THREE-WAY-CATALYST

Abstract

The present invention relates to a three-way catalyst (TWC) for treatment of exhaust gases of internal combustion engines operated with a predominantly stoichiometric air/fuel ratio, so called spark ignited engines.

Claims

1. Catalyst comprising a carrier substrate of the length L extending between substrate ends a and b and at least three washcoat zones A, B, and C, wherein washcoat zone A comprises Rh and a supporting oxide and extends starting from substrate end (a) over a part of the length L, and washcoat zone C comprises one or more platinum group metals, and a supporting oxide, and extends starting from substrate end (b) over a part of the length L, and washcoat zone B comprises Pd and a supporting oxide, and extends between washcoat zones A and C, wherein L=L.sub.A+L.sub.B+L.sub.C, wherein L.sub.A is the length of washcoat zone A, L.sub.B is the length of washcoat zone B and L.sub.C is the length of washcoat zone C.

2. Catalyst according to claim 1, wherein zone A comprises Rh in an amount of from 0.2 g/L to 4.0 g/L.

3. Catalyst according to claim 1, wherein zone B comprises Pd in an amount of from 0.4 g/L to 20 g/L.

4. Catalyst according to claim 1, wherein zone A comprises only Rh as the PGM.

5. Catalyst according to claim 1, wherein zone B comprises only Pd as the PGM.

6. Catalyst according to claim 1, wherein the supporting oxide is selected from the group consisting of alumina, silica, magnesia, titania, zirconia, ceria, rare earths such as lanthanum neodymium, praseodymium, yttrium, mixtures comprising at least one of these materials and mixed oxides comprising at least one of these materials.

7. Catalyst according to claim 1, wherein washcoat zone A extends over 15 to 50% of the length L of the carrier substrate, washcoat zone B extends over 7 to 30% of the length L of the carrier substrate and washcoat zone C extends over 20 to 78% of the length L of the carrier substrate.

8. Catalyst according to claim 1, wherein the carrier substrate of the length L is a flow-through or filter substrate.

9. Method for the manufacturing of a catalyst according to claim 1 comprising the steps in this order: a. applying a hydrophobic masking zone extending from substrate end (a) over the length L.sub.A, b. coating the carrier substrate from substrate end (a) with a coating to establish a PGM containing washcoat zone B over the length L.sub.B, c. removing the masking zone, d. coating the remainder of the carrier substrate to establish a PGM containing washcoat zones A over length L.sub.A and optionally a PGM containing washcoat zone C of length L.sub.C, e. drying and/or heating the coated carrier.

10. Catalyst system comprising a first catalyst according to claim 1 and another three-way catalyst, a gasoline particulate filter, a HC trap and/or a NOx trap.

11. Catalyst system according to claim 10, wherein substrate end (b) of said first catalyst is followed by the another three-way catalyst.

12. Catalyst system according to claim 10, wherein substrate end (a) of said first catalyst is followed by the another three-way catalyst.

13. Method for treating exhaust gases of a combustion engine, wherein the exhaust gas is passed over the catalyst of claim 1, wherein it enters the catalyst at substrate end (a) and exits it at substrate end (b).

14. Method according to claim 13, wherein said catalyst is arranged in close coupled position.

15. Method for treating the exhaust gas of a spark ignition engine, characterized in that the exhaust gas is passed over the catalyst of claim 1, wherein it enters the catalyst at substrate end (a) and exits it at substrate end (b).

Description

[0043] FIG. 1 illustrates catalysts according to the present invention. The upper part of the figure shows a detail of an inventive catalyst (1) which comprises a carrier substrate (3) which extends between substrate ends (a) and (b) and which carries washcoat zone A (4), washcoat zone B (5) and washcoat zone C (6).

[0044] The lower part of the figure shows a detail of an inventive catalyst (2) which comprises a carrier substrate (3) which extends between substrate ends (a) and (b) and which carries washcoat zone A (4), washcoat zone B (5). Washcoat zone C (6) is coated as a layer of the whole length of the catalyst.

[0045] FIG. 2 illustrates catalyst systems according to the present invention.

[0046] The upper part shows an inventive catalyst system (13) which comprises an inventive catalyst (1) and a conventional three-way catalyst (15). Both catalysts are arranged so that washcoat zone C (6) is followed by the conventional three-way catalyst (15).

[0047] The lower part shows an inventive catalyst system (14) which comprises an inventive catalyst (1) and a conventional three-way catalyst (15). Both catalysts are arranged so that washcoat zone A (4) follows the conventional three-way catalyst (15).

[0048] FIGS. 3a and b shows the results of a poisoned fresh (a) and thermally aged (b) catalyst of the invention in comparison with reversed design and prior art designs.

[0049] FIG. 4 shows the production process in a graphical fashion,

[0050] FIG. 5 depicts the better NH.sub.3 performance of poisoned catalyst according to the invention in comparison with reversed design and prior art designs under rich conditions.

EXPERIMENTAL PART

[0051] The four experimental parts, TWC_1, TWC_2, TWC_3 and TWC_4 shown e.g. in FIG. 3a, FIG. 3b have the same total PGM and washcoat loading. The total Pd loading was 4.51 g/L, the total Rh loading was 0.12 g/L, and the total washcoat loading was 159 g/L. The substrates utilized were of identical dimensions and cell density and consisted of ceramic substrates that were 118.4 mmT91 mm, 900 cell/2.5 mill cell structure. TWC_1 is the reference experimental part with the homogeneous washcoat layer.

[0052] TWC_2, TWC_3 and TWC_4 were built as follows. The monoliths were coated with a homogeneous washcoat load of 159 g/L containing 1.75 g/IL of Pd and 0.12 g/L of Rh (zone C), a high Pd loading of 10 g/L (zone B) and a high Rh loading of 1 g/L (zone A).

[0053] The masking band was applied by vacuuming from one end of the monolith using 55 g of 59 wt.-% solid content of polyurethane emulsion to a length of 25.4 mm from the one end of the monolith. Different emulsions can also be used to mask a zone. The masked part was dried in an oven at 110 C. for 12 hours so the masking agent formed a solid uniform water-impervious layer over the zone in the inlet of the part.

[0054] The application of the Pd-band or zone B was carried out as follows. An aqueous solution consisting of a thickening agent in water was prepared. This was added to control and limit wicking of the aqueous Pd-solution when applied to give the banded zone. The thickening agent was added at 0.5 wt % based on the total weight of solution. Different surfactants can also be used to lower the surface tension of the Pd-solution and minimize wicking thus improving control of the Pd-band length. To this solution was added Pd tetra-amine acetate at a concentration that was determined based on the Pd-loading target in the banded zone, the band/zone length and the amount of solution need to reach the end of the banded zone when injected over the masked zone assuming no solution or Pd uptake on the masked zone. To determine the Pd-solution concentration an initial wet weight uptake for the monolith was measured using a solution of the thickening agent in water without the Pd-salt present. In the current example the masked zone length was 25.4 mm and the target Pd-zone/band length was 25.4 mm. After application of the Pd-band, the excess solution was removed by vacuuming from the injection end of the monolith. The banded/zoned part was then calcined in an up-flow forced air oven with the masking band located at the top of the monolith. The calcination temperature was 550 C. for 30 minutes.

[0055] The application of the Rh-band or zone was carried out as follows. An aqueous solution consisting of a thickening agent in water was prepared. This was added to control and limit wicking of the aqueous Rh-solution when applied to give the banded zone. The thickening agent was added at 0.5 wt % based on the total weight of solution. Different surfactants can also be used to lower the surface tension of the Rh-solution and minimize wicking thus improving control of the Rh-band length. To this solution was added Rh tetra-amine acetate at a concentration that was determined based on the Rh-loading target in the banded zone, the band/zone length and the amount of solution need to reach the end of the banded zone. To determine the Rh-solution concentration an initial wet weight uptake for the monolith was measured using a solution of the thickening agent in water without the Rh-salt present. After application of the Rh-band, the excess solution was removed by vacuuming from the injection end of the monolith. The banded/zoned part was then calcined in an up-flow forced air oven. The calcination temperature was 550 C. for 30 minutes.

[0056] TWC_2 was built in the process order shown in FIG. 4. The masking process was carried out first, then the Pd-band was established to 50.8 mm from the inlet which was followed by the Rh-band process to 25.4 mm from the inlet, after removal of the masking zone. The PGM-zone C was afterwards applied from outlet end (b).

[0057] Comparison testing was carried out using TWC_3 and TWC_4.

[0058] TWC_3 was built by switching the process order of Pd-band process and Rh-band process on TWC_2. TWC_4 was built as the reference experiment part. Pd-band process was carried out and then Rh-band was applied in the same zone as the Pd-band.

Evaluation on Engine Dyno Bench

[0059] Four parts of TWC_1, TWC_2, TWC_3 and TWC_4 were engine aged to a full useful life 100,000 miles condition using a specific accelerated aging cycle. The cycle consisted of repetitive two seconds rich/rich followed by 5 seconds of air-injection for 50 hours. The peak temperature during air injection measured one inch from the catalyst inlet face was 1050 C.

[0060] After the above aging, poison aging was carried out on the same engine using a fuel that was doped with 0.1 wt % of a phosphorous compound. The doping level was such that after 50 hours of stoichiometric aging at 700 C. the catalysts was loaded with 6.6 g of P.sub.2O.sub.5 assuming all the phosphorous was adsorbed by the catalyst.

[0061] The aged catalysts were evaluated on a stand dyno using a 6.0 L GM engine before/after poisoning aging. The catalysts were connected to the exhaust manifold using a stainless-steel pipe. The test results are shown in FIGS. 3-a, 3-b and 5.

[0062] The FLO (Fast Light-Off) testing was carried out using a 21.4 g/sec exhaust gas flow. The mean lambda of the exhaust gas was 1.000 with a lambda modulation of 0.045 at 1 Hz. Data was collected at 1 Hz. Initially the catalyst was heated by the exhaust gas to 500 C. or close to 500 C. after which it was cooled down. During cool-down the exhaust was switched to a bypass line so that it did not pass through the catalyst. When the bed temperature of the catalyst was cooled to 50 C. the exhaust was switched from the by-pass line to the on-line position, so exhaust now passed through the catalyst resulting in the catalyst temperature increasing rapidly. The time needed to reach 50% HC-conversion (T.sub.50) was measured and compared for the four catalysts. The results are shown in FIGS. 3-a, 3-b. The catalyst having the lowest T.sub.50 number is the preferred one. FIG. 3-a shows the comparisons for the P.sub.2O.sub.5 poisoned parts. FIG. 3-b shows the comparisons after thermal aging and before poisoning. It is observed that TWC_2 of the current invention showed the best performance as it had the lowest T.sub.5 time. This was especially true for CO performance before and after poisoning evaluated using this FLO test.

[0063] In order to investigate NH.sub.3 production from NOx, a lambda sweep test on the same stand dyno engine was carried out. A lambda sweep at 600 C. from 1.044/Lean.fwdarw.0.948/Rich with a modulation of 0.055 at 1 Hz was carried out at an exhaust flow of 54.5 g/sec. The sweeping time was 680 seconds. As shown in FIG. 5, it was found that TWC_1 and TWC_2 had lower NH.sub.3 conversion from NOx on the rich side at lambda values below 0.98 lambda. Based on these results it is evident that the TWC_2 design gave the lowest NH.sub.3 formation from NOx as well as having the lowest T.sub.50 for the fast-light-off (FLO) test.