JP5632518B2 - Fuel cell - Google Patents

Description

  The technology disclosed herein relates to a solid oxide fuel cell including a support substrate.

Conventionally, a solid oxide fuel cell including a support substrate containing MgO, MgAl ₂ O _4, and Ni and a plurality of power generation units formed on the support substrate is known (see Patent Document 1).

JP-A-2005-340164

  However, the support substrate of Patent Document 1 has a problem that cracks are likely to occur because the dimensional change of the support substrate before and after reduction is large when exposed to a reducing atmosphere.

  The technique disclosed here is made in view of such a situation, and it aims at providing the fuel cell which can suppress that a crack generate | occur | produces in a support substrate by a reduction process.

  The fuel battery cell disclosed here includes a support substrate and a power generation unit. The support substrate has a fuel gas channel inside, and contains an insulating ceramic containing MgO as a main component. The power generation unit is formed on the support substrate. The power generation unit includes a fuel electrode, an air electrode, and a solid electrolyte layer disposed between the fuel electrode and the air electrode. When the support substrate is reduced, the average joining length of two adjacent insulating ceramic particles is 0.5 μm or more and 2.2 μm or less. The support substrate does not contain Ni.

  According to the technology disclosed herein, it is possible to provide a fuel battery cell capable of suppressing the generation of cracks in the support substrate due to the reduction treatment.

The perspective view which shows the structure of a horizontal stripe type fuel cell II sectional view of FIG.

  A horizontal stripe fuel cell 100 will be described as an example of a solid oxide fuel cell (SOFC).

[Configuration of Horizontally Striped Fuel Cell 100]
FIG. 1 is a perspective view showing an outline of a horizontally-striped fuel cell (hereinafter simply referred to as “cell”) 100. 2 is a cross-sectional view taken along the line II of FIG.

  As shown in FIGS. 1 and 2, the cell 100 includes a support substrate 10, a plurality of power generation units 20, a plurality of interconnectors 30, and a current collection unit 40. In FIG. 1, the current collector 40 is omitted for convenience of explanation.

1. Support substrate 10
The support substrate 10 is a plate-like member that is flat and extends in one direction (hereinafter referred to as “longitudinal direction”). The support substrate 10 is composed of a porous body having insulating properties. A flow path 10 a extending along the longitudinal direction of the support substrate 10 is formed inside the support substrate 10. At the time of power generation, a fuel gas containing hydrogen or the like is caused to flow through the flow path 10a, whereby the fuel gas is supplied to the plurality of power generation units 20 via the support substrate 10. At this time, the support substrate 10 is exposed to a reducing atmosphere.

The support substrate 10 contains an insulating ceramic containing MgO as a main component. The insulating ceramic may contain only MgO. Further, the insulating ceramic may contain at least one of MgAl ₂ O ₄ , CaZrO ₃ and CaTiO ₃ in addition to MgO. In the present embodiment, the composition X “comprising the substance Y as the main component” means that the substance Y preferably occupies 60% by weight or more, more preferably 70% by weight or more in the entire composition X. More preferably, it means 90% by weight or more.

The support substrate 10 does not contain Ni and NiO. However, “the support substrate 10 does not contain Ni” means not only when the Ni content is 0 mol%, but also when the Ni content is greater than 0 mol% and 1 mol% or less. It is a concept that includes. Further, since Ni or NiO may be diffused from the power generation unit 20 in the region of the support substrate 10 that contacts the power generation unit 20, the Ni content in the support substrate 10 is predetermined from the interface with the power generation unit 20. It is preferable to measure at a position separated by a distance (for example, 100 μm). The support substrate 10 may contain at least one of Fe ₂ O ₃ , SiO ₂ , B ₂ O ₃ , and Al ₂ O ₃ .

Further, the support substrate 10 may contain Fe. The support substrate 10 may contain Fe as an oxide (Fe ₂ O ₃ ). Fe ₂ O ₃ may be dissolved in MgO. If the supporting substrate 10 contains _{Fe 2 O 3, Fe 2 O} 3 may be reduced to Fe by exposure to a reducing atmosphere at the time of power generation. Further, the support substrate 10 may contain at least one of SiO ₂ , B ₂ O ₃ , and Al ₂ O ₃ .

  Here, in the support substrate 10, the Si content is 200 ppm or less, the P content is 50 ppm or less, the Cr content is 100 ppm or less, the B content is 100 ppm or less, and The S content is preferably 100 ppm or less. As a result, the resistance of the support substrate 10 to the re-oxidation / re-reduction test (Redox test) can be improved. In particular, the Si content is 100 ppm or less, the P content is 30 ppm or less, the Cr content is 50 ppm or less, the B content is 50 ppm or less, and the S content is 30 ppm. The following is more preferable.

  The microstructure of the support substrate 10 in the reduced state will be described later.

2. Power generation unit 20
As shown in FIG. 2, the power generation unit 20 is formed on the support substrate 10. The power generation unit 20 includes a fuel electrode 21, a solid electrolyte layer 22, a reaction prevention layer 23, and an air electrode 24.

  The fuel electrode 21 is formed on the support substrate 10 and functions as an anode. The thickness of the fuel electrode 21 may be 50 μm to 500 μm. The fuel electrode 21 includes a porous fuel electrode current collecting layer 21a excellent in gas permeability and a dense fuel electrode active layer 21b.

The anode current collecting layer 21 a is formed on the support substrate 10. The anode current collecting layer 21a includes a transition metal and an oxygen ion conductive material. The anode current collecting layer 21a may be a porous plate-like fired body. The anode current collecting layer 21a contains nickel oxide (NiO) and / or nickel (Ni) as a transition metal. The anode current collecting layer 21a includes, as an oxygen ion conductive substance, zirconia-based materials such as yttria stabilized zirconia (8YSZ, 10YSZ, etc.) and scandia stabilized zirconia (ScSZ), and gadolinium doped ceria (GDC: (Ce, Gd). ) O ₂ ) and ceria-based materials such as samarium-doped ceria (SDC: (Ce, Sm) O ₂ ) may be included. The thickness of the anode current collecting layer 21a can be set to 0.2 mm to 5.0 mm.

The anode active layer 21b is formed on the anode current collecting layer 21a. The anode active layer 21b contains a transition metal and an oxygen ion conductive material. The fuel electrode active layer 21b may be a dense plate-like fired body. The anode active layer 21b contains at least NiO and / or Ni as a transition metal. The anode active layer 21b may contain, in addition to NiO and / or Ni, Fe ₂ O ₃ (or FeO) and / or Fe, or CuO and / or Cu as a transition metal. The anode active layer 21b is made of zirconia-based materials such as yttria-stabilized zirconia (8YSZ, 10YSZ) and scandia-stabilized zirconia (ScSZ), and gadolinium-doped ceria (GDC: (Ce, Gd)). Ceria-based materials such as O ₂ ) and samarium-doped ceria (SDC: (Ce, Sm) O ₂ ) are included. The thickness of the anode active layer 21b can be 5.0 μm to 30 μm. In the anode active layer 21b, the volume ratio of Ni and / or NiO can be 25 to 50% by volume in terms of Ni, and the volume ratio of the oxygen ion conductive material can be 50 to 75% by volume. Thus, the content rate of the oxygen ion conductive material may be larger in the anode active layer 21b than in the anode current collecting layer 21a.

The solid electrolyte layer 22 is formed between the fuel electrode 21 and the air electrode 24. The solid electrolyte layer 22 can contain zirconia (ZrO ₂ ) as a main component. The solid electrolyte layer 22 can contain additives such as Y ₂ O ₃ and / or Sc ₂ O ₃ in addition to zirconia. These additives function as stabilizers. The addition amount of the additive in the solid electrolyte layer 22 is about 3 to 20 mol%. That is, examples of the material for the solid electrolyte layer 22 include zirconia-based materials such as yttria-stabilized zirconia such as 3YSZ, 8YSZ, and 10YSZ; and ScSZ (scandia-stabilized zirconia). The thickness of the solid electrolyte layer 22 is preferably 3 μm or more and 50 μm or less.

The reaction preventing layer 23 is formed on the solid electrolyte layer 22. The reaction preventing layer 23 may contain ceria (cerium oxide) as a main component. Specifically, examples of the material for the reaction preventing layer 23 include ceria and a ceria-based material containing a rare earth metal oxide solid-dissolved in ceria. Examples of the ceria-based material include GDC ((Ce, Gd) O ₂ : Gadolinium doped ceria), SDC ((Ce, Sm) O ₂ : samarium doped ceria) and the like. The thickness of the reaction preventing layer 23 is preferably 3 μm or more and 50 μm or less.

  The air electrode 24 is formed on the reaction preventing layer 23. The air electrode 24 functions as a cathode. The air electrode 24 may be made of, for example, a lanthanum-containing perovskite complex oxide. Examples of the lanthanum-containing perovskite complex oxide include LSCF (lanthanum strontium cobalt ferrite), lanthanum manganite, lanthanum cobaltite, and lanthanum ferrite. The lanthanum-containing perovskite complex oxide may be doped with strontium, calcium, chromium, cobalt, iron, nickel, aluminum, or the like. The thickness of the air electrode 24 is preferably 10 μm or more and 100 μm or less.

3. Interconnector 30
The interconnector 30 is formed on the anode active layer 21b. The interconnector 30 electrically connects two adjacent power generation units 20 in series. Specifically, the interconnector 30 is connected to the current collector 40 formed on one power generation unit 20 and the anode active layer 21 b of the other power generation unit 20. The interconnector 30 contains a perovskite complex oxide as a main component. In particular, examples of the perovskite complex oxide used in the interconnector 30 include chromite materials such as lanthanum chromite (LaCrO ₃ ) and titanate materials such as SrTiO ₃ . The thickness of the interconnector 30 is preferably 10 μm or more and 100 μm or less.

4). Current collector 40
The current collector 40 is formed on the air electrode 24 of one power generation unit 20. In addition, the current collector 40 extends on the interconnector 30 disposed between the two power generators 20. The current collector 40 only needs to have conductivity, and can be made of the same material as the interconnector 30 and the air electrode 24, for example. The thickness of the current collection part 40 can be 50 micrometers or more and 500 micrometers or less.

[Microstructure of Support Substrate 10]
Hereinafter, the microstructure of the support substrate 10 in the reduced state will be described.

1. Average value of bonding length In the support substrate 10, two adjacent insulating ceramic particles are bonded to each other. When the support substrate 10 includes only MgO as the insulating ceramic, the MgO particles are joined to each other, and the support substrate 10 is not only MgO but also MgAl ₂ O ₄ , CaZrO _3, and CaTiO ₃ as the insulating ceramic. When at least one is included, the same kind or different kinds of particles among the MgO particles, MgAl ₂ O ₄ particles, CaZrO ₃ particles, and CaTiO ₃ particles are bonded.

  Such a bonding state of the insulating ceramic particles is 5,000 times the cross section of the support substrate 10 by an FE-SEM (Field Emission Scanning Electron Microscope) using an in-lens secondary electron detector. The image can be observed by analyzing the image with the image analysis software HALCON manufactured by MVTec (Germany). As FE-SEM, FE-SEM (model: ULTRA55) manufactured by Zeiss (Germany) set to acceleration voltage: 1 kV, working distance: 2 mm, and visual field size: 18.75 μm × 25 μm can be used.

  When measuring the joining length of insulating ceramic particles in the image after analysis by the image analysis software, a minute joining portion having a predetermined value (for example, 0.1 μm) or less recognized by the image analysis software is not counted. May be. This is because the presence of minute joints recognized by image analysis software is uncertain, and there is little need to take it into account as a factor affecting cell performance.

  The average value of the length of the bonding line between the insulating ceramic particles (hereinafter referred to as “average value of the bonding length”) is preferably 0.5 μm or more and 2.2 μm or less. The average value of the bonding length can be obtained by dividing the sum of the bonding lengths of the insulating ceramic particles by the number of bonding points. The average junction length may be calculated from the analysis results in one visual field or may be calculated from the analysis results in a plurality of visual fields.

  Such an average value of the joining length is an index indicating the neck thickness between the insulating ceramic particles after reduction. In the present embodiment, the average junction length of the support substrate 10 in the reduced state is 0.5 μm or more and 2.2 μm or less, and sufficient strength of the skeleton structure of the support substrate 10 is maintained even after the reduction. Thereby, since the dimensional change of the support substrate 10 before and after reduction is suppressed, the cracks generated in the support substrate 10 due to the reduction process are suppressed.

In order to adjust the average joining length, for example, it is effective to control the particle size distribution of insulating ceramic raw materials (MgO raw material, MgAl ₂ O ₄ raw material, CaZrO ₃ raw material and CaTiO ₃ raw material). Further, the average value of the joining length can be finely adjusted by the average particle diameter and the added amount of the pore former, firing conditions, and the like.

An example of manufacturing conditions for controlling the average junction length to 0.5 μm or more and 2.2 μm or less is shown below.
・ The average particle size of the insulating ceramic material is 0.5 μm or more and 3.5 μm or less, and fine particles with a particle size of 0.2 μm or less are removed by classification treatment. ・ The average particle size of the pore former is 1 μm or more and 20 μm or less・ The amount of pore former added is 15% by volume or less of the ceramic raw material. ・ The co-firing temperature is 1350 ° C. or more and 1500 ° C. or less, and the treatment time is 2 hours or more and 20 hours or less.

2. Variation in average bonding length In an SEM image of 10 fields of view (5000 times magnification) that is arbitrarily acquired, the standard deviation of the average bonding length between insulating ceramic particles is preferably 1.5 or less.

  Such a standard deviation of the bond length average value is an index indicating the variation in the neck thickness between the insulating ceramic particles in the support substrate 10. By setting the standard deviation of the average bond length in the reduced state to 1.5 or less, the strength of the skeletal structure of the support substrate 10 after the reduction is made uniform, so that locally low strength portions are generated. Can be suppressed. Therefore, it is possible to improve the durability of the support substrate 10 in the process of repeating oxidation and reduction.

  In order to adjust the standard deviation of the average joining length, for example, it is effective to control the average particle diameter, mixing conditions, or firing conditions of the insulating ceramic raw material. Further, the standard deviation of the average junction length can be adjusted by controlling the mixing of impurities in the mixing process.

Below, an example of the production conditions for controlling the standard deviation of the bond length average value to 1.5 or less is shown.
・ Excludes fine particles of 0.15 μm or less and coarse particles of 5 μm or more in the MgO raw material
Leaving ・ Removing fine particles of 0.1 μm or less and coarse particles of 3.5 μm or more in MgAl ₂ O ₄ raw material, CaZrO ₃ raw material and CaTiO ₃ raw material ・ Co-firing temperature of 1350 ° C. or higher and 1500 ° C. or lower And the processing time is 1 hour or more and 10 hours or less

3. Porosity In the SEM image (one field of view) of the support substrate 10 magnified by 3000 times by FE-SEM using an in-lens secondary electron detector, the proportion of pores is 25% or more and 45% or less. It is preferable. Accordingly, the gas diffusion in the porous support substrate can be functioned without being inhibited.

[Method for Manufacturing Cell 100]
1. Formation of Support Substrate 10 The support substrate 10 can be formed by compacting. That is, the support substrate 10 includes a step of forming a green compact by putting powder mixed with the material of the support substrate 10 into a mold and compressing the powder.

As a material for the support substrate 10, an insulating ceramic material containing MgO can be used. Such a material may contain Fe ₂ O ₃ , SiO ₂ , B ₂ O ₃ , Al ₂ O _3, etc., but preferably does not contain NiO. The pressure applied to the powder at the time of compacting may be set so that the support substrate 10 has sufficient rigidity.

  However, the Si content in the material of the support substrate 10 is 200 ppm or less, the P content is 50 ppm or less, the Cr content is 100 ppm or less, the B content is 100 ppm or less, and The S content is preferably 100 ppm or less.

  Moreover, the gas flow path 10a is formed by performing compaction molding in a state where a cellulose sheet or the like that disappears by firing is embedded in the powder, and then performing firing.

2. Formation of Fuel Electrode 21 The fuel electrode 21 can be formed by compacting. That is, the fuel electrode 21 may include forming a green compact by putting powder mixed with the material of the fuel electrode 21 into a mold and compressing the powder. The fuel electrode 21 can be formed by a printing method. That is, the fuel electrode 21 may be formed on the support substrate 10 by screen printing using a paste containing the material of the fuel electrode 21.

  As the material of the fuel electrode 21, as described above, for example, nickel oxide, zirconia, and, if necessary, a pore former are used. The pore former is an additive for providing pores in the fuel electrode. As the pore former, a material that disappears in a later step is used. An example of such a material is cellulose powder.

  The anode 21 (specifically, the anode active layer 21b) includes 200 ppm or less Si, 50 ppm or less P, 100 ppm or less Cr, 100 ppm or less B, 100 ppm or less S, Is preferably added.

3. Formation of Solid Electrolyte Layer 22 The solid electrolyte layer 22 can be formed by, for example, CIP (cold isostatic pressing), thermocompression bonding, or slurry dip method. Examples of the material of the solid electrolyte layer 22 include zirconia-based materials such as yttria-stabilized zirconia such as 3YSZ, 8YSZ, and 10YSZ; and ScSZ (scandia-stabilized zirconia). In addition, the pressure at the time of pressure bonding of the sheet in the CIP method is preferably 50 to 300 MPa.

4). Formation of Reaction Prevention Layer 23 The reaction prevention layer 23 can be formed by a slurry dip method or the like. Examples of the material of the reaction preventing layer 23 include GDC ((Ce, Gd) O ₂ : Gadolinium doped ceria), SDC ((Ce, Sm) O ₂ : samarium doped ceria) and the like.

5. Firing includes co-firing (co-sintering) of the support substrate 10, the fuel electrode 21, the solid electrolyte layer 22, and the reaction prevention layer 23 that have been compacted. The firing temperature and time are set according to the cell material and the like.

6). Formation of the air electrode 24 After the air electrode 24 is formed, for example, on the laminate (fired body) of the fuel electrode 21, the electrolyte layer 22, and the reaction prevention layer 23, a material layer of the air electrode 24 is formed by a printing method or the like. It is formed by firing.

7). Formation of Current Collector 40 The current collector 40 is made of, for example, a material of the current collector 40 by a printing method or a slurry dip method on a laminate (fired body) of the fuel electrode 21, the electrolyte layer 22, and the reaction preventing layer 23. After the layer is formed, it is formed by firing. The air electrode 24 and the current collecting layer 40 may be formed by firing separately, or may be formed by sequentially stacking and firing together.

[Other Embodiments]
The present invention is not limited to the embodiment described above, and various modifications or changes can be made without departing from the scope of the present invention.

  In the above embodiment, an SEM image with a magnification of 5000 is used, but the present invention is not limited to this. The magnification of FE-SEM can be set arbitrarily.

  In the above embodiment, in the support substrate 10, the average joining length is 0.5 μm or more and 2.2 μm or less. However, this does not mean that there is no region where the average junction length is less than 0.5 μm or more than 2.2 μm. The support substrate 10 only needs to include at least one region having an average junction length of 0.5 μm or more and 2.2 μm or less.

  Moreover, in the said embodiment, although the electric power generation part 20 decided to have the reaction prevention layer 23, it does not need to have the reaction prevention layer 23. FIG.

1. Sample No. 1-No. Production of Fuel Battery Cell According to Sample No. 10 provided with a support substrate and a power generation unit as follows. 1-No. A fuel battery cell according to No. 10 was produced.

  First, the support materials described in Table 1 were weighed and mixed for 48 hours in a pot mill. Then, the slurry was produced by adding polyvinyl alcohol (PVA) as a binder and mixing for further 12 hours.

  Next, the slurry was dried with a drier and then granulated through a sieve having an opening of 150 μm.

Next, the granulated powder is compressed by a uniaxial press (molding pressure: 0.4 tf / cm ² ) to obtain a plate-shaped green compact (supporting substrate) having a length of 150 mm × width of 40 mm × thickness of 3.5 mm. Molded. In addition, ten φ2.0 mm channels for circulating the fuel gas were formed inside the plate-shaped green compact (support substrate). Further, in order to measure the dimensional change of the support substrate before and after the reduction treatment described later, sample No. 1-No. 10 support substrates were adjusted to dimensions of 45 mm in length, 4.0 mm in width, and 3.0 mm in thickness, and stored in units of 20 pieces.

  Next, an anode current collecting layer (NiO: 8YSZ = 50: 50 (Ni volume% conversion)) having a thickness of 150 μm formed by a tape forming method is adhesively formed on a plate-like green compact (support substrate), A fuel electrode active layer having a thickness of 20 μm was formed thereon by a printing method.

  Next, an 8YSZ electrolyte having a thickness of 5 μm and a GDC barrier film having a thickness of 5 μm were sequentially printed on the anode active layer to produce a laminate.

  Next, the laminate was co-sintered at 1400 ° C. for 2 hours to obtain a co-fired body. Thereafter, the LSCF air electrode having a thickness of 30 μm was baked at 1000 ° C. for 2 hours, whereby the sample No. of the horizontal stripe fuel cell shown in FIG. 1-No. 10 was produced.

2. Sample No. 1-No. No. 10 reduction treatment Sample No. 1-No. Each of the fuel cells and the support substrate according to 10 was subjected to reduction treatment. Specifically, each sample was heated from room temperature to 750 ° C. at 100 ° C./hr, and then hydrogen gas was passed through the flow path of the support substrate for 100 hours. Thereafter, Ar and hydrogen (4% with respect to Ar) were allowed to flow through the flow path, and the temperature of each sample was lowered from 750 ° C. to room temperature over 12 hours while maintaining the reducing atmosphere.

3. Sample No. 1-No. 10. Observation of cross section of fuel cell according to Sample No. 10 subjected to the reduction treatment 1-No. The cross section of the fuel cell according to No. 10 was observed to confirm the presence or absence of cracks in the support substrate.

  Moreover, the presence or absence of a crack was also confirmed by observing the presence or absence of foaming in a positive pressure environment (100 to 200 kPa) in which each sample was immersed in water.

  Sample No. 1-No. Table 1 shows the results of confirmation of cracks in each of the 10 support substrates.

4). Sample No. 1-No. Measurement of dimensional change of support substrate according to Sample No. 10 before and after the reduction treatment. 1-No. The dimensional change of 20 support substrates according to 10 was measured. Specifically, before and after the reduction treatment, the vertical dimension of each support substrate was measured with a precision micrometer at room temperature.

  In Table 1, the average value data obtained by measuring the vertical dimension of each support substrate 5 times each is obtained as a measurement result by obtaining 20 samples. In Table 1, “+” indicates expansion and “−” indicates contraction.

5. Average length of bonding between insulating ceramics Sample No. 1-No. By observing the cross section of the fuel cell according to No. 10, the average value of the joining lengths of the insulating ceramics was calculated.

  First, the support substrate of each sample was subjected to precision mechanical polishing, and then subjected to ion milling processing using IM4000 manufactured by Hitachi High-Technologies Corporation.

  Next, the SEM image of the cross section of the support substrate magnified by 3000 times by FE-SEM using an in-lens secondary electron detector was acquired. For obtaining SEM images, FE-SEM (model: ULTRA55) manufactured by Zeiss (Germany) set to acceleration voltage: 1 kV, working distance: 2 mm, visual field size: vertical 18.75 μm × horizontal 25 μm is used. It was. The cross section of the support substrate was preliminarily subjected to precision mechanical polishing treatment and ion milling treatment treatment using IM4000 of Hitachi High-Technologies Corporation.

  Next, the SEM image of the cross section was analyzed by image analysis software HALCON manufactured by MVTec (Germany), thereby detecting a joining line between the insulating ceramics. In addition, it set so that the fine joining line of 0.1 micrometer or less recognized by image analysis software may not be detected.

  Next, the average value of the joining length in the SEM image of one visual field was calculated using the analysis result of the image analysis software. The calculation results are shown in Table 1. In addition, in the sample containing only MgO, the joining length average value of MgO particles was calculated, and in the sample containing insulating ceramics other than MgO, the joining length average value of each insulating ceramic was calculated.

  As shown in Table 1, sample no. 2-No. In No. 9, it was possible to suppress cracks after reduction by controlling the average value of the junction length to 0.5 μm or more and 2.2 μm or less. This is because the amount of strain at the time of reduction of the porous support substrate could be suppressed by controlling the average value of the joining length to be 0.5 μm or more and 2.2 μm or less.

  On the other hand, sample no. In No. 1, since the average length of the joint was too small, that is, the neck thickness at the joint was small, the dimensional change due to the reduction treatment was large. Therefore, a crack occurred in the support substrate after reduction.

  Sample No. In No. 10, since the average value of the bonding length was too large, the variation in the average value of the bonding length was increased, and it is considered that local dimensional change occurred in the support substrate. As a result, the dimensional change due to the reduction treatment increased, and cracks occurred in the support substrate after reduction.

6). Durability of fuel cell Sample no. Sample No. 3 by mixing the amount of the mixture shown in Table 2 (silicon (Si), phosphorus (P), chromium (Cr), boron (B), and sulfur (S)) with reference to 3 on the support substrate. . Fuel cells according to 3-1 to 3-5 were produced. The mixture is premixed in the raw material of the support substrate, but may be mixed in the process of manufacturing the support substrate.

  And sample no. Regarding 3-1 to 3-5, the temperature is raised to 750 ° C. while supplying nitrogen gas to the fuel electrode side and air to the air electrode side, and when 750 ° C. is reached, reduction treatment is performed while supplying hydrogen gas to the fuel electrode. For 3 hours.

  Subsequently, a cycle in which each sample was heated from room temperature to 750 ° C. at 200 ° C./hr and then cooled from 750 ° C. to room temperature at 100 ° C./hr was repeated 10 times.

  Thereafter, the sample No. subjected to the above heat cycle treatment. The cross section of the fuel cell according to 3-1 to 3-5 was observed to confirm the presence or absence of peeling at the interface between the support substrate and the fuel electrode.

  As shown in Table 2, sample no. In 3-2 to 3-5, the Si content is 200 ppm or less, the P content is 50 ppm or less, the Cr content is 100 ppm or less, the B content is 100 ppm or less, and It was confirmed that when the S content is 100 ppm or less, peeling at the interface between the support substrate and the fuel electrode can be suppressed.

  Such a result was obtained because the bonding strength between the support substrate and the fuel electrode was improved by connecting Si, P, Cr, B and S mixed with the support substrate to the material constituting the fuel electrode. This is because.

  On the other hand, sample no. In 3-1, peeling was confirmed at the interface between the support substrate and the fuel electrode. This is because the bonding strength between the support substrate and the fuel electrode is reduced by the segregation of the mixture at the interface between the two.

  It has been experimentally confirmed that it is preferable to mix at least 1 ppm of Si, P, Cr, B, and S.

DESCRIPTION OF SYMBOLS 100 Horizontal stripe type fuel cell 10 Support substrate 20 Power generation part 21 Fuel electrode 21a Fuel electrode current collection layer 21b Fuel electrode active layer 22 Solid electrolyte layer 23 Reaction prevention layer 24 Air electrode 30 Interconnector 40 Current collection part

Claims (3)

Has a fuel gas flow passage therein, a supporting substrate made of an insulating ceramic containing MgO,
A power generation unit formed on the support substrate and having a fuel electrode, an air electrode, and a solid electrolyte layer disposed between the fuel electrode and the air electrode;
With
In the case where the support substrate is reduced, the average joining length of two adjacent insulating ceramic particles is 0.5 μm or more and 2.2 μm or less,
The support substrate does not contain Ni,
Ni is not dissolved in each of the two insulating ceramic particles.
Horizontally striped solid oxide fuel cell. The insulating ceramic contains at least one of MgAl ₂ O ₄ , CaZrO ₃ and CaTiO ₃ ,
The two insulating ceramic particles are MgO particles, MgAl ₂ O ₄ particles, CaZrO ₃ particles, and CaTiO ₃ particles, the same kind or different kinds of particles.
The horizontally-striped solid oxide fuel cell according to claim 1. The insulating ceramic does not contain Y ₂ O ₃ ,
The horizontally striped solid oxide fuel cell according to claim 2 .

Images