter of 0.8 μm or more.

<Average neck diameter>
Based on the microstructure image of the air electrode, the shortest distance of the contact surface (neck) between the particles was measured to obtain the average neck diameter. The magnification of the microstructure image used was about 8000 to 10,000 times, and image analysis software (Adobe Photoshop) was used for length measurement. Specifically, all the neck diameters obtained in one field of view are arranged in order of size, and the neck diameter data of a total of 30 points of the top 15 points and the bottom 15 points is collected in light of the average value thereof. The measurement was performed for a total of three visual fields, and the average value of the neck diameters for the three visual fields was calculated as the average neck diameter. FIG. 13 shows the relationship between the average neck diameter and the reaction resistance. The results shown in FIG. 13 indicate that the reaction resistance decreases when the microstructure has an average neck diameter of 0.35 μm or more.

<Relationship between microstructure and reaction resistance>
Table 2 shows the relationship between the five elements of the microstructure and the reaction resistance.

<Multivariate analysis (multiple regression analysis)>
Multivariate analysis (multiple regression analysis) was performed to examine which elements of the air electrode microstructure contributed greatly to the reaction resistance. Multivariate analysis is performed with spreadsheet software (Excel, manufactured by Microsoft) and explains the five elements in the microstructure (average length of conductive path, average ratio of conductive path, average porosity, average particle diameter, and average neck diameter). Multiple regression analysis was performed using variables (variables used for prediction) and reaction resistance values as objective variables (variables to be predicted). Here, the value of NNCF (1000 ° C.) was considered a singular point and was not included in the calculation. In addition, the multiple regression analysis is the average of the conductive path excluding the combination of the average porosity and average particle diameter, the average ratio of the conductive path and the average neck diameter, and the combination of the average neck diameter and the average particle diameter among the five elements. A multiple regression analysis was performed by limiting to three variables: ratio, average porosity, and average length of the conductive path (interface connection). As a result, among the above five elements, the ratio of the conductive path at the interface and the length of the conductive path have a high contribution rate to the reaction resistance, and the influence of the conductive path between the reaction preventing layer and the air electrode is the largest. found. This is considered to be because oxygen permeation and carrier movement are performed more efficiently by increasing the conductive path.

In addition, although the said Example is related to the structure by which the reaction prevention layer was joined to the air electrode, the said knowledge is similarly applied also to the fuel cell of the structure directly joined to a solid electrolyte layer not via a reaction prevention layer. . This is because the reaction preventing layer and the solid electrolyte layer are common in that they have performance as a solid electrolyte.

Claims (5)

An air electrode composite for a solid oxide fuel cell, comprising: an air electrode; and a solid electrolyte layer and / or a reaction prevention layer provided to be joined to one side of the air electrode ,
The air electrode has a general formula: A ₂ BO _{4 ± δ} (wherein A is at least one rare earth element selected from La, Pr and Nd, and B is from Ni, Cu, and Fe, Ga and Al). A layered perovskite oxide having a composition represented by: at least one element selected, wherein δ represents an oxygen excess or oxygen deficiency but may be 0),
The average length of the conductive paths at a joint interface of the air electrode and the solid electrolyte layer and / or the reaction preventing layer have a microstructure which is a 0.5 to 2.0 [mu] m,
Particles constituting the air electrode has an average neck diameter of 0.35～1.0Myuemu, cathode composite. The junction interface when microscopic observation, the length average percentage of the conductive paths occupying the length of one field of view of the microstructure image obtained is 30 to 60% and the air electrode composite according to claim 1 . The air electrode composite according to claim 1 or 2, wherein the air electrode has an average porosity of 20 to 35%. The particles constituting the air electrode, as measured by the intercept method, having an average particle diameter of 0.8 to 2.0 [mu] m, air electrode composite according to any one of claims 1-3. The air electrode composite according to any one of claims 1 to 4 , wherein the air electrode has a thermal expansion coefficient of 12.5 to 13.5 ppm / K.