The present invention relates to a composite for an electrode, a composite precursor, a method of manufacturing a composite, and the composite obtained by said method. The invention further relates to an electrode comprising the composite, as well as a solid state electrochemical cell comprising said composite. The invention also relates to the use of the composite as a fuel electrode, solid oxide fuel cell, and/or solid oxide electrolyser.
State-of-the-art electrodes for solid state electrochemical cells, such as oxide ion conducting or proton conducting solid oxide fuel cells and electrolysis cells, have a three-dimensional porous microstructure to obtain high electrochemical reaction rates (also known as high performance). The reaction sites are locations where electrons, oxygen ions and chemical reactants and products can be transported to/from, and hence where the chemical reactions occur. Therefore, these types of porous electrodes must be comprised of materials with electronic and ionic conductivities and they must have sufficient porosity to allow diffusion of the reactants and products to and from reactions sites.
To provide the electronic, ionic and gas pathways, electrodes are typically mixed-particle composites composed of necked particles, which may be fabricated by cheap wet-ceramic techniques such as tape-casting, spraying, screen-printing and laminating, followed by sintering. One example is the composite (also known as cermet) of nickel (Ni) particles and yttria-stabilized zirconia (YSZ) particles. The Ni-YSZ composite is for instance used as fuel electrode in oxide ion conducting cells, and the Ni provides electronic pathways, the YSZ provides ionic pathways and the reaction sites are the three-phase boundaries between Ni, YSZ and the gas phase.
The Ni-YSZ type electrode has been optimized significantly in the past decade. To improve the performance of the electrode, a focus has been on reducing the particle size of the composite. However, smaller particles are known to be less stable, and the long-term stability or durability of the composite structure is an issue. The stability of the metal of the cermet (e.g. Ni) is known to be particularly low during operation of a fuel electrode. The Ni phase will migrate and sinter, resulting in growth, coarsening and agglomeration of the Ni particles, thereby causing loss of electronic pathways and decreased reaction rates.
Different approaches to minimise the structural degradation of the composite electrode, and thus improve the durability, have been described. U.S. Pat. No. 7,029,777 [1] describes a composite for a fuel electrode, where the pore size and morphology is selected to limit metal (e.g. Ni) agglomeration. It is illustrated that the agglomeration decreases with smaller and asymmetric pore size, which is mainly ascribed to steric limitations. The fuel electrode is formed within a coarser pore region of a passive support, which is exemplified to have a mean pore size of 5 μm, and the electrode is further exemplified to have a mean pore size less than 1 μm, and secondary porosity with mean pore size of approximately 50-500 nm. Performance or stability data of the described structures are not disclosed in [1].
Other approaches to reduce particle size and reduce porosity have been described. Kishimoto et al. [2] describes a wet ceramic processed Ni-YSZ composites with 25.8% porosity, and with average particle sizes of 1 μm for Ni, and 700 nm for YSZ. However, the particle size distributions were indicated to be broad, and from the electrochemical test it can be construed that the composite showed limited long-term stability.
Lay-Grindler et al. [3] describes a Ni-YSZ composite with a low porosity of 20%. However, very limited long-term stability was observed. This was seen as a doubling of the Ni particle sizes after 1000 hours of operation.
The present invention provides a novel composite for an electrode, wherein the composite provides high performance and surprisingly long-term durability. This may result in a significant reduction of system cost as the lifetime of the cells is increased remarkably. In addition to these advantages, the novel composite can be manufactured using simple and cheap wet ceramic processes.
A first aspect of the invention relates to a composite for an electrode, comprising a three-dimensional network of dispersed metal particles, stabilised zirconia particles and pores, wherein the size of the pores is smaller than the size of the metal particles, and wherein the size of the metal particles is essentially equal to or smaller than the size of the stabilised zirconia particles, wherein the porosity is below 33, 30, or 29 vol %, more preferably below 26 or 24 vol %, and most preferably below 23, 22, 21, 18, 15, or 13 vol %, and/or wherein the pores are essentially exclusively generated from the volume created by reducing a corresponding metal oxide to the metal particles.
A second aspect of the invention relates to an electrode for an electrochemical cell, comprising the composite according to the first aspect of the invention.
A third aspect of the invention relates to a solid state electrochemical cell, comprising the composite according to the first aspect of the invention.
A fourth aspect of the invention relates to the use of the electrochemical cell according to the third aspect of the invention as a solid oxide fuel cell, and/or a solid oxide electrolyser.
A fifth aspect of the invention relates to a composite precursor for an electrode, comprising a three-dimensional network of dispersed particles of a metal oxide and a stabilised zirconia,
wherein the characteristic size of the particle size distribution is d10 between 0.03-0.07 μm, d50 between 0.08-0.4 μm, d90 between 0.4-1.2 μm, more preferably d10 between 0.04-0.06 μm, d50 between 0.1-0.3 μm, d90 between 0.9-1.1 μm, and most preferably d10=0.05 μm, d50=0.2 μm, and d90=1 μm, and/or wherein the porosity is below 5 vol %, more preferably below 3 vol %, and most preferably below 1 vol %.
A sixth aspect of the invention relates to a method of manufacturing a composite, comprising the steps of:
A seventh aspect of the invention relates to the composite for an electrode according to the second aspect of the invention obtained by the the sixth aspect of the invention.
The invention will in the following be described in greater detail with reference to the accompanying drawings.
FIG. 2 shows SEM micrographs of a standard composite (Ex. 1) of an electrolysis cell. The initial microstructure is shown in (b), and the microstructure after 1000 hours testing at 800° C., −1 A/cm2, H2O/H2:90/10 (inlet), 56% H2O conversion, and O2 to the oxygen electrode, is shown in (a). Arrows mark places with significant changes in the microstructure after the long-term testing.
The data from
The invention provides a composite for an electrode showing high performance as well as surprisingly long-term durability. The high and improved properties are related to the composite structure.
In an embodiment of the invention, the composite comprises a three-dimensional network of dispersed metal particles, stabilised zirconia particles and pores, wherein the size of the pores is smaller than the size of the metal particles, and wherein the size of the metal particles is similar or smaller than the size of the stabilised zirconia particles. In another embodiment of the invention, the composite comprises a three-dimensional network of dispersed metal particles, stabilised zirconia particles and pores, wherein the size of the pores is smaller than the size of the metal particles, wherein the size of the metal particles is essentially equal to or smaller than the size of the stabilised zirconia particles, wherein the porosity is below 33, 30, or 29 vol %, more preferably below 26 or 24 vol %, and most preferably below 23, 22, 21, 18, 15, or 13 vol %, and/or wherein the pores are essentially exclusively generated from the volume created by reducing a corresponding metal oxide to the metal particles.
By the term “similar particle size” is meant particles that have essentially equal sizes, or are about or approximately or substantially the same sizes, as would be known to the skilled person within the field.
In a further embodiment of the invention, the composite for an electrode, comprising a three-dimensional network of dispersed metal particles, stabilised zirconia particles and pores, wherein the characteristic size of the pores is smaller than the characteristic size of the metal particles, and wherein the characteristic size of the metal particles is similar or smaller than the characteristic size of the stabilised zirconia particles. In another embodiment, the characteristic size of the pores is smaller than the characteristic size of the metal particles, and wherein the characteristic size of the metal particles is essentially equal to or smaller than the characteristic size of the stabilised zirconia particles.
By the term “composite” as used herein is meant a material made of two or more materials with different physical or chemical properties, such as a composite of a ceramic material and a metallic material (also known as a cermet). An example of a composite is NiO-YSZ, and an example of a cermet is the Ni-YSZ cermet.
By the term “metal” as used herein is meant a material, compound or alloy comprising a metal element, and further showing significant electronic conductivity. Examples of metals include: nickel (Ni), copper (Cu), chromium (Cr), iron (Fe), and a combination thereof.
By the term “stabilised zirconia” as used herein is meant zirconium dioxide (ZrO2) doped (or stabilised) with another oxide, such as yttria (Y2O3), scandia (Sc2O3), magnesia (MgO), calcia (CaO), ceria (CeO2), or a combination thereof. Example include yttria-stabilised zirconia (Y2O3 doped ZrO2, also denoted YSZ), and yttria- and scandia co-stabilised zirconia (ZrO2 co-doped with Y2O3 and Sc2O3, also denoted ScYSZ).
The composite may also be described as the stabilised zirconia particles providing a ceramic ion-conducting phase, and dispersed in such a way to form a fine, dense, and well dispersed microstructure that hinders migration and sintering of the metallic phase e.g. Ni phase, when the composite is operated as an electrode, whereby high performance and long-term durability is obtained. In particular, high performance and long-term durability of an electrode at high current density (i.e. high fuel production rate for electrolysis cells) can be obtained. Furthermore, the composite of the invention provides a fuel electrode, which may be used in an electrolysis cell and that enables the electrolysis cell to run for several years at thermoneutral potential with fuel production rate corresponding approximately 0.45 l/h H2 produced per cm2 electrode area.
The stability of the composite when e.g. operated as an electrode depends on the stability of the metal phase. At elevated temperatures, a metal phase will tend to migrate and sinter, and the smaller the size of the metal phase, the stronger the tendency to migrate and sinter. If the migration and sintering of the metal phase is impeded, the composite will be more long-term durable.
The migration of a metal may be sterically inhibited by other phases, and thus by low porosity. In the current invention, the sintering of the metal phase in the composite may be inhibited by low porosity, and/or the arrangement of the pores.
The arrangement of the pores will depend on how the porosity is formed. The porosity may be obtained by reducing a corresponding oxide of a metal to the metallic form. For example, the reduction of NiO to Ni will result in pore volume due to the difference in density between NiO and Ni.
The porosity in a produced Ni-YSZ cermet may arise nearly exclusively from the volume created by reducing NiO to Ni. In a cermet with a ratio of Ni/YSZ of 50/50 this leads to a porosity of approximately 26%. In this case, the porosity may be further restricted to the vicinity of the reduced metal phase. In
In an embodiment of the invention, the porosity of the composite is generated by reducing a corresponding oxide of the metal to the metallic form. In a further embodiment, the porosity of the composite is generated exclusively from the volume created by reducing NiO to Ni.
When the porosity of the composite is generated exclusively from the volume created by reducing NiO to Ni, the pores or the porosity of the composite may be defined as essentially equal to the geometrical volume of the original NiO minus the geometrical volume of the Ni.
In a further embodiment of the invention, the porosity of the composite is essentially equal to the volume of the original NiO minus the volume of the Ni.
In a further embodiment of the invention, the generated porosity is in the vicinity of the metal phase. In a further embodiment, the porosity is below 33, 30 or 29 vol %, more preferably below 26 or 24 vol %, and most preferably below 23, 22, 21, 18, 15, or 13 vol %.
The metal of the composite should preferably have a corresponding oxide, which can be easily reduced to the metal phase, e.g. by exposure to reducing atmospheres, such as gasses comprising hydrogen. In an embodiment of the invention, the metal is selected from the group consisting of Ni, Cu, Cr, Fe, and any combination thereof. In a further embodiment, the corresponding oxide of the metal is selected from the group consisting of NiO, CuO, Cu2O, CrO2, Cr2O3, FeO, Fe2O3, Fe3O4, and any combination thereof.
The performance and stability of a composite for an electrode comprising metal particles, stabilised zirconia particles and pores, may also depend on the composite mixture, and the chemcial composition of the stabilised zirconia.
In an embodiment of the invention, the stabilised zirconia is selected from the group consisting of yttria-stabilised zirconia, scandia-stabilised zirconia, yttria- and scandia co-stabilised zirconia, magnesia-stabilised zirconia, calcia-stabilised zirconia, ceria-stabilised zirconia, or a combination thereof.
The metal and the stabilised zirconia comprise the solid phase of the composite. Different ratios between the in the solid phase may be formed as shown in
In a further embodiment, the composite comprises a metal volume fraction of the solid phase of 70, 60, 50, 45, 40, 30, or 20 vol %, most preferably essentially 30 vol %.
In a further embodiment of the invention, the volume ratio between metal and stabilised zirconia is 70:30, 60:40, 50:50, 45:55, 40:60, 30:70, or 20:80, most preferably 40:60.
The particle size of a spherical particle is unambiguously defined by its diameter or radius. However, for most microstructures, including composites as for example shown in
In most cases, a particle size distribution exists. An example of a distribution is shown in
In a three-dimensional (3D) network of dispersed particles, the particles may interact, and the structure may not be well described as, or consist of individual discrete particles, but rather the structure consists of continuous phase networks of varying thickness. To describe the distribution of sizes (PSD) in such a network, a 3D continuous particle size distribution is used. The continuous PSD characterizes the network by covering all parts of the network with the largest sphere that will fit inside the structure. The spheres can overlap and each part of the structure is only assigned to the largest sphere that covers it even if smaller spheres cover it as well. The continuous PSD is identical to the the conventional PSD for discrete spherical particles but extends intuitively to more complex structures. As an example the continuous PSD of a long cylinder consists of a single peak at the radius of the cylinder, since the entire volume of the cylinder can be covered by spheres the same size as the cylinder radius. The 3D continuous PSD is a computational method that works by analysing a reconstructed 3D image of a microstructure. Thus, the result of the 3D continuous PSD is the particle size distribution of the fitted spheres used to describe the network.
By the term “characteristic size” as used herein is meant the size as evaluated by 3D continuous particle size distribution. The characteristic size can be used to describe the network of the metal particle phase, the stabilised zirconia phase, and/or the pore phase. The characteristic size distribution as obtained by the 3D continuous PSD may be further described by the “average characteristic diameter”, which is the average diameter (or radius) of the size distribution of fitted spheres. The characteristic size may also be described by the “maximum characteristic diameter”, which is the diameter of the biggest fitted sphere or particle. The characteristic size may also be described by the median diameter or radius (r50), which is the diameter or radius below which 50% of the particles or fitted spheres, have diameter or radius.
The 3D continuous PSD method may be used for a composite structure by combination with 3D image analyses, e.g. where the 3D image data are obtained through focused ion beam (FIB) tomography and scanning electron microscopy (SEM). Based on FIB-SEM, a 3D re-construction of the structure can be obtained, and then 3D continuous PSD is used to obtain a characteristic size distribution of the phases, such as the pores, metal, and stabilised zirconia.
The characteristic sizes of the pore phase may be the average or median (r50), the maximum characteristic pore diameter, and FWHM. Similar characteristic sizes for the metal phase, and the stabilised zirconia phase can be obtained.
The performance of a composite for an electrode is dependent on the particle sizes of the composite. In general, the finer the particle size, the better the performance. However, the finer the metal particle size, the worse the stability as described earlier. The current invention provides a composite with fine particle sizes, and high stability due to the combination of particle size, in particular metal particle size, and pore microstructure and pore sizes.
In an embodiment of the invention, the maximum characteristic pore diameter is below 700 nm, more preferably below 650 nm or below 600 nm, and most preferably below 550 nm, such as 500 nm.
In a further embodiment of the invention, the median characteristic pore diameter is between 100-630 nm, more preferably between 200-500 nm, and most preferably between 300-400 nm or 350-400 nm.
In a further embodiment of the invention, the maximum characteristic metal particle diameter is below 1100 nm, more preferably below 900 nm or below 800 nm, and most preferably equal to or below 700 nm.
In a further embodiment of the invention, the median characteristic metal diameter is between 200-900 nm, more preferably between 300-800 nm or 400-700 nm, and most preferably between 500-650 nm or 550-600 nm.
A three-dimensional (3D) network of dispersed particles of different phases may be described by other parameters than the characteristic size of the phases. The interface area describes the degree of contact between two phases, for example the degree of contact between the stabilised zirconia (e.g. YSZ), and the metal (e.g. Ni). The YSZ/Ni interface area may have the unit μm2/μm3, that is the interface area per electrode volume.
In a composite for an electrode, the active reaction sites are related to the interface between the phases. Thus, it is advantageous to increase the interface area per volume.
In an embodiment of the invention, the network of metal particles and stabilised zirconica particles are dispersed such that the interface area between the stabilised zirconia and metal is above 0.8 μm2/μm3, more preferably above 1 or 1.2 μm2/μm3, and most preferably equal to or above 1.4 μm2/μm3.
In a further embodiment, the network of metal particles and stabilised zirconica particles has a volume specific interface area between the stabilised zirconia and metal above 0.8 μm2/μm3, more preferably above 1 or 1.2 μm2/μm3, and most preferably equal to or above 1.4 μm2/μm3, and/or wherein the percolating TPB defined by the network of metal particles, stabilised zirconia particles and pores has a volume specific length above 1.5 μm/μm3, more preferably above 2 or 2.5 μm/μm3, and most preferably equal to or above 3.0 or 3.2 μm/μm3.
The interface area may be determined based on a 3D image analyses, such as a 3D reconstruction based on FIB/SEM. For example, to determine the YSZ/Ni interface of a composite, the YSZ/Ni interface is reconstructed as a mesh of polygons, and then taking the total sum of the area of all polygons.
In a composite for an electrode for a solid oxide cell, the active reaction sites are at the the percolating triple phase boundary (TPB). The TPBs are locations in the electrode structure at which electrons, ions and chemical reactants and products can be transported to/from, and where the electron conducting and ion conducting phase meets the pore. Thus, percolating TPB is the lines where the metal phase meets the stabilised zirconia and pore phase, and in addition where said metal phase, stabilised zirconia phase and pore phase are percolated, i.e. each phase is connected to the exterior of the composite.
From the 3D reconstruction, obtained via FIB-SEM, it is possible to determine the electrode volume specific length of the percolating triple phase boundary (TPB). The TPBs form connected curve loops inside the microstructure and are measured simply by measuring the total length of these lines from the reconstructed 3D image data. To obtain the percolating TPB length a connectivity test is made for each segment of the TPB curve to test whether the TPB segment has a percolating network pathway to the borders of the 3D reconstruction through each of the two solid phases and the pores. Only TPB curve segments that pass this test are counted in the percolating TPB length. The unit of the percolating TPB is μm/μm3; that is TPB curve length per electrode volume.
Since the active reaction sites are related to the percolating TPB, it is advantageous to increase the TPB length per volume.
In an embodiment of the invention, the network of metal particles, stabilised zirconia particles and pores are dispersed to form a percolating TPB, wherein the percolating TPB has a length above 1.5 μm/μm3, more preferably above 2 or 2.5 μm/μm3, and most preferably equal to or above 3.0 or 3.2 μm/μm3.
Before reduction of the metal oxide to the corresponding the metal, the composite comprises stabilised zirconia, metal oxide, and essentially none or low porosity, as the pores are mainly formed as the metal oxide is reduced. Thus, the structure comprising the metal oxide is considered a precursor to the composite.
The precursor may be fabricated from powders of the metal oxide and the stabilised zirconia, and the precursor and composite may be formed from the powders using ceramic processing means. Typical ceramic processing means involve wet ceramic processing, where the powder are dispersed in a liquid for mixing and milling of the powders, whereby the particle size of the powders are reduced or the particle shape changed.
The precursor particle size of the powders may be measured while dispersed and/or suspended in the liquid, also known as slurry. As for the composite product, the shape of the powder particles are in most cases not spherical and a size distribution exist.
When carrying out particle sizing for slurry preparation, laser diffraction is applied. Laser diffraction measures particle size distributions by measuring the angular variation in intensity of light scattered as a laser beam passes through a dispersed particulate sample. Large particles scatter light at small angles relative to the laser beam and small particles scatter light at large angles. The angular scattering intensity data is then analyzed to calculate the size of the particles responsible for creating the scattering pattern, using the Mie theory of light scattering, which describes the interaction of light with a particle of arbitrary size. The particle size is reported as a volume equivalent sphere diameter, i.e. for a non-spherical particle, the size is ascribed as the diameter of the sphere with same volume as the particle.
The precursor particle size distribution and the precursor size may be evaluated as precursor characteristic size or diameter, where the precursor characteristic diameter means the diameter as obtained by volume equivalent sphere and/or laser diffraction means. The precursor characteristic diameters may be a size distribution, which may be described by the parameters “d10”, “d50”, and “d90”. “d50” (also written as d50) denotes the precursor characteristic particle diameter below which half (50%) of the particles have diameters below, and the other half have diameters above (i.e. it is a median value). Correspondingly, “d10” and “d90” denotes the precursor characteristic particle diameter below which respectively 10% and 90% of the particles have diameters below.
The parameters d10, d50, and d90 may be obtained by laser diffraction means. For example, for a slurry of NiO and YSZ, the particle sizes of the combined NiO and YSZ may be measured by use of LS 13 320 Laser Diffraction Particle Size Analyser from Beckman Coulter. The measurement may be carried out in Ethanol with a PIDS sample loading of 45%. For Refractive index and imaginary part 2.2 and 0.1 may be used.
The particle sizes of the precursor will be reflected in the characteristic particle size distribution of the composite, and the porosity (or packing density) of the precursor (also called green body) will be reflected in the porosity of the composite.
In an embodiment of the invention, the composite precursor for an electrode, comprises a three-dimensional network of dispersed particles of a metal oxide and a stabilised zirconia, wherein the precursor characteristic particle size distribution is tri-modal, and wherein the precursor characteristic size is d10 between 0.03-0.07 μm, d50 between 0.08-0.4 μm, d90 between 0.4-1.2 μm, more preferably d10 between 0.04-0.06 μm, d50 between 0.1-0.3 μm, d90 between 0.9-1.1 μm, and most preferably d10=0.05 μm, d50=0.2 μm, and d90=1 μm.
In a further embodiment of the invention, the porosity of the composite precursor is below 5 vol %, more preferably below 3 vol %, and most preferably below 1 vol %.
In another embodiment of the invention, the composite is obtained by a method of manufacturing a composite for an electrode, comprising the steps of:
In another embodiment of the invention, a method of manufacturing a composite is provided, comprising the steps of:
The ceramic processing means, such as mixing conditions and milling conditions, that are sufficient to obtain the precursor described above, and the precursor with the characteristic particle size distribution as described above, will be known to the person skilled within the art of ceramic processing.
The order of the ceramic processing means may influence the composite microstructure.
In a further embodiment of the invention, one or more ceramic processing means in step (b) is applied to the metal powder prior to the stabilised zirconia powder. In a further embodiment, the metal powder is milled before the stabilised zirconia powder.
A composite of Ni-YSZ is conventionally used as fuel electrode in solid oxide cells. The composite is typically manufactured using wet ceramic techniques, where the components of the composite (i.e. non-reduced NiO and YSZ) is first mixed by mixing suspensions of the powders of NiO and YSZ. The mixing is typically carried out by using ball milling of the oxide powders in a solvent with various additives. The particle size distribution of the suspension (i.e. the composite precursor) is typically evaluated by laser diffraction.
The precursor characteristic particle size distributions obtained by laser diffraction of the NiO-YSZ suspensions from Examples 1 and 2 are shown in
In contrast to the convention precursor particle size distribution, the precursor of the current invention shows a trimodal distribution, or a distribution with three or more peaks, and generally smaller precursor characteristic particle diameters (d10: 0.05 μm; d50: 0.2 μm; d90: 1 μm).
The different precursor size distribution of the current invention was obtained by separately stabilizing and milling the NiO particles to an adequate size prior to the dispersing and milling of the YSZ powder.
Furthermore, the precursor particle size distribution and the dispersion may be obtained by stabilizing the suspension with an electro-steric dispersant to avoid agglomeration during the further preparation.
Other ceramic processing means could be used to obtain the precursor characteristic particle size distribution of Example 2. The optimal particle size could also be obtained by synthesizing the composite either by spray-drying nitrates of respective oxides, followed by a calcination step or by other synthesis methods i.e. hydrothermal syntheses.
To form a Ni-YSZ fuel electrode, the NiO-YSZ precursor suspension is shaped by a wet ceramic technique such as tape casting, where a thin layer of suspension with well-defined thickness is formed. The shaped dried layer is the green body, and the thickness of the dried layer is the green thickness. The NiO-YSZ layer may be joined together with other layers e.g. the YSZ electrolyte layer by lamination or multilayer tape casting. The layer(s) are subsequently consolidated by heat treatment (sintering), in where the fuel electrode is densified during the co-sintering process with i.e electrolyte and support. The co-sintering shrinkage properties of the layers ensure the prober density of the fuel electrode and an oxygen electrode is hereafter applied by e.g. screen printing. The NiO component of the composite is reduced to Ni by exposure to a reducing atmosphere, such as a gas comprising hydrogen after heat treatment of the full cell including the oxygen electrode.
The performance of the Ni-YSZ fuel electrode may be tested in an electrolysis test set-up, using e.g. an alumina test house equipped with proper gas inlet, gas distribution system, electrical contacting to the cell and sealing. The performance of the cell is reflected by the cell voltage and the resistances (Ohmic resistance, Rs, and polarization resistance, Rp) of the cell, and the stability of the cell is reflected by the changes in these. Thus, degradation is seen as increased cell voltage or increased Rs or Rp over time.
The difference in long-term stability of the Ni-YSZ fuel electrode of Examples 1 and 2 can mainly be ascribed to the Ni-YSZ fuel electrode, and the changes of the microstructure, such as migration and sintering of the Ni phase. This was seen by comparing micrographs before test and after test for the two electrodes as shown in
In-lens micrographs of the Ni-YSZ electrodes of Examples 1 and 2 are also shown in
The micrographs of
From
The YSZ/Ni interface area and the percolating TPB is a measure of the dispersity of phases in the sintered and reduced structure, as all three phases (pore, electron conducting and oxygen ion conducting) shall be a part of a percolating network in the electrode as well as, a large YSZ/Ni interface area inhibits agglomeration during operation.
The invention is further described by the examples provided below.
The first step comprises individual tape-casting of four layers on a polymeric support. Layer 1: Fuel electrode support, Layer 2: Fuel electrode, Layer 3: Electrolyte and Layer 4: Barrier layer. The suspensions for tape-casting are manufactured by means of ball milling oxide powders in ethanol, with poly-pyrolidone (PVP), polyvinyl butral (PVB), polyethylene glycol (PEG) and S-2075 as additives.
Layer 1: The suspension comprises yttria stabilized zirconia (YSZ) and 53 vol % NiO, the porosity is about 30% after sintering and reduction. The green thickness is in the range of 350 μm.
Layer 2: The suspension comprises YSZ and 53 vol % NiO. The suspension is the precursor for the Ni-YSZ (40:60) composite. The particle size distribution is shown in
After tape casting, the green thickness is in the range of 12 μm. The porosity is 29% after sintering and reduction.
Layer 3: The suspension comprises YSZ with green thickness is in the range of 8 μm.
Layer 4: The suspension comprises Ce0.9Gd0.1O1.95 (CGO10) with green thickness is in the range of 5 μm.
The second step comprises assembling of the single tape-cast layers by lamination.
In the third step is the laminated half-cell cut into the desired shape by knife punching, resulting in sintered areas of 12×12 cm2.
In the fourth step is the half-cell sintered for about 12 h at 1300° C., with a heating ramp of 15° C./h to 600° C. and 60° C./h to 1300° C.
The fifth step comprises deposition of oxygen electrode by screen-printing. Screen-printing ink comprises 1:1 weight ratio mixture of (La0.6Sr0.24)xCoO3-δ (X=1.0-0.99) and CGO10 in a dowanol based PVB vehicle. The thickness is 30 μm.
The sixth step is sintering of the cell at about 930° C. for 24 h.
The seventh step is reduction of the cell, after sealing, at 850° C. for 2 hour in pure H2
The structure of the active fuel electrode has after this procedure among others the following characteristics: rpore: 0-700 nm; rYSZ, Ni: 0-1100 nm, rYSZ<r,Ni. YSZ/Ni interface area: 0.8 μm2/μm3 and percolating TPB: 1.5 μm/μm3
The cell was tested in a conventional set-up for 1000 hours under the conditions of 800° C., −1 A/cm2, H2O/H2:90/10 (inlet), 56% H2O conversion, and O2 to the oxygen electrode. The performance as a function of time was measured via recording the cell voltage over time and by recording electrochemical impedance spectra during the galvanostatic electrolysis test. The results are shown in
A significant degradation of the cell performance (increase in cell voltage) was observed. The cell voltage increased almost 250 mV (cf.
The cause of the degradation was evaluated to be due to the composite of Layer 2, which was investigated by scanning electron microscopy before and after test. SEM micrographs of the composite (Layer 2) before test (i.e. initial structures) are seen in
The pore and particle sizes (represented as r50), FWHM and other microstructural characteristics of the composite before test, were also evaluated by 3D image analysis obtained by focused ion beam (FIB) and scanning electron microscopy (SEM). Based on FIB-SEM, a 3D construction of the structure were obtained, and thus a size distribution of the particles. The characteristics of the initial microstructure are summarised in
SEM micrographs of the composite (Layer 2) after test are shown in
The first step comprises individual tape-casting of four layers on a polymeric support. Layer 1: Fuel electrode support, Layer 2: Fuel electrode, Layer 3: Electrolyte, Layer 4: Barrier layer. The suspensions for tape-casting are manufactured by means of ball milling oxide powders in ethanol, with poly-pyrrolidone (PVP), polyvinyl butyral (PVB), polyethylene glycol (PEG) and S-2075 as additives.
Layer 1: The suspension comprises yttria stabilized zirconia (YSZ) and 53 vol % NiO, the porosity is about 30% after sintering and reduction. The green thickness is in the range of 350 μm.
Layer 2: The suspension comprises YSZ and 53 vol % NiO. The suspension is the precursor for the Ni-YSZ (40:60) composite. NiO powder is dispersed and milled separately before addition of YSZ powder. The particle size distribution is shown in
After tape casting of the suspension, the green thickness is in the range of 12 μm. The porosity is 22% after sintering and reduction.
Layer 3: The suspension comprises YSZ with green thickness is in the range of 8 μm.
Layer 4: The suspension comprises CGO10 with green thickness is in the range of 5 μm.
Layer 3 and 4 are manufactured as a multilayer structure by co-cast of the electrolyte, Layer 3 on the top of the barrier layer, Layer 4.
The second step comprises assembling of the single tape-cast layers and the co-cast layer 3 and 4 by lamination.
In the third step is the laminated half-cell cut into the desired shape by knife punching, resulting in sintered areas of 12×12 cm2.
In the fourth step is the half-cell sintered for about 12 h at 1300° C., with a heating ramp of 15° C./h to 600° C. and 60° C./h to 1300° C.
The fifth step comprises deposition of oxygen electrode by screen-printing. Screen-printing ink comprises 1:1 weight ratio mixture of (La0.6Sr0.4)xCoO3-δ (X=1.0-0.99) and CGO10 in a dowanol based PVB vehicle. The thickness is 30 μm
The sixth step is sintering of the cell at about 930° C. for 24 h.
The seventh step is reduction of the cell, after sealing, at 850° C. for 2 hour in pure H2
The structure of the active fuel electrode has after this procedure among others the following characteristics: rpore: 0-700 nm; rYSZ,Ni: 0-300 nm, rYSZ˜r,Ni. YSZ/Ni interface area: 1.4 μm2/μm3 and percolating TPB: 3.2 μm/μm3
The cell was tested in a conventional set-up for 2000 hours under the conditions of 800° C., −1 A/cm2, H2O/H2:90/10 (inlet), 56% H2O conversion, and O2 to the oxygen electrode. The performance as a function of time was measured viavia recording the cell voltage over time and by recording electrochemical impedance spectra during the galvanostatic electrolysis test. The results are shown in
Compared to the standard cell of Example 1, a significant reduced degradation of the cell performance was observed. After 1000 hours of test, the cell voltage increase was ca. 100 mV (cf.
The improved degradation was evaluated to be due to a more stable microstructure of the composite of Layer 2. The microstructure was investigated by scanning electron microscopy before and after test. SEM micrographs of the composite (Layer 2) before test (i.e. initial structures) are seen in
The pore and particle sizes (represented as r50), FWHM and other microstructural characteristics of the composite before test, were also evaluated by 3D image analysis obtained by focused ion beam (FIB) and scanning electron microscopy (SEM). Based on FIB-SEM, a 3D construction of the structure were obtained, and thus a size distribution of the particles. The characteristics of the initial microstructure are summarised in
SEM micrographs of the composite (Layer 2) after 2000 hours of test are shown in
Manufactured as example 1, but Layer 2, in the first step comprises YSZ and 42 vol % NiO. The porosity is 17% after sintering and reduction.
The cell is produced similar to the cell in Example 2 until the fifth step where the (La07.5Sr0.25)0.9Co0.2Fe0.8 O3-δ/CGO10 composite oxygen electrode applied by screen printing.
In the sixed step is the cell sintered at 1050° C. for 12 h.
The seventh step is reduction of the cell, after sealing, at 850° C. for 2 hour in pure H2
The structure of the active fuel electrode has after this procedure the following characteristics: rpore: 0-200 nm; rYSZ, Ni: 0-300 nm and rYSZ˜r,Ni
Manufactured as example 2, but Layer 2, in the first step comprises YSZ and 30 vol % NiO.
The suspension for layer 2 is manufactured by milling the oxides in a vertical bead mill with a bead diameter of 1 mm. The suspension comprises YSZ and 53 vol % NiO. The suspension has a three modal particle size distribution with d10: 0.05 μm; d50: 0.2 μm; d90: 1 μm, after high efficiency bead milling with slurry:ball ratio of 1:12. The green thickness is in the range of 12 μm. The porosity is 12% after sintering and reduction. The cell is produced by tape casting and lamination, similar to the cell in Example 2 until the fifth step where the (La0.75Sr0.25)0.9Co0.2Fe0.8 O3-δ/CGO10 oxygen electrode is applied by screen printing.
In the sixed step is the cell sintered at 1050° C. for 10 h.
The seventh step is reduction of the cell, after sealing, at 850° C. for 2 hour in pure H2
The structure of the active fuel electrode has after this procedure the following characteristics: rpore: 0-200 nm; rYSZ, Ni: 0-300 nm and rYSZ>rNi
The first step comprises individual tape-casting of four layers on a polymeric support. Layer 1: Fuel electrode support, Layer 2: Fuel electrode, Layer 3: Electrolyte, Layer 4: Barrier layer and Layer 5: porous oxygen electrode impregnation layer.
The suspensions for tape-casting are manufactured by means of ball milling oxide powders in ethanol, with poly-pyrolidone (PVP), polyvinyl butral (PVB), polyethylene glycol (PEG) and S-2075 as additives.
Layer 1: The suspension comprises Scandia-yttria stabilized zirconia (ScYSZ) and 53 vol % NiO, the porosity is about 30% after sintering and reduction. The green thickness is in the range of 350 μm
Layer 2: The suspension comprises ScYSZ and 53 vol % NiO, prepared from a calcined powder, manufactured by spray drying of equivalent amount of metal nitrates such as Ni(NO3)6H2O. The suspension has a three modal particle size distribution with d10: 0.05 μm; d50: 0.2 μm; d90: 0.8 μm, after high efficiency ball milling with slurry:ball ratio of 1:12. The green thickness is in the range of 12 μm. The porosity is 22% after sintering and reduction.
Layer 3: The suspension comprises ScYSZ with green thickness is in the range of 8 μm
Layer 4: The suspension comprises CGO10 with green thickness is in the range of 5 μm
Layer 5: The suspension comprises ScYSZ and 70 vol % pore former such as PMMA and Graphite or combinations hereof. The green thickness 30 μm
Layer 3 and 4 are manufactured as a multilayer structure by co-cast of the electrolyte Layer 3 on the top of the barrier layer, Layer 4
The second step comprises assembling of the single tape-cast layers and the co-cast layer 3 and 4 by lamination.
In the third step is the laminated half-cell cut into the desired shape by knife punching, resulting in sintered areas of 12×12 cm2.
In the fourth step is the half-cell sintered for about 12 h at 1270° C., with a heating ramp of 15° C./h to 600° C. and 60° C./h to 1270° C.
The fifth step comprises coating of the porous ScYSZ oxygen electrode impregnation layer from a nitrate solution of ceria and gadolinium nitrate followed by impregnation of the electroactive oxygen electrode (LaxSr1−x)zCoy Fe1−yO3-δ(x=0-1, Y=0-1, Z=1-0.95) with a suitable precursors, such as Metal nitrates. In between every impregnation step is a decomposition and calcination step of the nitrates at 350C. The described infiltration and calcination steps are repeated until a 20 nm layer of CGO and LSCF is obtained. Example three times repetition for CGO and 8 times repetition for the electroactive oxygen electrode.
The sixth step is in-situ sintering of the cell during startup, at about 900° C. for 2 h.
The seventh step is reduction of the cell, at 850° C. in 9% H2 in N2 followed by 1 hour in pure H2
The structure of the active fuel electrode has after this procedure the following characteristics: rpore: 0-700 nm; rYSZ, Ni: 0-300 nm and rYSZ˜r,Ni
Manufactured as Example 2, except for Layer 1 and Layer 2.
Layer 1: The suspension comprises yttria stabilized zirconia with 20% Al2O3 and 53 vol % NiO, the porosity is about 30% after sintering and reduction. The green thickness is in the range of 350 μm
Layer 2: The suspension comprises YSZ and 63 vol % NiO. The suspension has a hexa modal particle size distribution with d10: 0.05 μm; d50: 0.2 μm; d90: 1 μm, after high efficiency ball milling with slurry:ball ratio of 1:12. The green thickness is in the range of 12 μm. The porosity is 26% after sintering and reduction.
The cell is produced similar to the cell in Example 2 until the fifth step where the (La0.75Sr0.25)0.95 MnO3 oxygen electrode is applied by screen printing.
In the sixed step is the cell sintered at 1290° C. for 12 h.
The seventh step is reduction of the cell, after sealing, at 1000° C. for 2 hour in pure H2
The structure of the active fuel electrode has after this procedure the following characteristics: rpore: 0-200 nm; rYSZ,Ni: 0-300 nm and rYSZ˜r,Ni
The cell is manufactured as Example 2 step 1 and 2 and further added a fifth layer (backbone for oxygen electrode): the fifth slurry is consisting CGO10, 40 vol % Graphite (4-6 μm) and 40 vol % PMMA (2 μm) and tape-cast so the dry thickness is 30 μm, the fifth layer is laminated on the fourth layer of the cell.
In the third step is the laminated half-cell cut into the desired shape by knife punching, resulting in sintered areas of 12×12 cm2.
In the fourth step is the half-cell sintered for about 12h at 1300° C., with a heating ramp of 15° C./h to 600° C. and 60° C./h to 1300° C.
The fifth step comprises infiltration of LSC in the porous oxygen electrode back-bone from respective nitrate salts, followed by a decomposition step of the nitrates at 350° C.
The seventh step is reduction of the cell, after sealing, at 1000° C. for 2 hour in pure H2
The structure of the active fuel electrode has after this procedure the following characteristics: rpore: 0-200 nm; rYSZ Ni: 0-300 nm and rYSZ˜r,Ni
Manufactured as example 6.
In the eighth step is the fuel electrode infiltrated in the stack with CGO10 and Cu2O from the respective nitrates.
The ninth step is a second reduction of the cell, at 800° C. for 2 hour in pure H2.
The same manufacturing procedures as described in Examples 1 and 2 were applied to produce cells with Ni-YSZ composites with Ni:YSZ ratio of 50:50, and respectively a standard microstructure, and a microstructure according to an embodiment of the invention with a microstructure optimized for stability.
The cells with 50 vol % Ni and 50 vol % YSZ with respectively standard and optimized fuel electrode structure were produced and tested under the same conditions as for the cells for which results are shown in
The results for the cells with at 50/50 volume ratios of Ni/YSZ are shown in
Compared to the data from
Thus, improved stability, or reduced degradation, may be obtained for Ni/YSZ composites with different Ni/YSZ ratios, and where further the microstructure is optimized for stability according to an embodiment of the invention.
Items
The invention can be further described by the items listed below.
Item 1
A composite for an electrode, comprising a three-dimensional network of dispersed metal particles, stabilised zirconia particles and pores,
wherein the size of the pores is smaller than the size of the metal particles, and
wherein the size of the metal particles is similar or smaller than the size of the stabilised zirconia particles.
Item 1A
A composite for an electrode, comprising a three-dimensional network of dispersed metal particles, stabilised zirconia particles and pores,
wherein the size of the pores is smaller than the size of the metal particles,
wherein the size of the metal particles is essentially equal to or smaller than the size of the stabilised zirconia particles,
wherein the porosity is below 33, 30, or 29 vol %, more preferably below 26 or 24 vol %, and most preferably below 23, 22, 21, 18, 15, or 13 vol %, and/or wherein the pores are essentially exclusively generated from the volume created by reducing a corresponding metal oxide to the metal particles.
Item 2
The composite according to item 1 or 1A, wherein the characteristic size of the pores is smaller than the characteristic size of the metal particles, and
wherein the characteristic size of the metal particles is similar or smaller than the characteristic size of the stabilised zirconia particles.
Item 3
The composite according any of the preceding items, wherein the porosity is generated by reducing a corresponding oxide of the metal to the metallic form.
Item 4
The composite according to any of the items, wherein the porosity is below 33, 30, or 29 vol %, more preferably below 26 or 24 vol %, and most preferably below 23, 22, 21, 18, 15, or 13 vol %.
Item 5
The composite according to any of the items, wherein the maximum characteristic pore diameter is below 700 nm, more preferably below 650 nm or below 600 nm, and most preferably below 550 nm, such as 500 nm.
Item 6
The composite according to any of the preceding items, wherein the median characteristic pore diameter is between 100-630 nm, more preferably between 200-500 nm, and most preferably between 300-400 nm or 350-400 nm.
Item 7
The composite according to any of the preceding items, wherein the maximum characteristic metal particle diameter is below 1100 nm, more preferably below 900 nm or below 800 nm, and most preferably equal to or below 700 nm.
Item 8
The composite according to any of the preceding items, wherein the median characteristic metal diameter is between 200-900 nm, more preferably between 300-800 nm or 400-700 nm, and most preferably between 500-650 nm or 550-600 nm.
Item 9
The composite according to any of the preceding items, wherein the network of metal particles and stabilised zirconica particles are dispersed such that the interface area between the stabilised zirconia and metal is above 0.8 μm2/μm3, more preferably above 1 or 1.2 μm2/μm3, and most preferably equal to or above 1.4 μm2/μm3.
Item 10
The composite according to any of the preceding items, wherein the network of metal particles, stabilised zirconia particles and pores are dispersed to form a percolating TPB, wherein the percolating TPB has a length above 1.5 μm/μm3, more preferably above 2 or 2.5 μm/μm3, and most preferably equal to or above 3.0 or 3.2 μm/μm3.
Item 10A
The composite according to any of the preceding items, wherein the network of metal particles and stabilised zirconica particles has a volume specific interface area between the stabilised zirconia and metal above 0.8 μm2/μm3, more preferably above 1 or 1.2 μm2/μm3, and most preferably equal to or above 1.4 μm2/μm3, and/or wherein the percolating TPB defined by the network of metal particles, stabilised zirconia particles and pores has a volume specific length above 1.5 μm/μm3, more preferably above 2 or 2.5 μm/μm3, and most preferably equal to or above 3.0 or 3.2 μm/μm3.
Item 11
The composite according to any of the preceding items, wherein the metal is selected from the group consisting of Ni, Cu, Cr, Fe, and any combination thereof.
Item 12
The composite according to any of the preceding items, wherein the corresponding oxide of the metal is selected from the group consisting of NiO, CuO, Cu2O, CrO2, Cr2O3, FeO, Fe2O3, Fe3O4, and any combination thereof.
Item 13
The composite according to any of the preceding items, wherein the stabilised zirconia is selected from the group consisting of yttria-stabilised zirconia, scandia-stabilised zirconia, yttria- and scandia co-stabilised zirconia, magnesia-stabilised zirconia, calcia-stabilised zirconia, ceria-stabilised zirconia, or a combination thereof.
Item 14
The composite according to any of the preceding items, wherein the volume fraction of metal of the solid phase is 70, 60, 50, 45, 40, 30, or 20 vol %, most preferably essentially 30 vol %.
Item 15
The composite according to any of the preceding items, wherein the volume ratio between metal and stabilised zirconia is 70:30, 60:40, 50:50, 45:55, 40:60, 30:70, or 20:80, most preferably 40:60.
Item 16
An electrode for an electrochemical cell, comprising the composite according to any of items 1-15.
Item 17
A solid state electrochemical cell, comprising the composite according to any of items 1-15.
Item 18
The solid state electrochemical cell according to item 17, wherein the composite is a fuel electrode.
Item 19
Use of the electrochemical cell according to any of items 17-18 as a solid oxide fuel cell, and/or a solid oxide electrolyser.
Item 20
A composite precursor for an electrode, comprising a three-dimensional network of dispersed particles of a metal oxide and a stabilised zirconia,
wherein the precursor characteristic size is d10 between 0.03-0.07 μm, d50 between 0.08-0.4 μm, d90 between 0.4-1.2 μm, more preferably d10 between 0.04-0.06 μm, d50 between 0.1-0.3 μm, d90 between 0.9-1.1 μm, and most preferably d10=0.05 μm, d50=0.2 μm, and d90=1 μm.
Item 21
The composite precursor according to items 20, wherein the porosity is below 5 vol %, more preferably below 3 vol %, and most preferably below 1 vol %.
Item 22
A method of manufacturing a composite for an electrode, comprising the steps of:
Item 22A
A method of manufacturing a composite, comprising the steps of:
Item 23
The method according to item 22 or 22A, wherein one or more ceramic processing means in step (b) is applied to the metal powder prior to the stabilised zirconia powder.
Item 24
The method according to items 22-23, wherein the metal powder is milled before the stabilised zirconia powder.
Item 25
A composite for an electrode obtained by the method of items 22-24.
Item 25A
The composite for an electrode according to item 16 obtained by the method of items 22-24.
[1] U.S. Pat. No. 7,029,777.
[2] Kishimoto et al., Fuel Cells, 13 (2013) 476-486.
[3] Lay-Grindler et al., Journal of Power Sources, 269 (2014) 927-936
[4] K. Thydén, et al., Solid State Ionics, 178 (2008) 1984-1989.
Number | Date | Country | Kind |
---|---|---|---|
15181381.3 | Aug 2015 | EP | regional |
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/EP2016/069580 | 8/18/2016 | WO | 00 |