Method for making ceramic wall-flow filter substrates supporting porous on-wall coatings - Patent Application 20070122997

Abstract

1. A method for making a ceramic honeycomb wall-flow filter substrate supporting an on-wall coating, wherein a section of the on-wall coated filter substrate analyzed by mercury porosimetry has a pore volume of 0.05 to 0.5 μm diameter as a percentage of the mercury intrusion volume of at least 15.0% of the section, and (i) an equivalent spherical diameter D(v,0.5) of 0.5 to 1 μm, which is equal to 10 to 35 wt. % of water, carboxylic acid, inorganic oxide particles, or a particulate insoluble cellulose pore former of heterogeneous morphology, with D(v,0.9) of 8 to 20 μm, relative to 100 wt. % of inorganic oxide particles. 4 μm and having a modal aspect ratio (width / length) of 0.3 to 0.9; (ii) coating the slurry onto at least first channels of a substrate, the substrate having an average pore size (D50) of 6 to 15 μm and a porosity of less than 60% before any coating; and (iii) drying and calcining the slurry-coated substrate, wherein the product of step (iii) has a porosity of 0.07 to 0.4 g / in based on the substrate weight before step (ii). 3 A method wherein sufficient slurry is coated onto a substrate to have a coating loading of (4.3-24.4 g / L) and an average on-wall coating thickness of the coated substrate is 5-70 μm.

Description

[Technical Field]

[0001] The present invention relates to a method of making a ceramic honeycomb wall-flow filter substrate bearing a porous on-wall coating for use in treating exhaust gases from a diesel engine or internal combustion engine configured to operate on a gaseous fuel having a major fuel mass of hydrogen (H2). The present invention also relates to a ceramic honeycomb wall-flow filter substrate bearing a porous on-wall coating obtained or obtainable by the method of the present invention, as well as an on-wall coated ceramic honeycomb wall-flow filter substrate for treating exhaust gases containing particulate matter emitted from an internal combustion engine, and an exhaust system for an internal combustion engine comprising the ceramic honeycomb wall-flow filter substrate. [Background technology]

[0002] As emissions regulations for vehicles continue to tighten around the world, particularly those with compression ignition engines in their powertrains, there is an increasing need to reduce emissions in both size and number of particulate matter throughout the vehicle's duty cycle (i.e., between key-on and key-off), including immediately after key-on / cold start.

[0003] This can be seen, for example, in the new exhaust gas regulations being introduced in Europe, where, from September 1, 2019, all new passenger cars being introduced to the European market for the first time are being tested according to new regulations called "Real Driving Emissions" (RDE) tests. RDE tests measure NOx, a major cause of air pollution. x This is a more realistic on-road test of emissions of diesel and other particulate matter under real driving conditions. The test is carried out with a portable emission measuring system (PEMS) attached to the vehicle while it is driven on the road under real-world conditions.

[0004] Additionally, a new laboratory-based test called the World Harmonised Light Vehicles Test Procedure (WLTP) was implemented on September 1, 2017, to replace the outdated New European Driving Cycle (NEDC) as the European vehicle homologation procedure. WLTP is a globally harmonised standard for determining the pollutant CO2 emissions and fuel economy of conventional and hybrid vehicles, and was developed by the United Nations Economic Commission for Europe (UNECE).

[0005] Euro 6d-TEMP and the third package of RDE tests for Euro 6d vehicles (Regulation (EC) 2017 / 1154) introduced a particle number (PN) measurement protocol and an associated conformance factor (relative to the WLTP limit) of 1.5 (including a margin of 0.5) from September 1, 2017 for new vehicle models (or "type approval") and until September 1, 2019 for existing models. Furthermore, Euro 6e (Regulation (EC) 2017 / 1151), which came into effect in September 2023, introduced revisions to the RDE test procedure, whereby the previously defined "conformance factor" is redefined as a "PEMS error margin," including a reduced PN margin of 0.34, corresponding to a conformance factor of 1.34, down from 1.5.

[0006] Additionally, the European Parliament, Council, and Commission are negotiating the long-anticipated Euro 7 proposal for new emissions standards for road vehicles, including both light vehicles (cars and vans) and heavy vehicles (trucks and buses) sold in the EU. The proposal would combine the successor regulations to Euro 6 (Regulation (EC) 715 / 2007) and Euro VI (Regulation (EC) 595 / 2009) into a single regulation called Euro 7 for both light and heavy vehicles (no longer Euro VII for heavy vehicles). According to a press release from the European Parliament and the Council of December 18, 2023, as of the filing date, the EU 7 proposal being advanced by the European Parliament has been provisionally agreed upon and includes the maintenance of the current Euro 6e exhaust emission limits for cars and vans (light vehicles) but includes limits on emissions of solid particles with diameters starting at 10 nm (PM10), instead of the current 23 nm for Euro 6. For heavy vehicles, in addition to limits on solid particle emissions from PM10, the Euro 7 proposal includes a NOx limit set at 200 mg / kWh (up from 460 mg / kWh) over the World Harmonised Transient Cycle, and limits for ammonia and N2O are introduced.

[0007] The EU Commission's proposed date for the Euro 7 regulation to come into force was July 1, 2025 for new light vehicles and July 1, 2027 for new heavy vehicles. A provisional agreement reached in tripartite talks between the EU Parliament, Council, and Commission replaces these dates, returning to a two-date approach for new vehicle types first, after which all new vehicles are registered. These dates are stated as the number of months after the regulation comes into force. For light vehicles, the date has been provisionally agreed to be +30 months for "new types" and +42 months for "all types." For heavy vehicles, the date has been provisionally agreed to be +48 months for "new types" and +60 months for "all types."

[0008] As of the filing date of this application, adoption of the Euro 7 regulation is subject to final formal approval and adoption by the European Parliament and its member states.

[0009] Methods for meeting current and future particulate matter emission standards generally involve the use of filter substrates. Filter substrates are known that include an on-wall washcoat containing pores partially derived from the combustion of a pore-forming agent. For example, European Patent No. 0736503(A1) discloses an exhaust gas purification filter produced by coating the surface of a matrix (e.g., cordierite or silicon carbide) having a plurality of pores with a mixture of ceramic powder (e.g., zeolite, or preferably alumina) and an interconnected pore-forming material, and then firing the coating to sinter the ceramic powder and form a coating layer, while burning off the interconnected pore-forming material to form interconnected pores in the coating layer. The exhaust gas purification filter comprises a matrix having a large number of pores, and a coating layer supporting a catalyst for purifying exhaust gas is provided on the surface of the matrix. The coating layer has interconnected pores that connect from its surface to the matrix. Preferably, the coating layer is formed not only on the surface of the matrix, but also on the inner surfaces of the pores inside the matrix.

[0010] According to claim 5 of European Patent No. 0736503(A1), the size of the interconnected pore-forming material is preferably the same as or smaller than the pores of the matrix. This allows the interconnected pore-forming material to smoothly penetrate the pores and form interconnected pores within the coating layer that coats the inner surfaces of the pores. Alternatively, the interconnected pore-forming material may be a combustible material having a size equal to or greater than the thickness of the coating layer, such as carbon, resin, or wax. The combustible material may be in the form of, for example, whiskers, needles (such as fibers), spheres, or pillars. However, if the size of the interconnected pore-forming material is larger than the size of the pores of the matrix, the interconnected pore-forming material will not enter the pores, and ceramic powder may clog the pores, resulting in increased pressure loss in the filter. The average pore size of the interconnected pores is preferably 10 to 60 μm.

[0011] Example 1, referring to FIG. 1, contains the only disclosure of the average pore size of the matrix in EP 0736503(A1). Example 1 shows the average pore size of a cordierite honeycomb (all-flow) filter (Φ 140 mm × length 130 mm, 150 cells / in 2The paper discloses a process for manufacturing an exhaust gas purification filter using a ceramic powder (mesh, wall thickness 0.45 mm) as a matrix. A slurry of 95% by weight of activated alumina and 5% by weight of alumina solution was used as the ceramic powder, and the pH was adjusted to 1-3 by adding dilute nitric acid. The average particle size of the activated alumina used was 5-10 μm. To 100% by weight of the ceramic powder, 1.5% by weight of a gas-generating substance and 25% by weight of carbon powder were added as interconnected pore-forming materials, and distilled water was then added to obtain a slurry. The particle size of the carbon powder was 10-80 μm, which is larger than the thickness of the coating layer to be formed (5-50 μm). The gas-generating substance was a gas-enclosing material prepared by encapsulating butane gas in a thermoplastic resin and then granulating it. The matrix was immersed in the slurry while stirring and then removed. The excess slurry was then removed by air blowing, the mixture was dried at 120°C for 2 hours, and then heated at 700°C for 2 hours to sinter the ceramic powder. During heating, the carbon powder and the thermoplastic resin gas-generating substance were burned away, and the butane gas enclosed therein escaped from the coating layer, forming interconnected pores that connected the inside of the coating layer to the surface.

[0012] In the final product of Example 1, the interconnected pores are distributed within a pore diameter range D of 20 to 40 μm (average 30 μm) as measured by mercury intrusion porosimetry. The average pore diameter of the matrix 5 is 20 to 40 μm. The honeycomb filter matrix was coated with a coating layer of 40 g per liter of apparent volume, based on the weight difference between the matrix before and after coating. The thickness of the coating layer 1 is 5 to 50 μm. A catalyst such as Pt (platinum) or Rh (rhodium) is supported on the alumina.

[0013] Applicant's International Publication No. 2015 / 082892 discloses a catalyzed honeycomb wall-flow filter for treating exhaust gases containing particulate matter emitted from an internal combustion engine, the filter including a honeycomb structure having a first end and a second end, and including an array of interconnecting porous walls defining an array of longitudinally extending first and second channels, the first channels being bounded on their sides by second channels and having a larger hydraulic diameter than the second channels, the first channels being end-plugged at the first end of the honeycomb structure, and the second channels being end-plugged at the second end of the honeycomb structure, and the channel wall surfaces of the first channels including an on-wall catalytic washcoat. The disclosure also relates to exhaust systems including the catalyzed filter, and methods of making the catalyzed filter.

[0014] This disclosure explains that the surface porosity of a washcoat can be increased by including voids therein. "Voids" in a washcoat layer refer to the presence of spaces within the layer defined by the solid washcoat material. The voids may include any voids, microscopic pores, tunnels (cylindrical, prismatic), slits, etc., and can be introduced by including in the washcoat composition for coating on the filter substrate a material that burns during firing of the coated filter substrate (e.g., shredded cotton, plastic beads, or materials that create pores by the formation of gases upon decomposition or combustion, such as acetic acid, starch, or other organics). This disclosure further explains that when the method of the present invention involves applying a washcoat to a partially plugged honeycomb structure, solid pore-forming agents such as polymer beads and shredded cotton can be filtered out within the filter along the axial length of the wall, resulting in the pore-forming agent collecting at one end of the axial washcoat, in which case a liquid pore-forming agent such as citric acid is preferred. The washcoat may have an average porosity of 5 to 80%, while the average diameter of the pores may be 0.2 to 500 μm, for example 10 to 250 μm.

[0015] The arrangement of channels (or cells) at the inlet end face of the filter having a larger hydraulic diameter than the hydraulic diameter of the channels (or cells) at the outlet end face of the filter is known as an asymmetric channel design.

[0016] European Patent No. 2,158,956 (A1) discloses a second honeycomb filter comprising partition walls having a porous partition wall base material that separates and forms a plurality of cells that function as fluid flow paths, and surface layers provided only on the inlet side or on both the inlet and outlet sides of the partition wall base material, in which predetermined cells have one end open on one side and the other end blocked on the other side, and the remaining cells have one end blocked on one side and the other end open on the other side, and the surface layer provided on at least the inlet side of the partition wall base material is made of a catalyst layer containing at least one material selected from the group of materials consisting of alumina, zirconia, titania, zeolite, and ceria, and the second honeycomb filter satisfies the following conditions (I) to (V): (I) the peak pore diameter of the surface layer is equal to or smaller than the average pore diameter of the partition wall base material, and the porosity of the surface layer is greater than the porosity of the partition wall base material. (II) The peak pore diameter of the surface layer is 0.3 μm or more and less than 20 μm, and the porosity is 60% or more and less than 95% (measured by mercury porosimetry). (III) The thickness L1 of the surface layer is 0.5% or more and less than 30% of the thickness L2 of the partition wall. (IV) The mass of the surface layer per filtration area is 0.01 mg / cm. 2 More than 6mg / cm 2 Less than (0.06 to 38.71 mg / in 2 (equivalent to). and (V) the average pore size of the partition wall base material is 10 μm or more and less than 60 μm, and the porosity is 40% or more and less than 65%. In the disclosed second honeycomb filter, the surface layer composed of the catalyst layer preferably contains fine particles of one or both of platinum and palladium. The second honeycomb filter may have an asymmetric design (see, for example, claim 14).

[0017] European Patent No. 2158956 (A1) also discloses a method for producing a second honeycomb filter, the method including the steps of: pre-fabricating a honeycomb structure as a fired article, the honeycomb structure having a porous partition wall base material that separates and forms a plurality of cells that function as fluid flow paths; mixing at least one material selected from the group of materials consisting of alumina, zirconia, titania, zeolite, and ceria; a pore former that is combustible and dispersible at a temperature of 800°C or less, has an average particle size of less than 10 μm, and is contained in a ratio of 20% by mass to less than 90% by mass of the solid content of the material; and water to obtain a slurry; depositing the slurry on only the inlet side or on both the inlet and outlet sides of the formed honeycomb article; and then drying or drying and firing the slurry to provide a surface layer on only the inlet side or on both the inlet and outlet sides of the formed honeycomb article.

[0018] EP 2158956 A1 discloses a single embodiment (Example 44) containing a platinum group metal-bearing catalyst layer, in which a mixture of gamma alumina and ceria (CeO2) in an unspecified weight ratio with an average particle size of 5 μm is immersed in a platinum (Pt)-containing solution. Resin particles with an average particle size of 3 μm are added as a pore-forming agent, and acetic acid and water are further added to obtain a slurry. The amount of pore-forming agent contained in the slurry is 71% by mass of the solid content of the material. This slurry is then poured into a plugged honeycomb structure (average pore size 25 μm, diameter 144 mm (5.7 in), length 152 mm (6.0 in), and 46.5 cells cm). -2 The slurry was deposited on the inlet side of the partition wall base material by applying high vacuum to the outlet end face of the partition wall (300 cells per inch, 300 μm (11.8 mils (1 / 1000 of an inch))) to form a surface layer. The structure was then dried and fired at 600°C for 3 hours to obtain a honeycomb filter with a surface layer (average thickness 30 μm) composed of a catalyst layer on the plugged honeycomb structure. The amount of oxides (gamma-Al2O3 and CeO2) in the surface layer was 100% of the total amount of oxides per cm of partition wall filtration area. 2The amount of Pt is 0.75 mg per kiloliter of honeycomb structure volume, and the amount of Pt is 2 g per kiloliter of honeycomb structure volume, which the applicant calculates to be 0.13 g / in -3 (7.9g / L), and 0.0567g / ft -3 This corresponds to a washcoat loading of 0.002 g / L. The peak pore size of the catalyst layer was 3 μm, and the porosity was 82%. Applicant understands that the "Material solids ratio (mass %)" of the "Pore-forming agent contained in the slurry for the surface layer" in Table 2 means that a porosity of 82% was obtained in the catalyst layer by formulating the slurry for the surface layer so that 71 mass % of the material solids was the pore-forming agent.

[0019] European Patent No. 1961481(A2) discloses a catalytic converter comprising a three-dimensional structural support, optionally including a wall-flow filter as shown in FIG. 1 having a plurality of cells separated by porous cell walls, wherein the pore size of the cell walls is 10 μm to 50 μm, the porosity of the cell walls is 40% by volume or more, and a catalyst-coated layer containing a catalytic component. The catalyst-coated layer is coated on the surface of the cell walls of the three-dimensional structural support. 50% by mass or more of the total amount of the catalytic component loaded on the three-dimensional structural support is present in the region from the surface of the cell walls of the three-dimensional structural support to the surface of the catalyst-coated layer, i.e., up to 50% by mass of the total amount of the catalytic component loaded on the three-dimensional structural support is present within the porous cell walls. In Examples 4 and 12, a catalytic converter having a pore density of 50 grams per liter (0.82 g / in ) derived from a slurry containing platinum-loaded aluminum oxide powder and activated carbon or polystyrene beads as a pore-forming agent is used. 3 The catalyst-coated layer, which supports a 144 mm diameter and 152 mm length, has a cell density of 400 / in 2 (1cm 2 The silicon carbide honeycomb substrate has an average pore size of 10-20 μm within the cell walls and a porosity of 50-60% by volume (62 cells per substrate). No details are given regarding the particle size of the aluminum oxide powder, activated carbon, or polystyrene beads.

[0020] WO 2008 / 153828(A2) discloses a method for preparing a porous inorganic coating on a porous support using a specific organic pore-forming agent, and a porous support coated with the porous inorganic coating. Inorganic particle components such as alpha alumina and gamma alumina are disclosed. The coating composition may contain 0.1 to 50 wt. % inorganic particles. While diesel particulate filter products are mentioned as applications for the present disclosure, details of suitable such filters or their physical parameters are not disclosed or exemplified. Furthermore, none of the exemplified inorganic coatings contain platinum group metals. Example 3 describes the deposition of an alpha alumina membrane (a blend of alpha alumina particles with average particle sizes of 0.8 to 1.1 μm and 0.3 to 0.4 μm) using a protein derived from skim milk with an average particle size of 0.4 μm as the pore-forming agent.

[0021] European Patent No. 2502662(A1) discloses a honeycomb filter including a honeycomb base material comprising a porous partition wall parent material that divides and forms a plurality of cells through which a fluid passes; blocked portions disposed in the pre-opening regions of certain cells in the end surface on the fluid inflow side and in the pre-opening regions of the remaining cells in the end surface on the fluid outflow side; and a porous collection layer disposed on the surface of the partition wall parent material within at least the remaining cells. The collection layer has a configuration in which a plurality of particles are intertwined or entangled with each other, and the collection layer includes a plurality of plate-like particles having a predetermined aspect ratio. The plurality of particles have an average major axis of 0.2 μm or more but less than 10 μm, an average ratio of the major axis to the minor axis (major axis / minor axis) of less than 3, and an average ratio of the major axis to the thickness (major axis / thickness) of 3 or more.

[0022] Example 14 of European Patent No. 2502662(A1) is the only example in which the collection layer contains a platinum group metal (platinum only, 0.5% by mass). In this example, the coating composition is obtained by coating alumina particles with an average major axis of 5 μm, an average major axis / minor axis ratio of 2, and an average major axis / thickness ratio of 20 with a slurry obtained by mixing 60 g of ceria sol containing 15% by mass of CeO, 20 g of alumina sol containing 20% ​​by mass of AlO, and 4 g of a platinum chloride aqueous solution containing 10% by mass of Pt. The collection layer contains composite plate-like particles. The reason for not using a pore-forming agent is understood to be that, during application of the slurry to the partition wall parent material, "multiple particles intertwine or interlock with each other without excessive alignment along the surface of the partition wall parent material." As a result, the high open area ratio described above is achieved, thereby suppressing "an increase in the initial pressure loss of the honeycomb filter." That is, it is essential that the aspect ratio of the composite plate-like particles is relatively high in order to achieve the claimed suppression of an increase in initial pressure loss.

[0023] International Publication No. 2017 / 209083(A1) discloses an exhaust gas purification filter that is said to be able to suppress the increase in pressure loss associated with the formation of a catalyst layer while simultaneously promoting the combustion of particulate matter. The exhaust gas purification filter according to the present invention includes a honeycomb substrate and a catalyst layer disposed on the honeycomb substrate. The catalyst layer contains a carrier and a metal catalyst. In an electron microscope image of the cross section of the catalyst layer, large pores with a circle-equivalent diameter of greater than 5 μm account for 45% or more of the area of ​​the catalyst layer, assuming the area of ​​the catalyst layer to be 100%. Examples disclose several platinum- and palladium-containing catalyst layers in which a solution of platinum group metal salts (platinum and palladium) is impregnated into "acicular Al2O3 powder" having an average fiber diameter of 0.5 μm and an average fiber length of 6 μm, or "acicular TiO2 powder" having an average fiber diameter of 0.2 μm and an average fiber length of 2 μm. However, examples are also disclosed of Al2O3, TiO2, or CeZr particulate powders, each milled by wet milling to an average particle size of 2 μm. In the examples, large pores are introduced by pore formers, which are melamine resin beads of various sizes (average particle sizes of 2 μm, 5 μm, and 10 μm), starch, or polymethyl methacrylate resin. Citric acid (a blowing agent) was used as a comparative "pore former." After calcination, during which the pore former was removed by combustion, the catalyst layer had a loading of 15 g / L (0.246 g / in 3 (corresponding to Example 19). Example 19 is a honeycomb substrate coated with a three-way catalyst composition containing multiple oxides of rhodium and ceria-zirconia, which is also disclosed. Therefore, it can be understood that Examples 1 to 18 are directed to products for treating diesel particulate matter.

[0024] However, in all of Examples 1 to 18, the thickness of the partition walls (12 thousandths of an inch (mils), which corresponds to 0.3048 mm) and the cell density (300 cells per square inch, or 46.5 cells / cm) were 2 ), no specific details of the silicon carbide honeycomb substrate used are provided. That is, there is no correlation between the particle size of the catalyst layer, the pore size derived from the pore former in the particulate layer, and the pore size of the honeycomb substrate.

[0025] European Patent No. 3623048 (A1) discloses an exhaust gas purification catalyst device including a honeycomb base material and an inlet-side coating layer, in which the honeycomb base material includes a plurality of cells separated by porous partition walls, the plurality of cells including inlet-side cells and outlet-side cells, and is configured so that exhaust gas that flows through the partition walls to the inlet-side cells is discharged from the outlet-side cells, the inlet-side coating layer is present on the surface side of the partition walls of the inlet-side cells, and the proportion of through pores of 4 to 9 μm in the through pore size distribution of the partition walls of the coated honeycomb base material measured according to the bubble point method using a perm porosimeter is at least 80 vol%, and the peak pore size measured using a mercury porosimeter is at least 3.0 μm larger than the peak through pore size measured according to the bubble point method also using a perm porosimeter.

[0026] An exhaust gas purification catalyst device can be prepared by coating the inlet cells of a honeycomb substrate with a slurry coating layer containing inorganic oxide particles and a pore-forming material, and then firing the slurry-coated honeycomb substrate. The pore-forming material may be organic polymer particles, such as (meth)acrylic resin particles, styrene-(meth)acrylic resin particles, polyurethane resin particles, maleic acid resin particles, styrene-maleic acid resin particles, alkyd resin particles, rosin-modified phenolic resin particles, and ketone resin particles, with an average particle size of 10 to 500 nm (0.01 to 0.5 μm). Generally, the median diameter of the inorganic oxide particles is, for example, 0.1 μm or more, 0.3 μm or more, 0.5 μm or more, 0.8 μm or more, or 1.0 μm or more, and, for example, 10 μm or less, 5.0 μm or less, 3.0 μm or less, 2.0 μm or less, 1.5 μm or less, or 1.0 μm or less.

[0027] In the examples of EP 3623048 A1, a cylindrical SiC diesel particulate filter substrate with a honeycomb structure having a diameter of 160 mm, a length of 135 mm (i.e., a volume of 2.71 liters), an average pore size of 12 μm (nominal), and a porosity of 42% by volume was filled with 100 g of Pd / Pt-alumina powder (36.9 g / L or 0.604 g / in 3 ), and coated with a slurry containing 50 g of ceria-zirconia composite oxide powder. These powders were pulverized to obtain a mixture having an average particle size of 1 μm, and then styrene-acrylic resin particles having an average particle size of 100 nm (0.1 μm) were added in an amount of 30 wt % based on the solid content in the slurry.

[0028] WO 2021 / 126685(A1) discloses a catalytic washcoat composition comprising a slurry containing at least one platinum group metal and / or at least one non-platinum group metal supported on at least one support, and at least one pore-forming agent having a particle size in the range of 100 nm to 5.0 μm, where the pore-forming agent is selected from carbon nanotubes, carbon nanofibers, activated carbon, resin, cellulose powder, and polymer spheres. A catalytic article for capturing particulate matter in the size range of 1.0 nm to 100 μm is also disclosed, the article comprising a catalytic washcoat deposited on a substrate and calcined to form 50% to 100% of pores having a pore size in the range of 100 nm to 5.0 μm. The only example is a catalytic composite / article coated with a washcoat containing both rhodium and an oxygen storage component containing ceria (CeO). That is, the example article is for use as a gasoline soot filter.

[0029] European Patent No. 3263214(A1) discloses a method for producing a catalyst for purifying exhaust gases, which includes the steps of: obtaining a catalyst slurry by mixing metal oxide particles having a cumulative 50% diameter value in the range of 3 to 10 μm in a cumulative particle size distribution based on volume measured by laser diffraction, a precious metal raw material, and a fibrous organic material having an average fiber diameter in the range of 1.7 to 8.0 μm and an average aspect ratio in the range of 9 to 40, so that the amount of fibrous organic material is in the range of 0.5 to 9.0 parts by mass per 100 parts by mass of the metal oxide particles; applying the catalyst slurry to the surface of a substrate so that the average thickness of the catalyst coating layer after calcination is in the range of 25 to 160 μm, thereby forming a catalyst slurry layer; and calcining the catalyst slurry layer to remove at least a portion of the fibrous organic material in the catalyst slurry layer, thereby obtaining a catalyst for purifying exhaust gases. According to the present disclosure, the "aspect ratio of the pores" in the obtained catalyst coating layer can be determined by analyzing three-dimensional information about the pores from a cross-sectional image of a cross section of the catalyst coating layer perpendicular to the flow direction of the exhaust gas in the substrate using a FIB-SEM (Focused Ion Beam-Scanning Electron Microscope).

[0030] In addition, European Patent No. 3263214(A1) explains that when the average aspect ratio is less than 9, the pore connectivity is insufficient, resulting in insufficient gas diffusion, and when the amount of fibrous organic material mixed exceeds 9.0 parts by mass per 100 parts by mass of metal oxide particles, the thickness of the catalyst coating layer increases, resulting in a large pressure drop and a deterioration in fuel efficiency. Furthermore, in the final catalyst product, the coating amount of the catalyst coating layer in the catalyst for purifying exhaust gas is preferably in the range of 50 to 300 g / L, which is 0.819 to 4.916 g / in 3This corresponds to the figure. If the coating amount is less than 50 g / L, the catalyst particles will not exhibit sufficient catalytic activity, and the resulting catalytic performance, such as NOx purification performance, tends to be insufficient. In this example, NOx purification performance ("Test to measure NOx removal rate") was performed under gasoline exhaust gas conditions, i.e., "Using an in-line 4-cylinder 2.4 L engine, air / fuel ratio feedback control was performed with a target of 14.1 to 15.1, and the NOx removal rate was calculated from the average amount of NOx emitted when the A / F was switched." Here, the engine operating conditions and piping settings were adjusted so that the intake air volume was 40 (g / sec) and the temperature of the gas flowing into the catalyst was 750°C.

[0031] For example, International Publication No. WO 2006 / 031600 discloses that coating the inlet channels of a wall-flow filter substrate with a catalyst washcoat from one substrate end to less than the entire axial length of the substrate, and coating the outlet channels with a catalyst washcoat from the other substrate end to the remainder of the axial length of the substrate, can beneficially reduce backpressure across the filter substrate compared to a filter substrate that is uniformly coated from the inlet channels to the entire axial length of the filter substrate. However, those skilled in the art will recognize that any gaps in the axial length coating can cause filtration failure. This is because exhaust gases, which are fluids such as water, pass through the wall-flow filter wall, which offers the least resistance to gas flow. This means that exhaust gases preferentially pass through axial gaps in the coating. For this reason, the actual production of catalyst-coated filters generally targets a small overlap between the inlet and outlet channel coatings.

[0032] It is common knowledge that the population of inorganic oxide particles for use in washcoats in the field of the present invention essentially has a particle size distribution that includes D(v, 0.5) (also called "D50") and D(v, 0.9) (also called "D90"). See, e.g., "Catalytic Air Pollution Control - Commercial Technology," 3 rd See chapter subheading 3.2.4 "Particle Size Distribution of the Carrier" in "The Inorganic Oxide Particles of the Present Invention," International Journal of Chemical Engineering, Vol. 1, No. 1, pp. 111-115, 2009. This particle size distribution can be adjusted to some extent by mechanical milling, e.g., using a ball mill, or by jet milling. See M. Blazek et al., Chemical Engineering Journal, 409 (2021) 128057. It can also be seen that the D50 and D90 of a population of inorganic oxide particles for use in a washcoat can be adjusted, for example, by combining two populations with different particle size distributions (see, e.g., U.S. Pat. No. 5,496,788, Example 1 and Figure 2). The difference between D(v,0.5) and D(v,0.9) can partially influence the position of the washcoat (e.g., on-wall or in-wall) when coated on a wall-flow filter substrate.

[0033] In the above-discussed International Publication No. 2015 / 082892, the applicant disclosed that the difference in hydraulic diameter between the inlet and outlet face channels in an asymmetric channel design can be used to produce an on-wall coating on the inlet channel having a larger hydraulic diameter than the outlet face channel. However, in more recent asymmetric designs from filter manufacturer Ibiden, as disclosed, for example, in Society of Automotive Engineers (SAE) Technical Paper 2014-01-1512 and European Patent No. 2862610(A1), by appropriately plugging the ends of an unplugged substrate containing channels of relatively large and relatively small hydraulic diameters, the channels extending from the inlet end face and plugged at the outlet end face have both larger and smaller hydraulic diameters, and surround a single larger hydraulic diameter channel extending from the outlet end face and plugged at the inlet end face in a repeating pattern. This design is called the "Valuable Plugging Layout" or VPL™ by Ibiden.

[0034] WO 2021 / 126685 discloses a catalyst for capturing particulate matter.

[0035] There is a need in the art for a filter device for modern diesel vehicles to meet proposed exhaust emission limits, such as EU7, including limits on PM10 emissions, that balances good cold flow backpressure and soot load backpressure characteristics with good filtration efficiency, and that can be manufactured cost-effectively and with as little environmental impact as possible. Looking beyond vehicular diesel internal combustion engines, there is also a need in the art for a filter device for treating particulate matter in exhaust gases from vehicular internal combustion engines powered by carbon-neutral fuels, such as hydrogen or hydrogen-rich fuel stocks. The present invention aims to meet these needs. Summary of the Invention

[0036] According to a first aspect, the present invention provides a method of making a ceramic honeycomb wall-flow filter substrate having an axial length L supporting an on-wall coating, wherein a section of the on-wall coated filter substrate analyzed by mercury porosimetry has a pore volume of 0.05 to 0.5 μm diameter as a percentage of the mercury intrusion volume of the section of at least 15.0%, the wall-flow filter substrate having a first end and a second end and comprising an array of axially extending first and second channels, each of the first and second channels having a first end and a second end. wherein the channels are defined in part by porous channel walls having axially extending porous channel wall surfaces, each first channel sharing a porous channel wall with an adjacent second channel, and in a scanning electron microscope image of a cross section of the coated substrate, each channel has a geometric hollow cross-sectional shape with at least one vertex defined in part by an apex angle between two channel surface edges, the first channel is end-plugged at the second end of the substrate, and the second channel is end-plugged at the first end of the substrate, and the method comprises: (i) preparing a slurry comprising water, carboxylic acid, inorganic oxide particles or a mixture of two or more inorganic oxide particles having an equivalent spherical diameter D(v,0.5) of 0.5 to 14 μm as determined by laser diffraction and a modal aspect ratio (width / length) of 0.3 to 0.9 as determined by direct particle measurement using a Flow Imaging Microscopy particle shape analyzer; (ii) coating the slurry onto at least the porous channel walls of the first channels of a substrate, the substrate having an average pore size (D50) of 6 to 15 μm before any coating and a porosity of less than 60% before any coating; (iii) drying and firing the slurry-coated substrate; The product of step (iii) is 0.07 to 0.4 g / in based on the weight of the substrate before step (ii). 3 The present invention provides a method in which sufficient slurry is coated onto a substrate to have a coating loading of 4.3 to 24.4 g / L, and the product of step (iii) has an average on-wall coating thickness of 5 to 70 μm, as measured by the angle bisecting the apex angle on a scanning electron microscope cross-sectional image of the coated substrate. The applicant has found that adding the insoluble cellulose pore-forming agent in the amounts indicated results in the applied washcoat being pre-positioned on the wall (on-wall), rather than in-wall, as can be seen, for example, from the analysis shown in Example 9. It will be understood that the term "on-wall" in the products of the first and fourth aspects of the present invention does not exclude the coating from entering or being present within the porous channel walls of the substrate. Nevertheless, the products of the first and fourth aspects of the present invention are defined by the on-wall coating thickness, and the washcoat applied to the substrate in step (ii) of the first aspect of the present invention is primarily "on-wall."

[0037] The use of natural cellulose pore formers in the first aspect of the present invention may beneficially reduce the environmental impact compared to the prior art use of, for example, thermoplastic resins, plastic beads, etc., in that the combustion of plastics risks the release of dioxins, furans, mercury and polychlorinated biphenyls (BCPs) or the need to wash such toxins from effluents generated by prior art manufacturing processes.

[0038] Applicant prefers that in the method of the first aspect of the invention, the wall-flow filter substrate used in step (ii) is "bare", i.e. not coated with another washcoat or impregnated with an aqueous solution of metal salts prior to the start of step (ii).

[0039] The porous channel walls of the second channels may be uncoated or may be coated with the same or a different slurry composition as in step (i), i.e., in the first aspect, step (ii) may comprise coating the porous channel walls of the second channels of the filter substrate with a slurry according to step (i), the slurry for coating the second channels having the same or a different composition as the slurry coated on the porous channel walls of the first channels. For the avoidance of doubt, if present, the coating on the channel walls of the second channels is also an on-wall coating.

[0040] In a second aspect, the present invention provides a ceramic honeycomb wall-flow filter substrate supporting an on-wall coating for treating exhaust gases containing particulate matter emitted from an internal combustion engine, obtainable by a process according to the first aspect of the invention.

[0041] In a third aspect, the present invention provides a ceramic honeycomb wall-flow filter substrate supporting an on-wall coating for treating exhaust gases containing particulate matter emitted from an internal combustion engine obtained by a process according to the first aspect of the invention.

[0042] A fourth aspect of the present invention is an on-wall coated ceramic honeycomb wall-flow filter substrate for treating particulate matter-containing exhaust gas emitted from an internal combustion engine, the wall-flow filter substrate comprising an array of axially extending first and second channels having first and second ends and a length L therebetween, each of the first and second channels being defined in part by a porous channel wall having an axially extending porous channel wall surface, each of the first channels sharing a porous channel wall with an adjacent second channel, and in a scanning electron microscope image of a cross section of the coated substrate, each channel has a geometric hollow cross-sectional shape having at least one vertex defined in part by an apex angle between two channel surface edges, the first channels being end-plugged at the second end of the substrate and the second channels being end-plugged at the first end of the substrate, and at least the channel wall surface of the first channels having a density of 0.07 to 0.4 g / in based on the substrate weight before coating. 3 and a ceramic honeycomb wall flow filter substrate supporting an on-wall washcoat (4.3 to 24.4 g / L) of inorganic oxide particles or a mixture of two or more inorganic oxide particles, wherein the washcoat has an average on-wall coating thickness of 5 to 70 μm as measured at the angle bisecting the apex angle on a scanning electron microscope cross-sectional image of the coated substrate, wherein fragments of the on-wall coated filter substrate analyzed by mercury porosimetry have a pore volume of 0.05 to 0.5 μm in diameter as a percentage of the mercury intrusion volume of at least 15.0% of the fragments, and wherein the on-wall washcoat coating contains geometrically non-uniform voids with an equivalent spherical diameter D(v,0.5) of 0.1 to 8 μm and a modal aspect ratio (width / length) of 0.3 to 0.9 as determined by Focused Ion Beam-Scanning Electron Microscopy (FIB-SEM).

[0043] In a fifth aspect, the present invention provides an exhaust system for an internal combustion engine comprising a ceramic honeycomb wall-flow filter substrate according to the fourth aspect of the invention, wherein a first end of the substrate is oriented upstream.

[0044] According to a sixth aspect, the present invention provides an internal combustion engine including an exhaust system according to the fifth aspect.

[0045] According to a seventh aspect, the present invention provides a vehicle including a diesel engine according to the sixth aspect of the invention.

[0046] According to a seventh aspect, the present invention provides a vehicle including an internal combustion engine configured to be powered by a mixture of air and fuel, wherein the fuel of the air and fuel mixture is a gaseous fuel comprising a predominant fuel mass of hydrogen (H2), the vehicle comprising a source of gaseous fuel comprising a predominant fuel mass of hydrogen (H2), and an exhaust system according to the sixth aspect of the invention. DETAILED DESCRIPTION OF THE INVENTION

[0047] Method - First Aspect of the Invention While washcoat adhesion can be improved during development by increasing the "fineness" of the inorganic oxide particles in the inorganic oxide particle size distribution through milling (e.g., washcoat solids content in the milling solution, milling time and milling rate, etc.), Applicant's inventors have found that backpressure, filtration efficiency, and washcoat adhesion are generally more reliably met when the inorganic oxide particles, or inorganic oxides that are components of a mixture of inorganic oxide particles, have a loose bulk density before milling of 275 to 850 g / L, preferably 350 to 750 g / L.

[0048] The refractory metal oxide is typically selected from the group consisting of alumina, silica, titania, magnesia, ceria, zirconia, and mixed or composite oxides thereof, e.g., mixed or composite oxides of two or more thereof. For example, the refractory metal oxide may be selected from the group consisting of silica-alumina, titania-alumina, zirconia-alumina, titania-silica, zirconia-silica, ceria-zirconia, zirconia-titania, and alumina-magnesium oxide. During development, applicant prepared one coated sample in which the inorganic oxide was rare earth-doped CeO, but the backpressure, filtration efficiency, and washcoat adhesion characteristics were unfavorable (see Example 7 herein below). Another sample containing the same CeO mixed with gamma alumina did not sufficiently improve these properties. With further development, applicant believes that ceria-containing inorganic oxides or mixtures containing ceria-containing inorganic oxides may be identified for use in related aspects of the present invention. However, in view of the results shown in Example 7, Applicants prefer to optionally exclude inorganic oxide particles in the washcoat containing CeO2, either by itself or in a mixture.

[0049] Preferably, the inorganic oxide of the inorganic oxide particles comprises or consists of alumina, silica, zirconia, or a mixed oxide or composite oxide of any two or more thereof. This includes inorganic oxide particles comprising gamma-alumina or alumina doped with one or more of zirconium (Zr), titanium (Ti), silicon (Si), yttrium (Y), lanthanum (La), praseodymium (Pr), samarium (Sm), and neodymium (Nd), preferably silicon, and the total weight percent of the dopant present in the doped alumina is preferably 1 to 10 weight percent. Preferably, the alumina is not α-alumina.

[0050] The inclusion of a dopant can thermally stabilize the refractory metal oxide or support material. Any reference to "doped" in this context should be understood to refer to a material in which the bulk or host lattice of the refractory metal oxide is substitutionally or interstitially doped with a dopant. In some cases, a small amount of dopant may be present on the surface of the refractory metal oxide. However, the majority of the dopant is generally present in the bulk of the refractory metal oxide. The chemical and / or physical properties of the refractory metal oxide are often affected by the presence of the dopant.

[0051] As shown in Example 7, gamma alumina and doped alumina meeting the pre-mill loose bulk density requirement of 275-850 g / L exhibit acceptable adhesion, while doped alumina outside this pre-mill loose bulk density range does not. However, the adhesion of ZrO2 that does not meet the pre-mill loose bulk density requirement of 275-850 g / L can be improved to meet the acceptable adhesion limit by combining it with gamma alumina and doped alumina that meet the pre-mill loose bulk density requirement of 275-850 g / L. Although alpha alumina does not meet the pre-mill loose bulk density requirement of 275-850 g / L, it is less preferred for use in, for example, catalytic washcoats because, due to their low specific surface area, non-rhodium platinum group metals can sinter more easily on particulate inorganic oxide supports with lower specific surface areas, and therefore their activity is less durable.

[0052] Although the present invention applies to wall-flow filter substrates in which substantially all juxtaposed channels have equal hydraulic diameters, in a preferred embodiment the method according to the first aspect of the invention is applied to so-called asymmetric wall-flow filter substrates in which a first channel extending axially of the wall-flow filter substrate has a larger hydraulic diameter than a second channel.

[0053] In this first preferred embodiment, step (ii) may comprise coating the porous channel walls of the second channels of the filter substrate with the slurry from step (i), wherein the porous channel walls of the first channels are coated to an axial length of 60% to 90% L, preferably 70% to 85% L, and the porous channel walls of the second channels are coated to an axial length of 10% to 40% L, preferably 15% to 35% L, and the sum of the axial lengths of the on-wall coatings in the first and second channels is 100% or more. Preferably, the upper limit of overlap is 130% L.

[0054] For the avoidance of doubt, when the porous channel walls of the second channel are coated with a slurry in accordance with step (i), the g / in 3 The coating loading (g / L) is the combined total of the coating loading on the porous channel walls of both the first and second channels (i.e., relative to the substrate weight before applying the coating to either the first or second channel). The same applies to the definition of the fourth aspect of the invention.

[0055] In a second preferred embodiment, the axially extending first channels of the wall-flow filter substrate are a combination of channels having a larger hydraulic diameter and channels having a smaller hydraulic diameter, and the hydraulic diameter of the second channels is larger than the first channels having a smaller hydraulic diameter. That is, this embodiment is directed to applying the first aspect of the invention to an Ibiden VPL™ filter substrate, as described herein above in connection with SAE Technical Paper 2014-01-1512 and EP 2862610 A1.

[0056] In this second preferred embodiment, step (ii) may comprise coating the porous channel walls of the second channels of the filter substrate with the slurry from step (i), wherein the porous channel walls of the first channels are coated to an axial length of 60% to 80% L and the porous channel walls of the second channels are coated to an axial length of 60% to 80% L, and the sum of the axial lengths of the on-wall coatings in the first and second channels is 120% or more, preferably 130% or more. Preferably, the upper limit of overlap is 160% L.

[0057] The D(v,0.5) of the pore former in step (i) is preferably smaller than the measured coating thickness to avoid adhesion problems and poor filtration efficiency.

[0058] Where the coated filter product of the first aspect of the invention is for use in treating particulate matter from an internal combustion engine powered by hydrogen fuel (so-called "H2-ICE"), the on-wall coating comprises inorganic oxide particles and may be free of platinum group metals. However, for application to diesel engines, and optionally also to H2-ICE, the on-wall coating is preferably an on-wall catalytic coating, and the slurry of step (i) is such that the product of step (iii) is present in an amount of 0.5 to 10 g / ft 3 (0.018 to 0.353 g / L), preferably 1 to 8 g / ft 3 and one or more non-rhodium platinum group metal salts at a platinum group metal salt concentration selected to have a total platinum group metal loading of 0.035 to 0.283 g / L. Through analysis, Applicant has determined that the non-rhodium platinum group metal is immobilized on the inorganic oxide and has substantially no affinity for the cellulose pore-former.

[0059] The non-rhodium platinum group metal may be platinum alone or a combination of both platinum and palladium, and the product of step (iii) may have a platinum to palladium weight ratio of 1:0 > 1:1. The non-rhodium platinum group metal is a combination of both platinum and palladium, and the Pt:Pd weight ratio in the product of step (iii) is from 10:1 to 1:1, preferably from 6:1 to 2:1.

[0060] Reference Example 4 herein below demonstrates how D(v,0.9) can beneficially balance filtration efficiency with backpressure and potentially adhesion (see Example 5). For at least these reasons, the inorganic oxide particles preferably have a D(v,0.9) of 9-15 μm, more preferably 10-14 μm. Further improvement in washcoat adhesion can be obtained when the inorganic oxide particles have a D(v,0.5) of 3.0-6.0 μm, in combination with the D(v,0.9) parameter ranges described above. Applicant believes this improvement in washcoat adhesion can be attributed to an increase in "fine particles" in the washcoat.

[0061] In Applicant's development work, the inventors have identified two commercially available particulate insoluble cellulose raw materials for use as pore-formers in the slurry of step (i) of the first aspect of the present invention that result in a product that meets the above-mentioned needs in the art. In a first such embodiment, the particulate insoluble cellulose is microcrystalline cellulose having an equivalent spherical diameter D(v, 0.5) of 1 to 8.0 μm as determined by laser diffraction and a modal aspect ratio (width / length) of 0.6 to 0.8 as determined by direct particle measurement using a Flow Imaging Microscopy particle shape analyzer. The resulting product, as determined by FIB-SEM, comprises the characteristics shown in Figure 4.

[0062] In a second embodiment, the particulate insoluble cellulose is a fibrous cellulose having an equivalent spherical diameter D(v, 0.5) of 8.5-13.0 μm as determined by laser diffraction and a modal aspect ratio (width / length) of 0.4-0.7 as determined by direct particle measurement using a Flow Imaging Microscopy particle shape analyzer, resulting in a product having the characteristics shown in Figure 3 as determined by FIB-SEM.

[0063] The prior art discloses the use of washcoat slurries containing inorganic, ceramic, or metal fibers for coating on filter substrates. Such fibers not only pose potential health and safety hazards to manufacturing, such as asbestosis, but can also make coating control during manufacturing difficult (e.g., clogging of channels, agglomeration of the slurry, etc.). The use of a pore-forming agent in the first aspect of the present invention results in a reliable coating composition that produces a repeatable, more uniform product.

[0064] The filter substrate for use in the first aspect of the present invention is a ceramic filter substrate, for example constructed from silicon carbide or cordierite. When the filter substrate is a silicon carbide filter substrate, the porosity of the filter substrate is preferably 38-45% prior to coating. Applicant has found that this enables the resulting product to meet the EU7 emission standard for PM10.

[0065] In mercury porosimetry, Applicant determined that a silicon carbide filter substrate having a pre-coated porosity of 38-45% can result in a coated product having a porosity of less than 45.0% (see Example 6 herein below).

[0066] The coated silicon carbide filter substrate, a product of the first aspect of the present invention, having a pre-coated porosity of 38-45%, has a surface area of ​​at least 0.230 cm2 as determined by a combination of helium pycnometry and mercury densitometry.3 / g (see Example 6 hereinbelow).

[0067] Alternatively, the filter substrate may be a cordierite filter substrate, in which case the filter porosity before coating is preferably between 46 and 60%, for example to meet the EU7 emission standard for PM10.

[0068] In a preferred product of the present invention, the product of step (iii) is 0.15 to 0.3 gin -3 Sufficient slurry is coated onto the substrate to have a coating loading of (9.15-18.3 g / L). Applicants have found that selection of this parameter range can result in a product having a beneficial balance of back pressure, filtration efficiency, catalytic activity, and adhesion.

[0069] The average on-wall coating thickness is preferably between 10 and 60 μm to promote filtration efficiency and moderate back pressure.

[0070] The carboxylic acid can be a dicarboxylic acid or tricarboxylic acid, preferably citric acid or succinic acid. The carboxylic acid can promote a number of benefits, including slurry rheology, reduction of non-rhodium platinum group metal salts to improve nanoparticle dispersion, and some micropore generation when gas is evolved during drying and sintering. Preferably, the carboxylic acid concentration in step (i) is such that the product of step (ii) has a concentration of 50 to 400 g / ft 3 The lower end of this parameter range is preferred for non-platinum group metal-containing washcoats for H2-ICE applications.

[0071] The geometric shape of the first channel cross section in the filter substrate for use in the present invention is a square. Although filter substrates having channels with triangular cross sections are known, in Applicant's experience, coating channels with triangular cross sections can cause clogging of the channels.

[0072] The product itself - the fourth aspect of the invention In the fourth aspect of the present invention, preferably the porous channel walls of the second channels of the filter substrate are also coated with an on-wall washcoat, which has the same or a different composition as the washcoat coated on the porous channel walls of the first channels.

[0073] The refractory metal oxide is typically selected from the group consisting of alumina, silica, titania, magnesia, zirconia, and mixed oxides or composite oxides thereof, such as mixed oxides or composite oxides of two or more thereof. For example, the refractory metal oxide may be selected from the group consisting of silica-alumina, titania-alumina, zirconia-alumina, titania-silica, zirconia-silica, zirconia-titania, and alumina-magnesium oxide.

[0074] Preferably, the inorganic oxide of the inorganic oxide particles comprises or consists of alumina, silica, zirconia, or a mixed or composite oxide of any two or more thereof, but preferably is not alpha alumina, for at least the reasons explained hereinabove. Preferably, the inorganic oxide particles comprise gamma alumina or alumina doped with one or more of zirconium (Zr), titanium (Ti), silicon (Si), yttrium (Y), lanthanum (La), praseodymium (Pr), samarium (Sm), neodymium (Nd), preferably silicon, and the total weight percent of dopants present in the doped alumina is 1 to 10 weight percent.

[0075] The first axially extending channel of the wall-flow filter substrate may have a larger hydraulic diameter than the second channel.

[0076] In a first embodiment according to the fourth aspect of the present invention, the porous channel walls of the first channel are coated to an axial length of 60% to 90% L, preferably 70% to 85% L, and the porous channel walls of the second channel are coated to an axial length of 10% to 40% L, preferably 15% to 35% L, and the sum of the axial lengths of the on-wall coatings in the first and second channels is 100% or more. Preferably, the upper limit of the overlap is 130% L.

[0077] In a second embodiment of the fourth aspect of the present invention, the axially extending first channels of the wall-flow filter substrate are a combination of channels having larger hydraulic diameters and channels having smaller hydraulic diameters, and the hydraulic diameter of the second channels is larger than the first channels having smaller hydraulic diameters.

[0078] Applicant has found that when coating a Valuable Plugging Layout filter substrate design using known techniques, such as those described in Applicant's International Publication No. WO 99 / 47260, with a target of slight axial overlap between the inlet and outlet channel coatings, differences in the hydraulic diameters of the inlet channels result in the catalytic washcoat extending to a greater axial length in the inlet channel having a larger hydraulic diameter than in the adjacent inlet channel having a smaller hydraulic diameter. Thus, in this second embodiment, the porous channel walls of the first channel are coated to an axial length of 60%-80%L, and the porous channel walls of the second channel are coated to an axial length of 60%-80%L, with the sum of the axial lengths of the on-wall coatings in the first and second channels being 120% or greater, preferably 130% or greater. Preferably, the upper limit of the overlap is 160%L.

[0079] Where the on-wall coated ceramic honeycomb wall flow filter substrate is intended specifically for use in treating exhaust gases from diesel engines, but possibly H2-ICE as well, the on-wall coating preferably has a coating density of 0.5 to 10 g / ft 3 (0.018 to 0.353 g / L), preferably 1 to 8 g / ft 3 The on-wall catalytic coating comprises one or more non-rhodium platinum group metals at a total platinum group metal loading of (0.035-0.283 g / L).

[0080] The non-rhodium platinum group metal may be platinum alone or a combination of both platinum and palladium in a weight ratio of platinum to palladium of 1:0 > 1:1.

[0081] When the on-wall catalytic coating contains both platinum and palladium, the weight ratio of Pt:Pd may be from 10:1 to 1:1, preferably from 6:1 to 2:1.

[0082] Preferably, the on-wall washcoat does not contain inorganic, ceramic, or metallic fibers, for at least the reasons explained herein above.

[0083] In one embodiment, the filter substrate is a silicon carbide filter substrate having a porosity of 38-45% before coating, and the coated filter substrate has a porosity of less than 45.0% as determined by mercury porosimetry (see Example 6 hereinbelow). The coated filter substrate of this embodiment has a porosity of at least 0.230 cm2 as determined by a combination of helium pycnometry and mercury densitometry. 3 / g (see also Example 6 herein below).

[0084] In another embodiment, the filter substrate is a cordierite filter substrate having a porosity of 46-60% before coating.

[0085] Preferably, the on-wall washcoat coating is 0.1 to 0.3 gin -3 The coating is carried out with a loading of (6.1 to 18.3 g / L).

[0086] Preferably, the average on-wall coating thickness is between 10 and 60 μm.

[0087] Preferably, the geometric shape of the first channel cross section is rectangular.

[0088] Exhaust System - Fifth Aspect of the Invention In an exhaust system according to a fifth aspect of the present invention, a flow-through substrate supporting a diesel oxidation catalyst washcoat is preferably positioned upstream of the ceramic honeycomb wall-flow filter substrate. The diesel oxidation catalyst oxidizes NO in the exhaust gas to NO2 in order to combust particulate matter retained on the downstream filter substrate. PM advantageously burns in NO2 at lower exhaust gas temperatures than in O2 (the so-called CRT® effect).

[0089] Additionally or alternatively, the exhaust system of the fifth aspect of the present invention includes a flow-through substrate supporting a selective catalytic reduction catalyst or a wall-flow filter substrate supporting a selective catalytic reduction catalyst, the flow-through substrate or wall-flow filter substrate being positioned downstream from the ceramic honeycomb wall-flow filter substrate and including a source of nitrogen reductant (preferably urea) and an injector for injecting the nitrogen reductant into the exhaust gas flowing at a point between the catalyzed ceramic honeycomb wall-flow filter substrate and the flow-through substrate or wall-flow filter substrate supporting the selective catalytic reduction catalyst.

[0090] definition The term "mixed oxide," as used herein, generally refers to a mixture of oxides in a single phase, as conventionally known in the art. The term "complex oxide," as used herein, generally refers to a composition of oxides having two or more phases, as conventionally known in the art.

[0091] The acronym "PGM" as used herein refers to "platinum group metals." The term "platinum group metals" generally refers to the metals Ru, Rh, Pd, Os, Ir, and Pt of the periodic table, and particularly to the metals Ru, Rh, Pd, Ir, and Pt.

[0092] Herein, g ft -3 (grams per cubic foot) or g in -3 Any reference herein to amounts in units of grams per cubic inch (grams per cubic inch) refers to the average weight of an ingredient per volume of substrate. The volume of a substrate is calculated by its external dimensions, which is the total volume of the substrate, ignoring any channels extending through the substrate, e.g., 165 mm diameter x 140.5 mm length = h x π x radius. 2 = 3.0 liters is the volume of the cylindrical substrate. 1 g ft -3 A conversion from 1 g in to 0.035 g / L may be used. -3 A conversion from 0.01 to 61.0 g / L may be used.

[0093] As used herein, the phrase "consisting essentially of" limits the scope of a feature to a new, essential feature within the context of the defined element, e.g., the specified materials or steps, and any other materials or steps that do not substantially affect the essential characteristics of that feature, e.g., trace impurities. The phrase "consisting essentially of" encompasses the phrase "consisting of."

[0094] Total pore volume (cm) of a piece of on-wall coated filter substrate 3 / g) was determined by helium pycnometry and mercury densitometry. Helium pycnometry is the skeletal density (g / cm3 ) (sometimes called true, absolute, or helium density). Mercury densitometry determines geometric density (g / cm 3 ) cm 3 The total pore volume in g / g is calculated as the reciprocal of the geometric density minus the reciprocal of the skeletal density, i.e., total pore volume = (1 / geometric density) - (1 / skeletal density).

[0095] The porosity of the on-wall coated filter substrate pieces was also determined by helium pycnometry and mercury densitometry, where porosity = (1 - (skeletal density (g / cm 3 ) / geometric density(g / cm 3 ))×100.

[0096] Direct particle measurements using Flow Imaging Microscopy were performed using a Flowcam 8100 (Yokogawa Fluid Imagining Technologies, Inc., see https: / / www.fluidimaging.com / products / flowcam-nano-submicron-particle-imaging) controlled by Visual Spreadsheet 5 software equipped with a 20X objective, a FOV50 flow cell, a 0.5 mL syringe, and a grayscale camera. Samples were dispersed in deionized water at a concentration of 0.1 wt% for 2 hours at 400 rpm using a stir plate equipped with a magnetic follower. The suspension was then filtered through a 35 μm nylon mesh. Two drops of sample were diluted with 1 mL of deionized water for analysis. The solution was analyzed at a flow rate of 150 μL / min. Particle detection was performed using a dark pixel threshold of 20 and four closed-hole replicates. Data sets were obtained from over 10,000 particles. The list of individual particle features was exported to Microsoft Excel spreadsheet software and analyzed.

[0097] Particle size measurements were obtained by laser diffraction particle size analysis using a Malvern Mastersizer 2000, which is a volume-based technique (i.e., D(v,0.1), D(v,0.5), D(v,0.9), and D(v,0.98) are DV10, DV50, DV90, and DV98, respectively (or may also be referred to as D10, D50, D90, and D98, respectively), applying the mathematical Mie theory model to determine particle size distribution. Diluted washcoat samples were prepared in surfactant-free distilled water by sonication at 35 watts for 30 seconds.

[0098] The term "modal" from "modal" average has its usual meaning in statistics of "the most frequently occurring value in a set of data values." In a graphical representation of the aspect ratios of particles in a population, the "mode" may appear as the highest "peak."

[0099] Hydraulic diameter is a well-known term in the art when dealing with flow in non-circular pipes and channels. This term can be used to calculate many things in the same way as for circular pipes. If the cross section is uniform along the length of the pipe or channel, the hydraulic diameter is D H = 4A / P, where A is the cross-sectional area of ​​the flow and P is the wetted perimeter of the cross section. Measurement of the cross section and wetted perimeter can be performed by simple measurement of the channel at the exposed coated filter substrate end surface.

[0100] The term "insoluble" as used herein requires that the pore-forming agent does not dissolve in the slurry (i.e., in water and under process conditions) during the washcoating process. That is, the pore-forming agent is insoluble in water and insoluble in the presence of additional washcoat components, such as carboxylic acids, in the process. This means that the pore-forming agent maintains its heterogeneous morphology throughout the process. Preferably, the pore-forming agent is completely insoluble in the washcoat at the coating temperature, regardless of how long it remains in the washcoat, and preferably remains insoluble even at a temperature of 100°C. The insolubility of a component can be assessed by any known technique, such as assessing the dry weight of the component before and after contact with the washcoat to determine that there is no mass loss. It can also be assessed that the particles do not lose their morphological heterogeneity.

[0101] The term "non-uniform morphology" is used herein to describe individual particles of pore-former and is interpreted in its ordinary sense in the art, i.e., the shape of the pore-former is not uniform (particularly not spherical, as is common for such particle sizes, but also not any other uniform shape, such as square, cylinder, etc.). Similarly, the term "geometrically non-uniform" is used to describe voids formed through the process after the non-uniformly shaped pore-former particles are calcined and removed. These terms are used synonymously but refer to different parts (pore-former or voids) to avoid confusion.

[0102] The present invention may also be defined according to one or more of the following definitions. 1. A method of making a ceramic honeycomb wall-flow filter substrate having an axial length L supporting an on-wall coating, wherein a section of the on-wall coated filter substrate analyzed by mercury porosimetry has a pore volume of 0.05 to 0.5 μm diameter as a percentage of the mercury intrusion volume of the section of at least 15.0%, the wall-flow filter substrate having a first end and a second end and comprising an array of axially extending first and second channels, each of the first and second channels having: the coated substrate is defined in part by porous channel walls having axially extending porous channel wall surfaces, each first channel sharing a porous channel wall with an adjacent second channel, and in a scanning electron microscope image of a cross section of the coated substrate, each channel has a geometric hollow cross-sectional shape with at least one vertex defined in part by an apex angle between two channel surface edges, the first channel being end-capped at the second end of the substrate and the second channel being end-capped at the first end of the substrate, the method comprising: (i) preparing a slurry comprising water, carboxylic acid, inorganic oxide particles or a mixture of two or more inorganic oxide particles having an equivalent spherical diameter D(v,0.5) of 0.5 to 14 μm as determined by laser diffraction and a modal aspect ratio (width / length) of 0.3 to 0.9 as determined by direct particle measurement using a Flow Imaging Microscopy particle shape analyzer; (ii) coating the slurry onto at least the porous channel walls of the first channels of a substrate, the substrate having an average pore size (D50) of 6 to 15 μm before any coating and a porosity of less than 60% before any coating; (iii) drying and firing the slurry-coated substrate; The product of step (iii) is 0.07 to 0.4 g / in based on the weight of the substrate before step (ii). 3 wherein sufficient slurry is coated onto the substrate to have a coating loading of (4.3 to 24.4 g / L) and the average on-wall coating thickness of the product of step (iii) is between 5 and 70 μm, as measured by bisecting the vertex angle on a scanning electron microscope cross-sectional image of the coated substrate. 2. The method according to item 1, wherein step (ii) comprises coating the porous channel walls of the second channels of the filter substrate with the slurry according to step (i), and the slurry for coating the porous channel walls of the second channels has the same or a different composition as the slurry coated on the porous channel walls of the first channels. 3. The method according to item 1 or 2, wherein the inorganic oxide particles or the mixture of inorganic oxide particles has a loose bulk density before pulverization of 275 to 850 g / L. 4. The method according to Item 3, wherein the inorganic oxide particles or the mixture of inorganic oxide particles has a loose bulk density before pulverization of 350 to 750 g / L. 5. The method according to item 1, 2, 3 or 4, wherein the inorganic oxide of the inorganic oxide particles comprises alumina, silica, zirconia, or a mixed oxide or composite oxide of any two or more thereof. 6. The method according to any one of paragraphs 1 to 5, wherein the inorganic oxide particles comprise gamma alumina or alumina doped with one or more of zirconium (Zr), titanium (Ti), silicon (Si), yttrium (Y), lanthanum (La), praseodymium (Pr), samarium (Sm), and neodymium (Nd), preferably silicon. 7. The method of claim 6, wherein the total weight percent of dopants present in the doped alumina is 1 to 10 weight percent. 8. The method according to any one of paragraphs 1 to 7, wherein the inorganic oxide particles themselves or the inorganic oxide particles in the mixture are either wholly or partially free of CeO2. 9. The method of any one of paragraphs 1 to 8, wherein the first axially extending channels of the wall-flow filter substrate have a larger hydraulic diameter than the second channels. 10. The method of item 9, wherein step (ii) comprises coating the porous channel walls of the second channels of the filter substrate with the slurry from step (i), wherein the porous channel walls of the first channels are coated to an axial length of 60% to 90% L, and the porous channel walls of the second channels are coated to an axial length of 10% to 40% L, and the sum of the axial lengths of the on-wall coating on the porous channel walls of the first and second channels is 100% or more. 11. The method of item 10, wherein the porous channel walls of the first channel are coated to an axial length of 70% to 85% L, and the porous channel walls of the second channel are coated to an axial length of 15% to 35% L. 12. The method of any one of paragraphs 1 to 8, wherein the first channels extending axially in the wall-flow filter substrate are a combination of channels having larger hydraulic diameters and channels having smaller hydraulic diameters, and the hydraulic diameter of the second channels is larger than the first channels having smaller hydraulic diameters. 13. The method of claim 12, wherein step (ii) comprises coating the porous channel walls of the second channels of the filter substrate with the slurry from step (i), wherein the porous channel walls of the first channels are coated to an axial length of 60% to 80% L, the porous channel walls of the second channels are coated to an axial length of 60% to 80% L, and the sum of the axial lengths of the on-wall coatings in the first and second channels is 120% or more. 14. The method according to item 13, wherein the sum of the axial lengths of the on-wall coatings in the first channel and the second channel is 130% or more. 15. The method according to any one of paragraphs 1 to 14, wherein D(v,0.5) of the pore-forming agent in step (i) is less than the measured coating thickness. 16. The on-wall coating is an on-wall catalytic coating, and the slurry of step (i) has a coating density of 0.5 to 10 g / ft 3 Item 16. The method of any one of items 1 to 15, comprising one or more non-rhodium platinum group metal salts at a platinum group metal salt concentration selected to have a total platinum group metal loading of (0.018 to 0.353 g / L). 17. Total platinum group metal loading is 1 to 8 g / ft 3 Item 17. The method according to Item 16, wherein the concentration is (0.035 to 0.283 g / L). 18. The method of any one of paragraphs 1 to 17, wherein the non-rhodium platinum group metal is platinum alone or a combination of both platinum and palladium, and the product of step (iii) has a weight ratio of platinum to palladium of 1:0 > 1:1. 19. The method according to item 18, wherein the weight ratio of Pt:Pd in ​​the product of step (iii) is 10:1 to 1:1, preferably 6:1 to 2:1. 20. The method according to any one of paragraphs 1 to 19, wherein the inorganic oxide particles have a D(v,0.9) of 9 to 15 μm. 21. The method according to item 20, wherein the inorganic oxide particles have a D(v,0.5) of 3.0 to 6.0 μm. 22. The method according to any one of paragraphs 1 to 21, wherein the particulate insoluble cellulose is microcrystalline cellulose having an equivalent spherical diameter D(v,0.5) of 1 to 8.0 μm as determined by laser diffraction and a mode aspect ratio (width / length) of 0.6 to 0.8 as determined by direct particle measurement using a Flow Imaging Microscopy particle shape analyzer. 23. The method according to any one of paragraphs 1 to 21, wherein the particulate insoluble cellulose is fibrous cellulose having an equivalent spherical diameter D(v,0.5) of 8.5 to 13.0 μm as determined by laser diffraction and a mode aspect ratio (width / length) of 0.4 to 0.7 as determined by direct particle measurement using a Flow Imaging Microscopy particle shape analyzer. 24. The method of any one of paragraphs 1 to 23, wherein the inorganic oxide particles are not alpha alumina. 25. The method according to any one of items 1 to 24, wherein the slurry does not contain inorganic fibers, ceramic fibers, or metal fibers. 26. The method of any one of paragraphs 1 to 25, wherein the filter substrate is a silicon carbide filter substrate having a porosity of 38 to 45% before coating. 27. The method of claim 26, wherein the coated filter substrate has a porosity of less than 45.0%. 28. The coated filter substrate has a surface area of ​​at least 0.230 cm2 as determined by a combination of helium pycnometry and mercury densitometry. 3 Item 28. The method according to item 26 or 27, having a total pore volume of / g. 29. The method of any one of paragraphs 1 to 25, wherein the filter substrate is a cordierite filter substrate having a porosity of 46 to 60% before coating. 30. The product of step (iii) is 0.15 to 0.3 gin -3 30. The method of any one of paragraphs 1 to 29, wherein sufficient slurry is coated onto the substrate to have a coating loading of (9.15 to 18.3 g / L). 31. The method of claim 30, wherein the average on-wall coating thickness is between 10 and 60 μm. 32. The method according to any one of items 1 to 31, wherein the carboxylic acid is a dicarboxylic acid or tricarboxylic acid, preferably citric acid or succinic acid. 33. The carboxylic acid concentration in step (i) is 50 to 400 g / ft 3 Item 33. The method according to any one of Items 1 to 32, wherein the carboxylic acid concentration is selected to be in the range of 1.77 to 14.1 g / L. 34. The method according to any one of items 1 to 33, wherein the firing temperature is higher than 400°C ± 20°C. 35. The method of any one of paragraphs 1 to 34, wherein the geometric shape of the first channel cross section is a rectangle. 36. A ceramic honeycomb wall-flow filter substrate carrying an on-wall coating for treating exhaust gases containing particulate matter emitted from an internal combustion engine, obtainable by the process according to any one of paragraphs 1 to 35. 37. A ceramic honeycomb wall-flow filter substrate supporting an on-wall coating for treating exhaust gases containing particulate matter emitted from an internal combustion engine, obtained by the process of any one of paragraphs 1 to 35. 38. An on-wall coated ceramic honeycomb wall-flow filter substrate for treating exhaust gases containing particulate matter emitted from an internal combustion engine, the wall-flow filter substrate having a first end and a second end, a length L therebetween, and including an array of axially extending first and second channels, each of the first and second channels being defined in part by a porous channel wall having an axially extending porous channel wall surface, each of the first channels sharing a porous channel wall with an adjacent second channel, and in a scanning electron microscope image of a cross section of the coated substrate, each channel has a geometric hollow cross-sectional shape having at least one vertex defined in part by an apex angle between two channel surface edges, and the first channels are positioned at the second end of the substrate and terminate at the third end. and a second channel is plugged at a first end of the substrate, wherein at least the channel wall surface of the first channel supports an on-wall washcoat of 0.07 to 0.4 g / in3 (4.3 to 24.4 g / L) based on the substrate weight before coating, the washcoat comprising inorganic oxide particles or a mixture of two or more inorganic oxide particles, the average on-wall coating thickness measured at the angle bisecting the apex angle on a scanning electron microscope cross-sectional image of the coated substrate is 5 to 70 μm, and a fragment of the on-wall coated filter substrate analyzed by mercury porosimetry has a pore volume of 0.05 to 0.5 μm diameter as a percentage of the mercury intrusion volume of at least 15.0% of the fragment, and the on-wall washcoat coating is Focused On-wall coated ceramic honeycomb wall-flow filter substrates containing geometrically non-uniform voids with equivalent spherical diameters D(v,0.5) between 0.1 and 8 μm and modal aspect ratios (width / length) between 0.3 and 0.9, as determined by Ion Beam-Scanning Electron Microscopy (FIB-SEM). 39. An on-wall coated ceramic honeycomb wall-flow filter substrate according to paragraph 38, wherein the porous channel walls of the second channels of the filter substrate are also coated with an on-wall washcoat, the on-wall washcoat having the same or a different composition as the washcoat coated on the porous channel walls of the first channels. 40. The on-wall coated ceramic honeycomb wall-flow filter substrate according to paragraph 38 or 39, wherein the inorganic oxide of the inorganic oxide particles comprises alumina, silica, zirconia, or a mixed oxide or composite oxide of any two or more thereof. 41. The on-wall coated ceramic honeycomb wall flow filter substrate of paragraph 38, 39 or 40, wherein the inorganic oxide particles comprise gamma alumina or alumina doped with one or more of zirconium (Zr), titanium (Ti), silicon (Si), yttrium (Y), lanthanum (La), praseodymium (Pr), samarium (Sm), neodymium (Nd), preferably silicon. 42. The on-wall coated ceramic honeycomb wall-flow filter substrate of paragraph 41, wherein the total weight percent of dopants present in the doped alumina is 1 to 10 weight percent. 43. An on-wall coated ceramic honeycomb wall flow filter substrate according to any one of paragraphs 38 to 42, wherein the inorganic oxide particles themselves or the inorganic oxide particles in the mixture do not contain CeO2, either in whole or in part. 44. An on-wall coated ceramic honeycomb wall-flow filter substrate according to paragraph 38, 39, 40, 41 or 42, wherein the first axially extending channels of the wall-flow filter substrate have a larger hydraulic diameter than the second channels. 45. The on-wall coated ceramic honeycomb wall-flow filter substrate according to any one of items 38 to 44, wherein the porous channel walls of the first channel are coated to an axial length of 60% to 90% L, and the porous channel walls of the second channel are coated to an axial length of 10% to 40% L, and the sum of the axial lengths of the on-wall coatings in the first channel and the second channel is 100% or more. 46. ​​An on-wall coated ceramic honeycomb wall flow filter substrate according to paragraph 45, wherein the porous channel walls of the first channel are coated to an axial length of 70% to 85% L, and the porous channel walls of the second channel are coated to an axial length of 15% to 35% L. 47. An on-wall coated ceramic honeycomb wall-flow filter substrate according to any one of paragraphs 38 to 43, wherein the first channels extending axially of the wall-flow filter substrate are a combination of channels having a larger hydraulic diameter and channels having a smaller hydraulic diameter, and the hydraulic diameter of the second channels is larger than the first channels having the smaller hydraulic diameter. 48. An on-wall coated ceramic honeycomb wall flow filter substrate according to item 47, wherein the porous channel walls of the first channel are coated to an axial length of 60% to 80% L, the porous channel walls of the second channel are coated to an axial length of 60% to 80% L, and the sum of the axial lengths of the on-wall coatings in the first channel and the second channel is 120% or more. 49. The on-wall coated ceramic honeycomb wall-flow filter substrate according to item 48, wherein the sum of the axial lengths of the on-wall coating in the first channel and the second channel is 130% or more. 50. On-wall coating: 0.5-10g / ft 3Item 38 to 49, wherein the on-wall catalytic coating comprises one or more non-rhodium platinum group metals at a total platinum group metal loading of (0.018 to 0.353 g / L). An on-wall coated ceramic honeycomb wall flow filter substrate according to any one of items 38 to 49. 51. Total platinum group metal loading is 1 to 8 g / ft 3 Item 50. An on-wall coated ceramic honeycomb wall flow filter substrate according to item 50, wherein the solubility is (0.035 to 0.283 g / L). 52. The on-wall coated ceramic honeycomb wall flow filter substrate of any one of paragraphs 38 to 48, wherein the non-rhodium platinum group metal is platinum alone or a combination of both platinum and palladium in a platinum to palladium weight ratio of 1:0 > 1:1. 53. An on-wall coated ceramic honeycomb wall-flow filter substrate according to item 52, wherein the weight ratio of Pt:Pd is 10:1 to 1:1, preferably 6:1 to 2:1. 54. The on-wall coated ceramic honeycomb wall-flow filter substrate of any one of paragraphs 38 to 53, wherein the inorganic oxide particles are not alpha alumina. 55. The on-wall coated ceramic honeycomb wall flow filter substrate according to any one of paragraphs 38 to 54, wherein the on-wall washcoat does not contain inorganic fibers, ceramic fibers, or metal fibers. 56. An on-wall coated ceramic honeycomb wall-flow filter substrate according to any one of paragraphs 38 to 55, wherein the filter substrate is a silicon carbide filter substrate having a porosity of 38 to 45% before coating. 57. An on-wall coated ceramic honeycomb wall-flow filter substrate according to paragraph 56, wherein the coated filter substrate has a porosity of less than 45.0%. 58. The coated filter substrate has a surface area of ​​at least 0.230 cm2 as determined by a combination of helium pycnometry and mercury densitometry. 3Item 58. The on-wall coated ceramic honeycomb wall flow filter substrate of item 56 or 57, having a total pore volume of 1 / g. 59. An on-wall coated ceramic honeycomb wall flow filter substrate according to any one of paragraphs 38 to 55, wherein the filter substrate is a cordierite filter substrate having a porosity of 46 to 60% before coating. 60. On-wall washcoat coating is 0.1 to 0.3 gin -3 Item 38 to 59, wherein the on-wall coated ceramic honeycomb wall flow filter substrate is coated with a loading of (6.1 to 18.3 g / L). 61. An on-wall coated ceramic honeycomb wall flow filter substrate according to paragraph 60, wherein the average on-wall coating thickness is between 10 and 60 μm. 62. The on-wall coated ceramic honeycomb wall-flow filter substrate of any one of paragraphs 38 to 61, wherein the geometric shape of the first channel cross section is a square. 63. An exhaust system for an internal combustion engine comprising the ceramic honeycomb wall-flow filter substrate of any one of paragraphs 38 to 62, wherein a first end of the substrate is directed upstream. 64. An exhaust system as described in paragraph 63, including a flow-through substrate supporting a diesel oxidation catalyst washcoat, positioned upstream of the ceramic honeycomb wall-flow filter substrate. 65. An exhaust system according to paragraph 63 or 64, comprising a flow-through substrate supporting a selective catalytic reduction catalyst or a wall-flow filter substrate supporting a selective catalytic reduction catalyst, the flow-through substrate or wall-flow filter substrate being positioned downstream from the ceramic honeycomb wall-flow filter substrate, and comprising a source of nitrogen reducing agent (preferably urea), and an injector for injecting the nitrogen reducing agent into the exhaust gas flowing at a point between the catalyzed ceramic honeycomb wall-flow filter substrate and the flow-through substrate or wall-flow filter substrate supporting the selective catalytic reduction catalyst. 66. An internal combustion engine comprising an exhaust system according to paragraph 60, 64 or 65. 67. A vehicle including a diesel engine according to paragraph 66. 68. A vehicle including an internal combustion engine configured to be powered by a mixture of air and fuel, wherein the fuel of the air and fuel mixture is a gaseous fuel comprising a predominant fuel mass of hydrogen (H2), and the vehicle comprises a source of gaseous fuel comprising a predominant fuel mass of hydrogen (H2), and an exhaust system according to paragraph 63, 64 or 65. [Brief explanation of the drawings]

[0103] In order that the invention may be more fully understood, the following examples are given, by way of example only, and with reference to the accompanying drawings, in which: [Figure 1] 1 is a plot of mercury log differential intrusion versus pore size for some of the samples from Example 6, Table 13, for pores in the 0.05 to 0.5 μm diameter range. [Figure 2] 1 is a backscattered electron scanning electron microscope (BSE-SEM) image of a cross section of a coated filter substrate showing the juxtaposed inlet and outlet channels of an asymmetric filter substrate, the angle bisecting the apex angle between the two edges of the channel's rectangular cross section, and the orthogonally juxtaposed channel wall surfaces along which the thickness of the on-wall washcoat at the channel "corners" is measured. [Figure 3] Report of image analysis of local three-dimensional pores measured in the washcoat thickness of Sample 1 in Table 4 after A. Varambhia et al., Johnson Matthey Technol. Rev., 2022, 66, (3), 355-371. [Figure 4] 1 is a report of an image analysis of localized three-dimensional pores measured in the washcoat thickness of Sample 1 in Table 4. [Example]

[0104] Example 1 - Evaluation of the effect of pore formers of different particle sizes on filtration and back pressure Three different samples were prepared, each containing a cylindrical silicon carbide wall-flow filter substrate (165 mm diameter x 140.5 mm length) with an asymmetric channel arrangement (larger hydraulic diameter inlet channels, smaller hydraulic diameter outlet channels) of 46.5 cells per square centimeter (300 cells per square inch). The uncoated filter had a porosity of 42% and an average pore size of 14 μm.

[0105] In each untreated substrate, the inlet channel with the larger hydraulic diameter of the asymmetric channel configuration was coated with the washcoat for 80% of its axial length, and the outlet channel with the smaller hydraulic diameter was coated with the same washcoat for 20% of its axial length. The washcoat in each sample contained particulate gamma alumina, citric acid, platinum nitrate, a pore former, and a rheology modifier to achieve a desired viscosity for coating the substrate according to the method disclosed in WO 99 / 47260, i.e., (a) placing a containment means on top of the substrate; (b) introducing a predetermined amount of liquid component into the containment means, in either order (a) then (b) or (b) then (a); and (c) applying a high vacuum to draw the entire amount of liquid component into at least a portion of the substrate and retaining substantially all of the amount within the support without recovery.

[0106] The washcoat component ratio was 0.3 g / in of the coated substrate before baking. 3 (18.3g / L) of gamma alumina, 250g / ft 3 (8.83g / L) citric acid, 3g / ft 3 (0.11 g / L) Pt, 0.09 g / in 3 (5.49 g / L) of pore former, i.e., the applied washcoat was selected to achieve a target of 0.3 g / in in the calcined final product. 3The catalyst contained 30 wt. % pore former based on the gamma alumina washcoat solids required to obtain a loading of 18.3 g / L (18.3 g / L). All coated parts were statically calcined at 500°C, a process that resulted in combustion and removal of the pore former from the final product. The catalyst coating on both the inlet and outlet channels of the final product was substantially on-wall, as visible by SEM.

[0107] The pore former in each sample is listed in Table 1 below.

[0108] [Table 1] + D(v,0.5) / D(v,0.9) as determined by laser diffraction, assuming equivalent spherical diameters.

[0109] Sample Evaluation - Low Temperature Backpressure Cold flow backpressure analysis of the coated and aged filters was performed using a commercially available Superflow SF1020 instrument at https: / / superflow.com / products / flowbenches / .

[0110] Ambient temperature of 20°C ± 2°C and 600m 3 At a flow rate of 1 / h, the results are shown in Table 2.

[0111] [Table 2]

[0112] Sample evaluation - soot deposition back pressure Soot Loaded Back Pressure ("SLBP") testing was performed using the apparatus and method described in European Patent No. 1850068 and available for purchase from Cambustion® (Cambridge, UK) as the "DPG Particulate Filter Testing System" (see https: / / www.cambustion.com / products / engine-exhaust-emissions / dpg-particulate-filter-testing-system), namely: (i) an apparatus for producing and collecting particulate matter derived from the combustion of a liquid carbon-containing fuel, the apparatus comprising a fuel burner including a nozzle, the nozzle housed within a vessel, the vessel including a gas inlet and a gas outlet connected to a conduit for transporting gas from the gas outlet to the atmosphere, means for detecting a velocity of gas flowing through the gas inlet, means for forcing an oxidizing gas from the gas inlet through the vessel, the gas outlet and the conduit to the atmosphere, a station for collecting particulate matter from the gas flowing through the conduit, and means for controlling the means for forcing gas flow in response to the gas flow rate detected at the gas inlet, whereby the gas flow rate at the gas inlet is maintained at a desired rate to provide for sub-stoichiometric combustion of the fuel in the vessel, thereby promoting particulate matter formation; and (ii) A method for producing and collecting particulate matter derived from the combustion of a liquid carbon-containing fuel in an oxidizing gas, the method comprising: burning the fuel in a sub-stoichiometric amount of oxidizing gas in a fuel burner, the fuel burner including a nozzle housed within a vessel; forcing the oxidizing gas to flow from a gas inlet to the vessel, to the atmosphere, through a gas outlet, and into a conduit connected to the vessel and the gas inlet; collecting the particulate matter at a station located within the conduit; detecting the velocity of the oxidizing gas flow at the gas inlet; and controlling the velocity of the oxidizing gas flow so that a desired velocity of the oxidizing gas flow is maintained at the gas inlet.

[0113] The filter is inserted into the station to collect particulate matter from the gas flowing through the conduit. A new filter is first preconditioned using low-sulfur diesel fuel (10 ppm S) in a lean-burn combustion stream at an air flow rate of 80 kg / hr, raising the filter inlet temperature to 650°C, a temperature typically used in vehicles to regenerate soot-laden filters. The temperature of this preconditioning step is sufficiently higher than the soot combustion temperature to ensure that the filter under test is clean from the start. Pressure sensors located upstream and downstream of the station monitor the backpressure across the filter. The SLBP test is conducted at an air flow rate of 180 kg / hr and a filter inlet temperature of 250°C while burning low-sulfur diesel fuel (10 ppm S).

[0114] Filtration efficiency testing was also performed on the DPG Particulate Filter Testing System as follows. At a constant flow rate, the initial test detected the particulate number (PN) every second upstream of the filter using a PN counter. The upstream PN raw data was averaged (mean) over an 11-second detection period to generate an 11-second averaging window that moved every second to reduce the effect of any fluctuations in PN generation. Here, time "t" corresponds to the midpoint of the 11-second window. The sample filter was then tested by detecting post-filter PN, also using a PN counter. The instantaneous filtration efficiency at time "t" in the test is given by the following calculation: FE t [%] = [average uplink PN at time t - downlink PN at time t] / [average uplink PN at time t] × 100%

[0115] The sample SLBP and filtration efficiency analysis results are shown in Table 3.

[0116] [Table 3]

[0117] From the results shown in Tables 2 and 3, it can be seen that cellulose fibers having an equivalent spherical diameter D(v,0.5) between 0.5 and 14 μm, when used as pore formers in coating ceramic honeycomb wall-flow filter substrates according to the present invention, generate lower cold flow backpressures and soot deposition backpressures than cellulose fiber pore formers with higher D(v,0.5).

[0118] Example 2 - Laboratory and Engine Evaluation of the Effect of Pore Formers of Different Particle Sizes on Filtration and Backpressure Six different samples were prepared, each containing a cylindrical cordierite wall-flow filter substrate (143.8 mm diameter x 152.4 mm length) with an asymmetric channel configuration of 46.5 cells per square centimeter (300 cells per square inch). The uncoated filter had a porosity of 47% and an average pore size of 12 μm, with an asymmetric channel configuration of 46.5 cells per square centimeter (300 cells per square inch). The substrate used was different from that used in Example 1.

[0119] Each sample had a different particle size of 0.09 g / in 3 (5.5g / L) (30% by weight of 100% is 0.3g / in 3 The method of Example 1 was followed, except that a different pore former was used (18.3 g / L gamma alumina).

[0120] The pore former in each sample is listed in Table 4 below.

[0121] [Table 4] + D(v,0.5) / D(v,0.9) as determined by laser diffraction, assuming equivalent spherical diameters.

[0122] Sample Evaluation - Low Temperature Backpressure In the same manner as described in Example 1, the temperature was 20°C ± 2°C / flow rate 600 m 3The cold stream back pressure was evaluated at 1000 kJ / h and the results are shown in Table 5.

[0123] [Table 5]

[0124] Sample evaluation - soot deposition back pressure Soot load backpressure was performed as described herein above in Example 1. The results are shown in Table 6 below.

[0125] [Table 6]

[0126] Sample evaluation - engine test Three different samples were prepared, each containing a different cylindrical silicon carbide wall-flow filter substrate (304.8 mm diameter x 228.6 mm length) with an asymmetric channel configuration (larger hydraulic diameter inlet channels, smaller hydraulic diameter outlet channels) of 46.5 cells per square centimeter (300 cells per square inch), but coated with the coatings described in Table 7. The uncoated filter had a porosity of 41% and an average pore size of 9 μm. Each sample was prepared according to the method of Example 1.

[0127] Particle number (PN) evaluations were performed using a 12 L bench-mounted diesel engine running on EUVI B7 fuel, with the diesel particulate filter (DPF) being evaluated mounted downstream of the diesel oxidation catalyst using removable canning, and a PN counter (AVL) mounted downstream of the DPF. Each DPF was pre-loaded with 4 g / L of soot per filter substrate volume before evaluation over a "REAL Driving Emissions" (RDE) compliance test cycle, and post-DPF PN10 data were processed using a moving average window analysis to generate 90% percentile values ​​of PN / kWh. Results are shown in Table 7.

[0128] [Table 7]

[0129] Based on the current EU7 proposal, all three samples were 9.0 × 10 11 The RDE PN number per kWh for 10 nm is met over the RDE test cycle limit of 10 nm.

[0130] Example 3 - Pore-Forming Agent Loading Study Six different samples were prepared, each containing a cylindrical silicon carbide wall-flow filter substrate (143.8 mm diameter x 150.5 mm length) with an asymmetric channel arrangement (larger hydraulic diameter inlet channels, smaller hydraulic diameter outlet channels) of 46.5 cells per square centimeter (300 cells per square inch). The uncoated filter had a porosity of 42% and an average pore size of 11 μm. The substrate used was different from that used in Example 1 or Example 2.

[0131] The samples were prepared similarly to Example 1, except that the pore former used in all of the samples in Example 3 was the pore former from Sample 1.1 in Table 1, but the pore former was added at different weight percent levels relative to the 30 weight percent of Sample 1.1. The pore former loadings added to each sample washcoat are listed below in Table 8. Sample 1.1 was included as a reference sample.

[0132] [Table 8]

[0133] Sample Evaluation - Soot Deposition Backpressure and Filtration Efficiency Soot load backpressure was performed as described herein above in Example 1. The results are shown in Table 9 below.

[0134] [Table 9]

[0135] The results shown in Table 9 are unexpected. One skilled in the art would expect that increasing cellulose pore former loading would correlate with lower filtration efficiency because the introduction of increased pore volume creates more paths for soot to pass through. However, the results in Table 9 show that lower soot deposition backpressure and lower filtration efficiency stabilize at about 98% filtration efficiency above about 15 wt.% pore former loading.

[0136] Example 4 - Effect of washcoat particle size distribution on backpressure and filtration Six different samples were prepared, each containing a cylindrical silicon carbide wall-flow filter substrate (143.8 mm diameter x 150.5 mm length) with an asymmetric channel arrangement (larger hydraulic diameter inlet channels, smaller hydraulic diameter outlet channels) of 46.5 cells per square centimeter (300 cells per square inch). The uncoated filter had a porosity of 42% and an average pore size of 11 μm. The substrate used was different from that used in Example 1 or Example 2.

[0137] Each sample was prepared according to the method of Example 1 using the pore former of Sample 1. The washcoat component ratio was 0.4 g / in with various particle size distributions on the coated substrate before calcination. 3 (24.4g / L) of gamma alumina, 250g / ft 3 (8.83g / L) citric acid, 3g / ft 3 (0.11 g / L) 10:1 Pt:Pd, 0.125 g / in 3 (7.63 g / L) of pore former, i.e., the applied washcoat was selected to achieve a target of 0.4 g / in in the calcined final product. 3The gamma alumina washcoat solids required to achieve a loading of 24.4 g / L contained 31.25 wt. % pore former in addition to the gamma alumina washcoat solids. The difference between each sample was that the D(v, 0.9) of the gamma alumina had a 2 μm difference from the previous sample in that run, as shown in Table 10 below. The D(v, 0.9) of the gamma alumina was adjusted by ball milling.

[0138] [Table 10]

[0139] Sample Evaluation - Soot Deposition Backpressure and Filtration Efficiency Soot load backpressure was performed as described herein above in Example 1. The results are shown in Table 11 below.

[0140] [Table 11]

[0141] From the results shown in Table 11, it can be seen that filtration efficiency tends to increase as the back pressure increases with a decrease in D(v, 0.9). Generally, 3 For practical applications of a washcoat loading of (24.4 g / L), the filtration efficiency of the reference embodiment in this example is too low and the soot deposition backpressure is too high. However, in light of the results of this example, in order to target diesel exhaust particles to meet existing and future emission standards with acceptable adhesion (see Example 5 herein below), Applicant prefers a D(v,0.9) particle size range of 9-15 μm, more preferably 10-14 μm, for washcoats for use in the present invention.

[0142] Example 5. Adhesion test Prior to washcoat adhesion testing in the laboratory apparatus described below, each sample was held in hand and tapped on a table or smacked with the hand. If the washcoat fell off the sample, the sample immediately failed the adhesion test and was not subjected to laboratory apparatus testing. This test is referred to as the "tap test."

[0143] The washcoat adhesion of samples prepared as in Example 1 was evaluated in a laboratory test apparatus using a nozzle that emitted high-pressure air. A high-pressure air flow rate of 425 L / min was used to deliver a jet of air vertically downward onto the end face of a vertically positioned filter from 0.5 inches (1.3 cm) above the end face. The nozzle was robotically controlled to move across and laterally relative to the end face of the filter in a zigzag manner at a speed of 6.7 mm / sec, so that the entire end face of the filter was ultimately exposed to the high-pressure air treatment.

[0144] The filter samples to be tested were first stabilized by heating in an oven at 115°C for 30 minutes, after which the samples were weighed. The washcoat adhesion was then tested by exposing the filter samples to high-pressure air in the following order: outlet end, then inlet end, and finally outlet end, after which the samples were again heat stabilized and weighed. The difference in weight before and after exposure to the high-pressure air test (i.e., weight loss) was calculated as a percentage of the total baked coating applied to the filter. Results for various samples from the foregoing examples are shown in Table 12 hereinbelow.

[0145] Comparative samples 2 and 3 failed the "tap test" and therefore were not subjected to full adhesion testing.

[0146] [Table 12]

[0147] A % adhesion loss of greater than 2.0 is considered a "fail." From the results shown in Table 12, it can be seen that when more than 35 wt. % pore former is used (Sample 7.7), adhesion loss is observed.

[0148] Example 6 - Mercury (Hg) Porosimetry Silicon carbide wall-flow filters with 300 cells per square inch (equivalent to 46.5 cells per square centimeter) were coated with the test formulations of Examples 9.1-9.8. Cylindrical cores approximately 10 mm in diameter and 10 mm long were cut from the filters for analysis. The inlet "face" of the coated substrate had an area of ​​0.79 square centimeters, thus 0.79 x 46.5 cells per centimeter = approximately 36 channels, half of which were "plugged" to account for the "checkerboard" pattern of clogging at the end face of the wall-flow filter. Thus, analysis was performed on at least 18 coated channel cells from the inlet end of the coated filter sample. Because the entire cross section of these cores would not fit into the 3 cc (cubic centimeter) penetrometer cup, analysts used a serrated scalpel similar to a chisel blade to cut the samples into substantial chunks, which were then packed into the cup. The sample chunk retained the appearance of recognizable cell channels at its "open" inlet end, and no further breakage was observed after analysis. Therefore, the sample was representative of the coated portion of the wall-flow filter as a whole. [Fragments of filter substrate without on-wall coating are not representative of the filter and do not exhibit the required volume of 0.05-0.5 μm pores. This is because the pores in such substrates are larger. In particular, the average pore diameter D50 of the substrate is typically 6-15 microns.]

[0149] Porosity data were measured by mercury porosimetry using a Micromeritics AutoPore 9600 according to ASTM Method D4284-03 (Test Method for Determining Pore Volume Distributions of Catalysts by Mercury Intrusion Porosimetry). Intrusion curves were measured over a pressure range of 0.5 to 60,000 psia (pounds per square inch absolute) (3.45 kPa to 414 MPa) and then extruded to atmospheric pressure. A 15-second equilibration time was used for each data point on both the intrusion and extrusion curves, with a mercury contact angle of 130° and a mercury surface tension of 485 dynes / cm.

[0150] Prior to analysis, samples were dried overnight in an oven at 115 °C. Temperature and pressure effects during porosimetry runs were accounted for by performing a blank correction with an empty penetrometer tube, which was then subtracted from the experimental data. Data generated up to 60,000 psia (414 MPa) included the pore volume and pore size distribution of the alumina component of the washcoat. Alumina was unaffected by the addition of pore formers. Therefore, the pore size range of interest for pore formers to affect washcoat porosity was found to be between 0.05 μm and 5 μm in diameter, which falls within the pressure range of 0.5 psia to 1,000 psia (3.45 kPa to 6,890 kPa).

[0151] The Micromeritics AutoPore 9600 instrument has a low-pressure circuit and a high-pressure circuit. The low-pressure circuit uses compressed air to pressurize the sample up to 36 psia. The high-pressure circuit uses an oil-filled chamber. The low-pressure circuit is used to analyze pores down to 300 μm, i.e., the 0.5 psia end of the pressure range referenced in this Example 6. It will be appreciated that because this pressure is below atmospheric pressure (i.e., 14.7 psia), the 0.5 psia end of the range is under high vacuum. It will also be recognized that to analyze the full pressure range of 0.5 to 60,000 psia, the sample / penetrometer must be switched from the low-pressure circuit to the high-pressure circuit. This is because the high-pressure circuit cannot resolve down to the 300 μm end of the analytical range. To ensure a seamless transition of analysis when moving from the low-pressure circuit to the high-pressure circuit, care is taken to ensure that the pressure transducer values ​​in each circuit match when the switch is performed. Once the maximum pressure was reached, the penetrometer was returned to atmospheric pressure and the sample was removed. The 0.05–0.5 μm pore volume as a percentage of the intrusion volume data was calculated from the geometric density derived from the mercury porosimetry intrusion data and the skeletal density measured by helium pycnometry using a Micromeritics AccuPyc II 1340 pycnometer according to ISO 12154:2014—Determination of Density by Volumetric Displacement—Skeletal Density by gas Pycnometry. Samples were dried overnight at 115°C in an oven prior to analysis.

[0152] Six samples were prepared in accordance with the present invention using the methods and materials described for Sample 1 in Example 1. Details of the loading and pore former content are shown in Table 13. The uncoated filter used in each case was a silicon carbide filter of asymmetric design with 41% porosity, an average pore size of 9 μm, 300 cells per square inch, and a wall thickness of 7 thousandths of an inch (mils) (0.1778 mm).

[0153] [Table 13]

[0154] The results are shown in Table 14, including the untreated substrate as a reference.

[0155] [Table 14]

[0156] It can be seen from Table 14 that increasing the amount of pore former at a constant washcoat loading tends to increase the total pore volume, the 0.05-0.5 μm pore volume as a % of mercury intrusion volume, and the % porosity, but if too much pore former is added, the washcoat adhesion will be affected (see the results in Example 5 below). It can also be seen that as the washcoat loading increases at a constant amount of added pore former, the total pore volume, the 0.05-0.5 μm pore volume as a % of mercury intrusion volume, and the % porosity tend to increase.

[0157] Example 7 - Screening of Support Materials - Effect of Bulk Density on SLBP, Filtration Efficiency and Adhesion A series of samples was prepared similarly to Example 1, as described in Table 15, using the cylindrical cordierite wall-flow filter substrate (143.8 mm diameter x 152.4 mm length) used in Example 2. Each filter substrate sample was coated with a washcoat having the coating formulation and axial length described in Example 1 for Sample 1, except that the gamma alumina support material was replaced with an alternative support material as described in Table 15. Prior to adding the pore former, the support materials were wet-milled to the indicated D50 and D90, respectively.

[0158] [Table 15]

[0159] The cold stream backpressure, SLBP, and filtration efficiency of each sample were evaluated by DPG according to the methodology in Example 1, and adhesion testing was performed according to the methodology in Example 5. The results are shown in Table 16. Note that Sample 10.1 from Example 7 and Sample 1 from Example 1 are similarly formulated. However, Sample 10.1 is a freshly prepared sample and therefore different from Sample 1. Therefore, it is expected that there may be some variation in the reported cold stream backpressure in Table 2 and the filtration efficiency in Table 3.

[0160] From the data presented in Table 15, Applicant notes that the D50 and D90 for each of the 10.1-10.5 samples are broadly similar, and therefore any differences in backpressure, filtration efficiency, and adhesion are expected to result from differences in the properties of the raw materials used.

[0161] [Table 16]

[0162] From the results shown in Table 16, it can be seen that despite having good cold flow backpressure and soot deposition backpressure, the CeO2-supported material, Sample 10.4, had poor filtration efficiency and significant adhesion problems. Microscopic examination revealed that the surface coating of 10.4 exhibited cracking, with a dried mud appearance, or so-called "cracked pavement." The cracking could be due to both loss of adhesion and lack of filtration efficiency, although reduced backpressure is also a potential explanation. Applicant attempted to improve adhesion by combining Sample 10.1 with an increasing ratio of CeO2 in 10.4, but the negative results of 10.4 were not overcome (results not shown).

[0163] Sample 10.5 had acceptable soot backpressure and filtration efficiency compared to the Sample 10.1 reference, but unacceptable adhesion loss. However, Applicant found that combining a rare earth-stabilized ZrO2 support material with the gamma alumina support material of Sample 10.1 in a ratio of 80:20 to 60:40 (ZrO2:Al2O3) ultimately resulted in progressively improved adhesion (for the 60:40 embodiment) and met the adhesion requirements (0.6% adhesion loss), and soot backpressure and filtration efficiency were also improved compared to the Sample 10.1 reference (results not shown).

[0164] The higher pore volume 5 wt % silica doped alumina sample (sample 10.3) also has poor adhesion and surprisingly high back pressure.

[0165] It should be noted that, relative to the gamma alumina and 5 wt. % silica-doped alumina samples, i.e., Samples 10.1 and 10.2, the support materials of Samples 10.3, 10.4, and 10.5, by themselves, i.e., without being mixed with gamma alumina, have pre-milled loose bulk densities of less than about 850 g / L and greater than about 275 g / L. While not wishing to be bound by any theory, Applicant believes that there exists a pre-milled loose bulk density parameter range of 275 to 850 g / L from which refractory metal oxide support materials with acceptable backpressure, filtration efficiency, and adhesion for use in the present invention may be advantageously selected. The results shown in Table 16 and discussed above also indicate that materials with loose bulk densities in the range below 275 g / L can be mixed with support materials in the range of 275 to 850 g / L to improve the adhesion characteristics of the former.

[0166] Example 8 - FIB-SEM The samples from Table 2, Examples 1 and 5 were analyzed using a focused ion beam backscattered electron (BSE) scanning electron microscope (FIB-SEM) using a gallium beam and an image "slice" interval of 50 nm, as described in A. Varambhia et al., Johnson Matthey Technol. Rev., 2022, 66, (3), 355-371. Each analyzed sample was cured in resin after applying a high vacuum to ensure that all pores in the sample were filled with resin. The results are shown in Table 17 below and in Figures 3 and 4.

[0167] [Table 17]

[0168] From these data, it can be seen that the combustion of the cellulose pore former during calcination and the evolution of gas from the citric acid in the applied washcoat slurry result in a pore structure within the washcoat layer. It can also be seen that the pore size is not necessarily proportional to the particle size distribution of the dry pore former due to in-situ compression of the washcoat slurry. However, it can be seen from Figures 3 and 4 that the pore former-derived pore size of Sample 1 in Table 4 is larger than the pore size of Sample 5 in Table 4.

[0169] Example 9 - Washcoat Layer Thickness Three 152.4 mm long segment samples of asymmetric silicon carbide filters were coated from the inlet end to 60% of the axial length of the segment and from the outlet end to 40% with washcoats containing those described in Example 1, where the amount of pore former was varied. The inlet channels had a larger hydraulic diameter than the outlet channels.

[0170] It is also well known that silicon carbide filters are typically manufactured as rectangular cross-section segments that are bonded together to form, for example, a cylindrical overall filter, because silicon carbide itself has a relatively high coefficient of thermal expansion (CTE), which could lead to catastrophic stress-induced crack formation within the canned filter during use if a segmented arrangement were not used.

[0171] The amount of pore former used for each sample is shown in Table 18 below. Backscattered electron (BSE)-SEM image of a cross section of a channel in a coated sample part. 0.4 g / in 3 (24.4 g / L) for lumber (not according to the present invention) and 0.2 g / in 3 (12.2 g / L) of pore former at a point 50.8 mm inside the coated segment in an axial direction (i.e., in both the smaller and larger hydraulic diameter channels) from the inlet of the segment coated with the washcoat containing 0.3 g / in 3 SEM images were taken 50.8 mm from the outlet for the sample loaded with 18.3 g / L of pore-forming agent. The thickness of the washcoat coating was measured at the angle bisecting the vertex angle between the two edges of the square cross section of the channel, representing the orthogonally juxtaposed channel wall surfaces, as shown in Figure 1. Thickness measurements were taken from five cells per location within each part, with four corner measurements per cell / channel.

[0172] [Table 18]

[0173] From the results shown in Table 18, it can be seen that as the loading of pore former increases, the thickness of the coating increases. From Applicant's WO 2015 / 082892, it is known that on-wall coatings are thicker on channels of asymmetric filter substrates with smaller hydraulic diameters than on channels with larger hydraulic diameters. However, the thickest washcoat loading (0.4 g / in), corresponding to Examples 8.3-8.6 of Reference Example 4, was 0.4 g / in. 3 (24.4 g / L)) generally indicates problems with too high soot deposition backpressure or too low filtration efficiency and / or adhesion (see Example 5 herein above).

[0174] For the avoidance of any doubt, the entire contents of any and all documents cited herein are incorporated by reference into this application.

Claims

1. 1. A method of making a ceramic honeycomb wall-flow filter substrate having an axial length L supporting an on-wall coating, wherein a section of the on-wall coated filter substrate analyzed by mercury porosimetry has a pore volume of 0.05 to 0.5 μm diameter as a percentage of the mercury intrusion volume of said section of at least 15.0%, the wall-flow filter substrate having a first end and a second end and comprising an array of axially extending first and second channels, each of the first and second channels having an axially extending length L. wherein each first channel shares a porous channel wall with an adjacent second channel, and in a scanning electron microscope image of a cross section of the coated substrate, each channel has a geometric hollow cross-sectional shape with at least one vertex defined in part by an apex angle between two channel surface edges, the first channel is end-capped at the second end of the substrate, and the second channel is end-capped at the first end of the substrate; and the method comprises: (i) preparing a slurry comprising water, carboxylic acid, inorganic oxide particles or a mixture of two or more inorganic oxide particles having a D(v,0.9) of 8 to 20 μm as determined by laser diffraction particle size analysis and 10 to 35 wt. % of a particulate insoluble cellulose pore former of heterogeneous morphology based on 100 wt. % of inorganic oxide particles, and an equivalent spherical diameter D(v,0.5) of 0.5 to 14 μm as determined by laser diffraction and a modal aspect ratio (width / length) of 0.3 to 0.9 as determined by direct particle measurement using a Flow Imaging Microscopy particle shape analyzer; (ii) coating the slurry onto at least the porous channel walls of the first channels of the substrate, wherein the substrate has an average pore size (D50) of 6 to 15 μm before any coating and a porosity of less than 60% before any coating; (iii) drying and firing the slurry-coated substrate; The product of step (iii) is 0.07 to 0.4 g / in based on the weight of the substrate before step (ii). ３ wherein sufficient slurry is coated onto the substrate to have a coating loading of (4.3 to 24.4 g / L) and the product of step (iii) has an average on-wall coating thickness of 5 to 70 μm as measured at the angle bisecting the vertex angle on a scanning electron microscope cross-sectional image of the coated substrate.

2. 2. The method of claim 1, wherein step (ii) comprises coating the porous channel walls of the second channels of the filter substrate with a slurry according to step (i), and wherein the slurry for coating the porous channel walls of the second channels has the same or a different composition as the slurry coated on the porous channel walls of the first channels.

3. 3. The method of claim 1 or 2, wherein the inorganic oxide particles or components of the mixture of inorganic oxide particles have a loose bulk density before milling of 275 to 850 g / L.

4. The method of any one of claims 1 to 3, wherein the first axially extending channels of the wall-flow filter substrate have a larger hydraulic diameter than the second channels.

5. 4. The method of claim 1, 2, or 3, wherein the axially extending first channels of the wall-flow filter substrate are a combination of channels having larger and smaller hydraulic diameters, and the hydraulic diameter of the second channels is larger than the first channels having smaller hydraulic diameters.

6. the on-wall coating is an on-wall catalytic coating, and the slurry of step (i) has a viscosity of 0.5 to 10 g / ft of the product of step (iii). ３ 6. The method of any one of claims 1 to 5, comprising one or more non-rhodium platinum group metal salts at a platinum group metal salt concentration selected to have a total platinum group metal loading of (0.018 to 0.353 g / L).

7. The method of any one of claims 1 to 6, wherein the filter substrate is a silicon carbide filter substrate having a porosity of 38 to 45% before coating.

8. 8. The method of claim 7, wherein the coated filter substrate has a porosity of less than 45.0%.

9. The coated filter substrate has a thickness of at least 0.230 cm as determined by a combination of helium pycnometry and mercury densitometry. ３ The method according to claim 7 or 8, wherein the total pore volume is 1 / g.

10. The method of any one of claims 1 to 6, wherein the filter substrate is a cordierite filter substrate having a porosity of 46 to 60% before coating.

11. 1. An on-wall coated ceramic honeycomb wall-flow filter substrate for treating particulate matter-laden exhaust gases emitted from an internal combustion engine, the wall-flow filter substrate comprising an array of axially extending first and second channels having first and second ends and a length L therebetween, each of the first and second channels being defined in part by a porous channel wall having an axially extending porous channel wall surface, each first channel sharing a porous channel wall with an adjacent second channel, and in a scanning electron microscope image of a cross section of the coated substrate, each channel has a geometric hollow cross-sectional shape with at least one vertex defined in part by an apex angle between two channel surface edges, the first channels being end-plugged at the second end of the substrate, and the second channels being end-plugged at the first end of the substrate, and at least the channel wall surface of the first channels having a density of 0.07 to 0.4 g / in based on the substrate weight before coating. ３ and a coating thickness of 4.3 to 24.4 g / L on the surface of the filter substrate, the coating thickness being measured at the angle bisecting the apex angle on a cross-sectional scanning electron microscope image of the coated substrate, the average on-wall coating thickness being 5 to 70 μm, the cross-sectional scanning electron microscope image of the coated substrate being measured at the angle bisecting the apex angle on a cross-sectional scanning electron microscope image of the coated substrate, the cross-sectional scanning electron microscope image of the coated substrate being measured at the angle bisecting the apex angle on the surface of the filter ... An on-wall coated ceramic honeycomb wall-flow filter substrate containing geometrically non-uniform voids with equivalent spherical diameters D(v,0.5) of 0.1-8 μm and modal aspect ratios (width / length) of 0.3-0.9 as determined by microscopy (FIB-SEM).

12. A ceramic honeycomb wall-flow filter substrate carrying an on-wall coating for treating exhaust gases containing particulate matter emitted from an internal combustion engine, obtainable or obtained by the process of any one of claims 1 to 10.

13. 13. An exhaust system for an internal combustion engine comprising the ceramic honeycomb wall-flow filter substrate of claim 11 or 12, wherein the first end of the substrate is oriented upstream.

14. An internal combustion engine comprising the exhaust system of claim 13.

Images