A THREE-WAY CONVERSION CATALYST IN GASOLINE-NATURAL GAS APPLICATIONS

Abstract

The disclosure relates to a three-way conversion catalyst for the treatment of an exhaust gas comprising nitrogen monoxide, carbon monoxide, and hydrocarbon, wherein the catalyst comprises: (i) a substrate; (ii) a first coating comprising rhodium supported on a first oxidic component; (iii) a second coating comprising palladium supported on a non-zeolitic oxidic material, wherein the non-zeolitic oxidic material comprises manganese and a second oxidic component, wherein the second coating consists of 0 weight-% to 0.001 weight-% of platinum; wherein the first coating is disposed on the substrate over x % of the axial length, with x ranging from 80 to 100; wherein the second coating extends over y % of the axial length from the inlet end to the outlet end and is disposed on the first coating, with y ranging from 20 to x.

Claims

1-15. (canceled)

16. A three-way conversion catalyst for the treatment of an exhaust gas comprising nitrogen monoxide, carbon monoxide and hydrocarbon, wherein the catalyst comprises: a substrate comprising an inlet end, an outlet end, a substrate axial length extending from the inlet end to the outlet end and a plurality of passages defined by internal walls of the substrate extending therethrough; (ii) a first coating comprising rhodium supported on a first oxidic component; and (iii) a second coating comprising palladium supported on a non-zeolitic oxidic material, wherein the non-zeolitic oxidic material comprises manganese and a second oxidic component, wherein from 0 weight-% to 0.001 weight-% of the second coating consists of platinum; wherein the first coating is disposed on the surface of the internal walls of the substrate over x % of the substrate axial length and x ranging from 80 to 100; wherein the second coating extends over y % of the substrate axial length from the inlet end to the outlet end and is disposed on the first coating and y ranging from 20 to x.

17. The three-way conversion catalyst of claim 16, wherein x ranges from 90 to 100 and wherein y ranges from 80 to x.

18. The three-way conversion catalyst of claim 16, wherein the first oxidic component, in the first coating, is chosen from alumina, ceria, silica, zirconia, titania, a mixture of two or more thereof, and a mixed oxide of two or more thereof.

19. The three-way conversion catalyst of claim 16, wherein the first coating further comprises a platinum group metal other than rhodium.

20. The three-way conversion catalyst of claim 19, wherein the platinum group metal other than rhodium is supported on a third oxidic component.

21. The three-way conversion catalyst of claim 20, wherein the platinum group metal other than rhodium is further supported on a first oxygen storage compound.

22. The three-way conversion catalyst of claim 16, wherein the first coating consists of from 0 weight-% to 0.001 weight-% of manganese.

23. The three-way conversion catalyst of claim 16, wherein the second coating comprises manganese an amount ranging from 1 weight-% to 10 weight-%, calculated as MnO.sub.2 and based on the weight of the non-zeolitic oxidic material.

24. The three-way conversion catalyst of claim 16, wherein the second oxidic component, in the non-zeolitic oxidic material of the second coating, is chosen from alumina, silica, ceria, zirconia, titania, a mixture of two or more thereof, and a mixed oxide of two or more thereof.

25. The three-way conversion catalyst of claim 16, wherein the second coating comprises palladium, calculated as elemental palladium, at a loading ranging from 20 g/ft.sup.3 to 200 g/ft.sup.3.

26. The three-way conversion catalyst of claim 16, wherein the second coating further comprises a second oxygen storage compound.

27. The three-way conversion catalyst of claim 16, wherein the substrate according to (i) is a flow-through substrate.

28. A process for preparing a three-way conversion catalyst comprising: (a) providing a substrate comprising an inlet end, an outlet end, a substrate axial length extending from the inlet end to the outlet end and a plurality of passages defined by internal walls of the substrate extending therethrough; (b) preparing a slurry comprising rhodium, a first oxidic component and water; disposing the slurry on the surface of the internal walls of the substrate over x % of the substrate axial length, from the inlet end to the outlet end of the substrate, or from the outlet end to the inlet end of the substrate, with x ranging from 80 to 100; calcining the obtained slurry-treated substrate, obtaining a substrate coated with a first coating; (c) preparing a slurry comprising water, palladium and a non-zeolitic oxidic material, comprising manganese and a second oxidic component; disposing the slurry on the first coating of the coated substrate obtained in (b) over y % of the substrate axial length from the inlet end to the outlet end of the substrate, with y ranging from 20 to x; calcining the obtained slurry-treated substrate, obtaining a substrate coated with a first coating and a second coating, wherein the second coating consist of 0 weight-% to 0.001 weight-% of platinum.

29. An exhaust gas treatment system downstream of and in fluid communication with an engine, the system comprising a three-way conversion catalyst according to claim 16.

30. A method of treating an exhaust gas comprising hydrocarbon, carbon monoxide, and nitrogen monoxide from a gasoline engine or a combined gasoline-natural gas engine comprising using the three-way conversion catalyst according to claim 16 for the treatment.

Description

EXAMPLES

Reference Example 1: Determination of BET Specific Surface Area of Alumina

[0331] The BET specific surface area of the alumina was determined according to DIN 66131 or DIN-ISO 9277 using liquid nitrogen.

Reference Example 2: Determination of the Volume-Based Particle Size Distributions

[0332] The particle size distributions were determined by a static light scattering method using a Sympatec HELOS/BR-OM & QUIXEL wet dispersion equipment, fitted with laser (HeNe) diffraction sensor with 31 channel multielement detection range comprising 5 modules covering 0.1-875 microns.

Reference Example 3: General Coating Method

[0333] In order to coat a flow-through substrate with a slurry, the flow-through substrate was immersed vertically in a portion of a given slurry for a specific length of the substrate resulting in the targeted length of the coating to be applied. In this manner, the washcoat contacted the walls of the substrate. The sample was left in the slurry for a specific period of time, usually for 1-15 seconds. The substrate was then removed from the slurry, rotated 180° about its vertical axis and shaken gently at first and vigorously later to rid it of excess slurry, followed by blowing with compressed air (in the direction of initial slurry uptake).

Comparative Example 1: Preparation of a Catalyst not According to the Present Invention (56 g/ft.SUP.3 .of Pd—no Manganese)

[0334] An alumina powder (Al.sub.2O.sub.3: about 100 weight-%, having a BET specific surface area of about 150 m.sup.2/g, a Dv50 of 35 micrometers, a mean pore radius of 11 nm and a total pore volume of 0.9 ml/g) was impregnated with an aqueous mixture of deionized water and a palladium nitrate, such that the incipient point may not be exceeded. The amount of alumina was calculated such that the alumina loading in the catalyst after calcination was of 1.40 g/in.sup.3. The obtained mixture of Pd-alumina (solid content: 65 weight-%) was calcined in a calciner at 590° C. for 2 hours (thermal fixation).

[0335] A mixture was prepared with distilled water, n-octanol (0.3 weight-% based on the weight of the washcoat after calcination) and barium nitrate. The amount of barium nitrate was calculated such that the final loading of BaO in the catalyst after calcination was 0.068 g/in.sup.3. The amount of octanol was calculated such that it was 0.3 weight-% of the final washcoat loading Said mixture was stirred for approximately 10 minutes in a container at room temperature.

[0336] The calcined Pd-alumina powder was added slowly to the obtained mixture while stirring to obtain a slurry. The solid content of the slurry was set to about 40 weight-%. After an initial pH adjustment with nitric acid to 3.8, the slurry was milled until the particles of the slurry had a Dv90, determined as in Reference Example 2, of 12 micrometers with a final pH adjustment to 3.5. The obtained slurry was disposed over the full length of an uncoated flow-through cordierite honeycomb substrate (diameter: 2.54 cm (1 inch)×length: 10.16 cm (4 inches)), cylindrical shaped substrate with 400/(2.54).sup.2 cells per square centimeter and 0.1 millimeter (4 mil) wall thickness) using the coating method described in Reference Example 3. Afterwards, the coated substrate was dried at 140° C. in air for 30 minutes and was calcined at 590° C. in air for 2 hours.

[0337] The washcoat loading in the catalyst after calcination was 1.50 g/in.sup.3, including 1.40 g/in.sup.3 of alumina, 0.068 g/in.sup.3 BaO and 56 g/ft.sup.3 of Pd.

Comparative Example 2: Preparation of a Catalyst not According to the Present Invention (112 g/ft.SUP.3 .of Pd—no Manganese)

[0338] The catalyst of Comparative Example 2 was prepared as the catalyst of Comparative Example 1, except that the washcoat loading in the catalyst after calcination was 1.53 g/in.sup.3, including 1.40 g/in.sup.3 of alumina, 0.068 g/in.sup.3 BaO and 112 g/ft.sup.3 of Pd.

Comparative Example 3: Preparation of a Catalyst not According to the Present Invention (167 g/ft.SUP.3 .of Pd—no Manganese)

[0339] The catalyst of Comparative Example 3 was prepared as the catalyst of Comparative Example 1, except that the washcoat loading in the catalyst after calcination was 1.56 g/in.sup.3, including 1.40 g/in.sup.3 of alumina, 0.068 g/in.sup.3 BaO and 167 g/ft.sup.3 of Pd.

Example 1: Preparation of a Catalyst (56 g/ft.SUP.3 .of Pd—Manganese)

[0340] The catalyst of Example 1 was prepared as the catalyst of Comparative Example 1, except that the alumina powder was replaced by an Mn-alumina powder (95 weight-% of Al.sub.2O.sub.3, 5 weight-% of MnO.sub.2, having a BET specific surface area of about 132 m.sup.2/g, a Dv50 of 37.5 micrometers, a mean pore radius of 11.5 nm and a total pore volume of 0.8 ml/g). The washcoat loading in the catalyst after calcination was 1.50 g/in.sup.3, including 1.40 g/in.sup.3 of Mn-alumina, 0.068 g/in.sup.3 BaO and 56 g/ft.sup.3 of Pd.

Example 2: Performance Evaluation of the Catalysts of Example 1 and of Comparative Examples 1-3—CO, NO and HC Conversions

[0341] All catalysts were aged together in one oven at 900° C. hydrothermally (10% O.sub.2 and 10% steam) for 4 hours. Oven aging was done in an oven equipped with several gas lines for simultaneous dosage of several gases under controlled flow conditions.

[0342] All aged samples were evaluated individually, one at a time using a Gasoline System Simulator (GSS) reactor operated using the New European Drive Cycle (NEDC) test cycle implemented from a real vehicle in the Compressed Natural Gas (CNG) mode. To evaluate the impact of hydrothermal aging on the various technologies, the same samples were evaluated in the fresh state on the same reactor prior to the oven-aging. The results are displayed in FIG. 1.

[0343] CO conversion: As may be taken from FIG. 1, the catalyst of Example 1 exhibits CO conversions of about 84.5% (fresh) and of about 80% (aged) while the catalyst of Comparative Example 1 exhibits lower CO conversions of about 76% (fresh) and of about 71% (aged). This shows that the catalyst of Example 1 exhibits improved CO conversions under fresh and aged conditions compared to a catalyst with the same Pd loading but free of manganese. Further, the use of the catalyst of Example 1 enhances CO conversion improvement after ageing. Furthermore, the catalyst of Example 1 exhibits a CO conversion comparable to the catalyst of Comparative Example 2 which contains twice the palladium loading of Example 1 under fresh conditions. The trend is seemingly the same after oven-ageing, with improvements for the manganese containing catalyst. Finally, the catalyst of Example 1 exhibits CO conversions comparable to the catalyst of Comparative Example 3 which contains three times the palladium loading of Example 1 under fresh and aged conditions.

[0344] NO conversion: As may be taken from FIG. 1, the catalyst of Example 1 exhibits NO conversions of about 96% (fresh) and of about 87% (aged) while the catalyst of Comparative Example 1 exhibits NO conversions of about 87% (fresh) and of about 81% (aged). This shows that the catalyst of Example 1 exhibits improved NO conversion in fresh and aged states compared to a catalyst with the same Pd loading but free of manganese. Furthermore, the catalyst of Example 1 exhibits improved NO conversions compared to those obtained with the catalysts of Comparative Examples 2 and 3 under fresh conditions and comparable or slightly lower NO conversions compared to those obtained with the catalysts of Comparative Examples 2 and 3 after ageing.

[0345] HC conversion: As may be taken from FIG. 1, the catalyst of Example 1 exhibits HC conversions of about 83% (fresh) and of about 60% (aged). The HC conversion under fresh conditions obtained with the catalyst of Example 1 is comparable to those of Comparative Example 1. The HC conversions obtained with the catalysts of Comparative Examples 2 and 3 are higher than those obtained with the catalyst of Example 1. Without wanting to be bound to any theory, it may be explained by the fact that HC conversions are directly linked with the platinum group metal (PGM) amount. This is confirmed by the results outlined in Example 4 below.

[0346] Example 2 demonstrates that the catalyst according to the present invention offers improved CO and NO conversions while maintaining competitive HC conversion activity under fresh and aged conditions. This example further shows that by using alumina with manganese oxide (MnO.sub.2) disposed thereon, it is possible to reduce the amount of a platinum group metal, in particular palladium, in a catalyst to obtain similar, or even improved, catalytic activities, especially CO and NOx.

Comparative Example 4: Preparation of a Catalyst not According to the Present Invention (112 g/ft.SUP.3 .Pt—Manganese)

[0347] An Mn-alumina powder (95 weight-% of Al.sub.2O.sub.3, 5 weight-% of MnO.sub.2, having a BET specific surface area of about 132 m.sup.2/g, a Dv50 of 37.5 micrometers, a mean pore radius of 11.2 nm and a total pore volume of 0.8 ml/g) was impregnated with an aqueous mixture of a platinum nitrate and deionized water. The amount of Mn-alumina was calculated such that the Mn-alumina loading in the catalyst after calcination was of 1.40 g/in.sup.3. The obtained mixture of Pt-alumina (solid content: 65 weight-%) was calcined in a calciner at 590° C. for 2 hours (thermal fixation).

[0348] A mixture was prepared with distilled water, n-octanol (0.3 weight-% based on the weight of the washcoat after calcination) and barium nitrate. The amount of barium nitrate was calculated such that the final loading of BaO in the catalyst after calcination was 0.068 g/in.sup.3. Said mixture was stirred for approximately 10 minutes in a container at room temperature.

[0349] The calcined Pt-alumina powder was added slowly to the obtained mixture while stirring in order to obtain a slurry. The solid content of the slurry was set to about 40 weight-%. After an initial pH adjustment with nitric acid to 3.8, the slurry was milled until the particles of the slurry had a Dv90, determined as in Reference Example 2, of 12 micrometers with a final pH adjustment to 3.5. The obtained slurry was disposed over the full length of an uncoated flow-through cordierite honeycomb substrate (diameter: 2.54 cm (1 inch)×length: 10.16 cm (4 inches)), cylindrical shaped substrate with 600/(2.54).sup.2 cells per square centimeter and 0.1 millimeter (4 mil) wall thickness) using the coating method described in Reference Example 3. Afterwards, the coated substrate was dried at 140° C. in air for 20 minutes and was calcined at 590° C. in air for 2 hours.

[0350] The washcoat loading in the catalyst after calcination was 1.53 g/in.sup.3, including 1.40 g/in.sup.3 of Mn-alumina, 0.068 g/in.sup.3 BaO and 112 g/ft.sup.3 of Pt.

Comparative Example 5: Preparation of a Catalyst not According to the Present Invention (112 g/ft.SUP.3 .Pt—no Manganese)

[0351] The catalyst of Comparative Example 5 was prepared as the catalyst of Comparative Example 4, except that the Mn-alumina powder was replaced by the alumina powder used in Comparative Example 1. The washcoat loading in the catalyst after calcination was 1.53 g/in.sup.3, including 1.40 g/in.sup.3 of alumina, 0.068 g/in.sup.3 BaO and 112 g/ft.sup.3 of Pt.

Comparative Example 6: Preparation of a Catalyst not According to the Present Invention (112 g/ft.SUP.3 .Pd—no Manganese)

[0352] The catalyst of Comparative Example 6 was prepared as the catalyst of Comparative Example 5, except that platinum nitrate was replaced by palladium nitrate. The washcoat loading in the catalyst after calcination was 1.53 g/in.sup.3, including 1.40 g/in.sup.3 of alumina, 0.068 g/in.sup.3 BaO and 112 g/ft.sup.3 of Pd.

Example 3: Preparation of a Catalyst (112 g/ft.SUP.3 .Pd—Manganese)

[0353] The catalyst of Example 3 was prepared as the catalyst of Comparative Example 4, except that platinum nitrate was replaced by palladium nitrate. The washcoat loading in the catalyst after calcination was 1.53 g/in.sup.3, including 1.40 g/in.sup.3 of Mn-alumina, 0.068 g/in.sup.3 BaO and 112 g/ft.sup.3 of Pd.

Example 4: Performance Evaluation of the Catalysts of Example 3 and of Comparative Examples 4, 5 and 6—CO, NO and HC Conversions

[0354] All catalysts were aged together in one oven at 900° C. hydrothermally (10% O.sub.2 and 10% steam) for 4 hours. Oven aging was done in an oven equipped with several gas lines for simultaneous dosage of several gases under controlled flow conditions. All aged samples were evaluated individually, one at a time using a Gasoline System Simulator (GSS) reactor operated using the New European Drive Cycle (NEDC) test cycle implemented from a real vehicle in the Compressed Natural Gas (CNG) mode. The results are displayed in FIG. 2.

a) Example 3 vs Comparative Examples 4 and 5

[0355] As may be taken from FIG. 2, the catalyst of Example 3 exhibits a HC conversion of about 69.5%, a CO conversion of about 83% and a NO conversion of about 92% while the catalyst of Comparative Example 4 exhibits a lower HC conversion of 33% and CO and NO conversions like those of Example 3. Further, the catalysts of Example 3 exhibit improved HC, CO and NO conversions compared to the catalyst of Comparative Example 5. Example 4 demonstrates that the catalyst of the present invention comprising palladium and manganese permits to obtain improved catalytic activities compared to a catalyst comprising platinum with, or without, manganese. This shows that palladium is mandatory in such catalyst formulations. Thus, the catalyst of the present invention permits to obtain improved HC conversion, which is particularly important in Compressed Natural Gas (CNG) applications, while exhibiting great NO and CO conversions under aged conditions. Essentially, the catalyst of the present invention shows superior three-way conversion catalyst performance (HC, CO, NOx).

b) Example 3 vs Comparative Example 6

[0356] As may be taken from FIG. 2, the catalyst of Comparative Example 6 exhibits lower HC, NO and CO conversions compared to the catalyst of Example 3. Example 4 further demonstrates that the catalyst of the present invention comprising palladium and manganese permits obtain improved catalytic activities compared to a catalyst comprising palladium but no manganese. Thus, from said example, there is an indication that palladium and manganese have a synergistic effect on the catalytic activities of a catalyst for the treatment of exhaust gas with respect to HC, CO and NO oxidations.

Comparative Example 7: Preparation of a Three-Way Conversion Catalyst not According to the Present Invention (Free of Manganese)

First Coating

[0357] A calcined powder of Pd (70 weight-% of the total weight of palladium in the first coating after calcination) of an oxygen storage compound comprising Ce (30 weight-% calculated as CeO.sub.2) and Zr (60 weight-% calculated as ZrO.sub.2) and further comprising lanthanum and yttrium (5 weight-% each calculated as X.sub.2O.sub.3) was added slowly under stirring to a container already filled with distilled water, n-octanol (0.3 weight-% based on the weight of the first coating after calcination), barium nitrate and zirconium nitrate (previously stirred for 10 minutes). The obtained slurry was stirred for 10 minutes followed by a pH adjustment to 3.8 and had a solid content of 40 weight-%. The obtained slurry was wet-milled until the particles of the slurry has a Dv90, as determined in Reference Example 2, of 7 micrometers.

[0358] Similarly, a first calcined powder of Pd (30 weight-% of the total weight of palladium in the first coating after calcination) on alumina (Al.sub.2O.sub.3: about 98.7 weight-%, having a BET specific surface area of about 145 m.sup.2/g, a Dv50 of 7.2 micrometers and a total pore volume of 0.537 ml/g) and a second calcined powder of Rh (100 weight-% of the total weight of rhodium in the first coating after calcination) on alumina (Al.sub.2O.sub.3: about 98.7 weight-%, having a BET specific surface area of about 145 m.sup.2/g, a Dv50 of 7.2 micrometers and a total pore volume of 0.537 ml/g) were stirred into a container with distilled water, n-octanol (0.3 weight-% based on the weight of the first coating after calcination), barium nitrate and zirconium nitrate. The obtained slurry was milled until the particles of the slurry had a Dv90, as determined in Reference Example 2, of 15 micrometers. The pH of the aqueous phase of the slurry was controlled prior to and after milling and was adjusted to 3.8 using nitric acid when necessary. The amount of barium nitrate in each slurry was calculated such that the final loading of BaO in the first coating of the catalyst after calcination was 0.068 g/in.sup.3 and the amount of zirconium nitrate in each slurry was calculated such that the final loading of ZrO.sub.2 in the first coating of the catalyst was 0.021 g/in.sup.3.

[0359] The two slurries were finally blended into a final slurry. The final slurry had a solid content of 38 weight-% and was stirred for 10 minutes. The final slurry was milled until the particles of the slurry had a Dv90, as determined in Reference Example 2, of 12 micrometers. The pH of the aqueous phase of the slurry was controlled and adjusted to 3.5 with nitric acid. The obtained slurry was disposed over the full length of an uncoated flow-through cordierite honeycomb substrate (diameter: 11.84 cm (4.66 inches)×length: 11.43 cm (4.5 inches), cylindrical shaped substrate with 600/(2.54).sup.2 cells per square centimeter and 0.1 millimeter (4 mil) wall thickness) using the coating method described in Reference Example 3. Afterwards, the coated substrate was dried at 140° C. in air for 30 minutes and was calcined at 590° C. in air for 2 hours. The first coating had a washcoat loading in the catalyst after calcination of 1.63 g/in.sup.3, including 0.825 g/in.sup.3 of alumina, 0.680 g/in.sup.3 of ceria-zirconia, 0.068 g/in.sup.3 of BaO, 0.021 g/in.sup.3 of ZrO.sub.2, 54 g/ft.sup.3 of Pd and 10 g/ft.sup.3 of Rh.

Second Coating

[0360] An alumina powder (Al.sub.2O.sub.3: about 100 weight-%, having a BET specific surface area of about 150 m.sup.2/g, a Dv50 of 35 micrometers, a mean pore radius of 11 nm and a total pore volume of 0.9 ml/g) was impregnated with an aqueous mixture of deionized water and a palladium nitrate. The amount of alumina was calculated such that the alumina loading (in the second coating) in the catalyst after calcination was of 0.6 g/in.sup.3. The frit of Pd (50 weight-% of the total weight of palladium in the second coating after calcination) on alumina (solid content: 65 weight-%) was calcined in a calciner at 590° C. for 2 hours (thermal fixation). Similarly, the remaining palladium (50 weight-% of the total weight of palladium in the second coating after calcination) was impregnated on an oxygen storage compound comprising Ce (30 weight-% calculated as CeO.sub.2) and Zr (60 weight-% calculated as ZrO.sub.2) and further comprising lanthanum and yttrium (5 weight-% each calculated as X.sub.2O.sub.3) and was calcined at 590° C. for 2 hours (thermally fixation).

[0361] The calcined Pd/ceria-zirconia was added slowly under stirring to a container already filled with distilled water, n-octanol (0.3 weight-% based on the weight of the first coating after calcination), and barium nitrate (previously stirred for 10 minutes). The obtained slurry was stirred for 10 minutes and had a solid content of 40 weight-%. After a pH adjustment to 3.8 with nitric acid, the obtained slurry was milled until the particles of the slurry had a Dv90, determined as defined in Reference Example 2, of 7 micrometers.

[0362] Similarly, the calcined Pd-alumina frit was added slowly to a container already filled with distilled water, n-octanol (0.3 weight-% based on the weight of the first coating after calcination), and barium nitrate (previously stirred for 10 minutes). After an initial pH adjustment with nitric acid to 3.8, the obtained slurry was milled until the particles of the slurry had a Dv90, determined as in Reference Example 2, of 15 micrometers with a final pH adjustment to 3.8 using nitric acid if necessary. The amount of barium nitrate in each slurry was calculated such that the final loading of BaO in the second coating of the catalyst after calcination was 0.12 g/in.sup.3. The two slurries were finally blended into a final slurry. The final slurry was stirred for 10 minutes and had a solid content of 38 weight-%. The final slurry was milled until the particles of the slurry had a Dv90, determined as in Reference Example 2, of 12 micrometers. The pH of the aqueous phase of the slurry was controlled and adjusted to 3.5 using nitric acid. The obtained slurry was disposed over the full length of the substrate coated with the first coating using the coating method described in Reference Example 3. Afterwards, the coated substrate was dried at 140° C. in air for 30 minutes and was calcined at 590° C. in air for 2 hours. The second coating had a washcoat loading in the catalyst after calcination of 1.55 g/in.sup.3, including 0.8 g/in.sup.3 of ceria-zirconia, 0.6 g/in.sup.3 of alumina, 0.12 g/in.sup.3 of BaO and 54 g/ft.sup.3 of Pd. The obtained catalyst after calcination had a total platinum group metal loading of 118 g/ft.sup.3, including 108 g/ft.sup.3 of Pd and 10 g/ft.sup.3 of Rh. The total washcoat loading of the catalyst after calcination was about 3.2 g/in.sup.3.

Example 5: Preparation of a Three-Way Conversion Catalyst (5 weight-% of MnO.SUB.2.)

First Coating

[0363] This coating was prepared and coated as the first coating of Comparative Example 7. The first coating had a washcoat loading in the catalyst after calcination of 1.63 g/in.sup.3, including 0.825 g/in.sup.3 of alumina, 0.680 of ceria-zirconia, 0.068 g/in.sup.3 of BaO, 0.021 of ZrO.sub.2, 54 g/ft.sup.3 of Pd and 10 g/ft.sup.3 of Rh.

Second Coating

[0364] The second coating was prepared as the second coating of Comparative Example 7, except that the alumina powder from Pd on alumina was replaced by an Mn-alumina powder (95 weight-% of Al.sub.2O.sub.3, 5 weight-% of MnO.sub.2 having a BET specific surface area of about 132 m.sup.2/g, a Dv50 of 37.5 micrometers, a mean pore radius of 11.2 nm and a total pore volume of 0.8 ml/g). The obtained slurry was disposed over the full length of the substrate coated with the first coating using the coating method described in Reference Example 3. Afterwards, the coated substrate was dried at 140° C. in air for 30 minutes and was calcined at 590° C. in air for 2 hours. The second coating had a washcoat loading in the catalyst after calcination of 1.55 g/in.sup.3, including 0.8 g/in.sup.3 of ceria-zirconia, 0.6 g/in.sup.3 of Mn-alumina, 0.12 g/in.sup.3 of BaO and 54 g/ft.sup.3 of Pd. The obtained catalyst after calcination had a total platinum group metal loading of 118 g/ft.sup.3, including 108 g/ft.sup.3 of Pd and 10 g/ft.sup.3 of Rh. The total washcoat loading of the catalyst after calcination was about 3.2 g/in.sup.3.

Example 6: Preparation of a Three-Way Conversion Catalyst (8 Weight-% of MnO.SUB.2.)

First Coating

[0365] This coating was prepared and coated as the first coating of Comparative Example 7. The first coating had a washcoat loading in the catalyst after calcination of 1.63 g/in.sup.3, including 0.825 g/in.sup.3 of alumina, 0.680 g/in.sup.3 of ceria-zirconia, 0.068 g/in.sup.3 of BaO, 0.021 g/in.sup.3 of ZrO.sub.2, 54 g/ft.sup.3 of Pd and 10 g/ft.sup.3 of Rh.

Second Coating

[0366] The second coating was prepared as the second coating of Example 5, except that the content of MnO.sub.2 in the Mn-alumina powder was of 8 weight.-% based on the total weight of Mn-alumina). The obtained slurry was disposed over the full length of the substrate coated with the first coating using the coating method described in Reference Example 3. Afterwards, the coated substrate was dried at 140° C. in air for 30 minutes and was calcined at 590° C. in air for 2 hours. The second coating had a washcoat loading in the catalyst after calcination of 1.55 g/in.sup.3, including 0.8 g/in.sup.3 of ceria-zirconia, 0.6 g/in.sup.3 of Mn-alumina, 0.12 g/in.sup.3 of BaO and 54 g/ft.sup.3 of Pd. The obtained catalyst after calcination had a total platinum group metal loading of 118 g/ft.sup.3, including 108 g/ft.sup.3 of Pd and 10 g/ft.sup.3 of Rh. The total washcoat loading of the catalyst after calcination was about 3.2 g/in.sup.3.

Example 7: Performance Evaluation of the Catalyst of Examples 5 and 6 and of Comparative Example 7—CO Conversion

[0367] The catalysts of Examples 5 and 6 and of Comparative Example 7 were 100 h fuel-cut aged in with 850° C. inlet temperature according to ZDAKW ageing cycle. The performance of the aged catalysts was measured under world light duty test cycle (WLTC) on 1.4 l dual mode gasoline-CNG vehicle on a chassis-dyno in the CNG mode. The results are displayed on FIG. 3. As shown in FIG. 3, the WLTC results of aged (ZDAKW, 850° C., 100 hours, engine) catalysts indicate that the catalysts of Examples 5 and 6 are superior in CO conversion compared to the catalyst of Comparative Example 7 throughout the entire cycle, starting from the first acceleration. This observation is suggestive of an earlier light-off for the catalysts of Examples 5 and 6, respectively. Furthermore, the catalyst of Example 6 is seemingly more receptive to the catalyst of Example 5 during the high speed high flow segment of the cycle, with the latter slipping more CO, which indicates that more manganese is beneficial. HC and NOx conversion during the same cycle were also recorded for the afore-mentioned catalysts, with the results showing that all three formulations perform similarly.

BRIEF DESCRIPTION OF THE FIGURES

[0368] FIG. 1 shows the HC, CO and NO conversions obtained with the catalysts of Example 1 and of Comparative Examples 1-3, fresh and after oven-aging.

[0369] FIG. 2 shows the HC, CO and NO conversions obtained with the catalysts of Example 3 and of Comparative Examples 4-6 after oven-aging.

[0370] FIG. 3 shows the CO conversion obtained from the catalysts of Examples 5 and 6 and of Comparative Example 7, after engine-ageing.

CITED LITERATURE

[0371] WO 2015/09058A1 [0372] US2015/202572 A1 [0373] US2015/202600 A1 [0374] US2015/202611 A1