Method for producing catalytically active wall flow filters

Abstract

The present invention relates to a wall flow filter, to a method for the production and the use of the filter for reducing harmful exhaust gases of an internal combustion engine. Particle filters are commonly used for filtering exhaust gases from a combustion process. Also disclosed are novel filter substrates and their specific use in exhaust gas aftertreatment.

Claims

1. A method for producing coated ceramic wall flow filters having at least two catalytically active zones, the wall flow filter having a first end face, a second end face and a length L and a porosity of at least 50% to at most 80% and a mean pore diameter of 5-50 μm, the method comprising the following steps: i) an excess of a first coating suspension is introduced into the first end face by applying a pressure difference via the wall flow filter; ii) with a pressure difference reversal, an excess of the first coating suspension is removed from the wall flow filter; iii) a second coating suspension without excess is introduced into the wall flow filter via the second end face by applying a pressure difference via the wall flow filter, and wherein step ii) is carried out after step i), and wherein the pressure difference reversal is applied to the wall flow filter both simultaneously and in a common direction to a flow direction of the second coating suspension being introduced into the wall flow filter in step iii).

2. The method according to claim 1, characterized in that in step i), the first coating suspension is introduced into a vertically locked wall flow filter from a lower, first end face into the wall flow filter, and in step iii) the second coating suspension is introduced from an upper, second end face into the still vertically locked wall flow filter.

3. The method according to claim 1, characterized in that a pressure pulse is utilized for the pressure difference reversal and is at least 150 mbar and at most 400 mbar.

4. The method according to claim 1, characterized in that one zone has a positive gradient for an amount of catalytically active material in the coating direction.

5. A catalytically active wall flow filter for the treatment of exhaust gases of a combustion process produced according to claim 1.

6. The catalytically active wall flow filter according to claim 5, characterized in that the catalytically active coatings of the filter are selected from the group consisting of three-way catalyst, SCR catalyst, nitrogen oxide storage catalyst, oxidation catalyst, soot-ignition coating, hydrocarbon storage.

7. The catalytically active wall flow filter according to claim 6, characterized in that the catalytically active coatings are located in the pores and/or on the surfaces of the channel walls of the filter.

8. A method for treating exhaust gas, comprising passing the exhaust gas as to contact the wall flow filter according to claim 5 in a method for oxidizing hydrocarbons and/or carbon monoxide and/or in a method for reducing nitrogen oxide.

9. The method according to claim 1, wherein step ii) is carried out directly after step i).

10. The method according to claim 1, wherein the first and second coating suspensions have a respective particle size differential such that the wall flow filter has both an in-wall coating and an on-wall coating.

11. The method according to claim 8, wherein the first end face is, in use, positioned at a downstream end of the wall flow filter.

12. The method according to claim 1, wherein the first and second coating suspensions are applied as to overlap by 7%-15% of the length L.

13. The method according to claim 1, wherein the first and second coating suspensions have a common composition, and the common composition is a TWC-TWC coating suspension combination.

14. The method according to claim 1, wherein a suction pulse provides the pressure difference reversal in step ii).

15. The method according to claim 1, wherein the wall flow filter is retained in a continuous vertically locked state through each of steps i); ii) and iii).

16. The method according to claim 1, wherein a pressureless holding time before the pressure difference reversal of between 0 and 2 s is maintained.

17. A method for producing coated ceramic wall flow filters having at least two catalytically active zones, the wall flow filter having a first end face, a second end face and a length L and a porosity of at least 50% to at most 80% and a mean pore diameter of 5-50 μm, the method comprising the following steps: i) an excess of a first coating suspension is introduced into the first end face by applying a pressure difference via the wall flow filter; ii) with a pressure difference reversal, an excess of the first coating suspension is removed from the wall flow filter; iii) a second coating suspension without excess is introduced into the wall flow filter via the second end face by applying a pressure difference via the wall flow filter, and wherein step ii) is carried out only after step iii).

18. The method according to claim 17, wherein step ii) is carried out with a pressure pulse that is applied, following introduction of the first and second coating suspensions into contact with the wall flow filter, to both the first and second coating suspensions while still wet.

19. A method for producing coated ceramic wall flow filters having at least two catalytically active zones, the wall flow filter having a first end face, a second end face and a length L and a porosity of at least 50% to at most 80% and a mean pore diameter of 5-50 μm, the method comprising the following steps: i) introducing an excess of a first coating suspension into the first end face by applying a pressure difference via the wall flow filter; ii) with a pressure difference reversal, removing an excess of the first coating suspension from the wall flow filter; iii) introducing a second coating suspension, without excess, into the wall flow filter via the second end face by applying a pressure difference via the wall flow filter, and wherein step ii) is carried out simultaneously with step iii) or only after step iii) is completed.

20. The method according to claim 19, characterized in that a pressureless holding time before the pressure difference reversal of up to 10 seconds is maintained.

21. The method according to claim 20 wherein the pressureless holding time before the pressure difference reversal is between 0 and 2 s.

Description

FIGURES

(1) FIG. 1 shows by way of example the effect of different combinations of the different coating architectures on the basis of Pattern 1 (top)-4 (below). With regard to their effect in the exhaust gas flow. The combination of two coatings with high permeability has the lowest pressure loss, but is otherwise weaker in filtration efficiency, light-off temperature and OSC (oxygen storage) than the other patterns of the present invention. The combination of two coatings with low permeability shows very good values with regard to light-off temperature, OSC and filtration efficiency, but results in an enormous increase in pressure. The best combination of all features (pressure loss, filtration efficiency, light-off temperature and OSC) shows the combination of a low permeability coating in the input cell (inlet side of the filter in the exhaust gas; E) with a porous coating according to the invention with a high permeability in the starting cell (outlet side of the filter in the exhaust gas; A). The optimum setting of the two zone lengths depends on the requirements of the respective motor. The quality criteria of the coated filter can thus be adjusted via the zone length as with a slide controller.

(2) FIG. 2 shows schematically the product with two application zones resulting from the combination of the coating methods. On the porous basic matrix of the filter 100 with the plug 160 is a porous coating 400, also termed a filter cake, with a high permeability, produced by the coating from below with an excess and a pressure difference reversal, and a low-porosity coating 500 with a low permeability, produced by the coating from above without pressure difference reversal. The exhaust gas 600 flows over the coating 500 and flows through the porous matrix of the filter 100 and the open-porous coating 400. After the flow, the layer 400 is also overflowed.

(3) FIG. 3 relates to the combination of an on-wall coating (10a) produced according to the invention and an additional in-wall coating (10b) coming from the other side. The preferred embodiment with an overlap is shown.

(4) FIG. 4 The figures show four exemplary embodiments of Pattern 1 to Pattern 4 from the experimental part

(5) FIG. 5 shows the coating concentration along the longitudinal direction of two inventive (Pattern 2 and 3 respectively) and two non-inventive wall flow filters (Pattern 1 and 4). All three variants have in total the same amount of coating as loading. The five disks per variant were shown (standardized) relative to the loading of the first disks on the left. The distribution of the oxide loading over the length of the filter measured via the determination of the BET surface is shown by way of example for various combinations of coatings for two zones on the wall.

(6) FIG. 6 shows the permeability distribution of two application zones produced according to the invention

EXAMPLES

(7) The exemplary experiments for the preparation of filters with application zones were carried out with the following starting materials.

(8) Filter substrates: Cordierite, 4.66″×4.66″×6.00″, 300/8, average pore diameter d50=17.5 μm

(9) Noble metal loading: 36 g/ft.sup.3 (Pt=0/Pd=30/Rh=6)

(10) Oxide loading: 100 g/L

(11) Specification for the particle-column distribution of the on-wall coatings: d50=4.2-5.0 μm, d90=9.0-18 μm

(12) Distribution of Washcoat Coating in the Wall Flow Filter:

(13) Since the filter used for pattern production has a volume of 1.68 liters, the oxide loading is 168 g. If only a zone of 80% of the length is applied, the zone for these comparisons contains the 168 g. If two zones with 60% of the length L of the filter are applied, each zone contains 50% of the oxides and thus 84 g of oxide. If a zone with 80% of the substrate length were combined with a zone with 40% of the substrate length, the 80% long zone in this comparison would have ⅔ of the amount of oxide (=112 g) and the 40% long zone would have ⅓ of the amount of oxide (=56 g).

(14) Experimental Part for the Production of the Zoned Outlay Pattern 1-4:

(15) General

(16) Aluminum oxide stabilized with lanthanum oxide was suspended in water with a first oxygen storage component, which comprised 40% by weight cerium oxide, zirconium oxide, lanthanum oxide and praseodymium oxide, and a second oxygen storage component, which comprised 24% by weight cerium oxide, zirconium oxide, lanthanum oxide and yttrium oxide. Both oxygen storage components were used in equal parts. The suspension thus obtained was then mixed with a palladium nitrate solution and a rhodium nitrate solution under constant stirring. The resulting coating suspension (washcoat) was used directly to coat a commercially available wall flow filter substrate.

(17) In the following, the methods for producing products are described which each have two application zones which have been coated by different end faces of the filter and each extend over approximately 60% of the length of the filter. The loading of the finished catalysts is composed of 100 g/L of ceramic oxides and 36 g/cft (=1.27 g/L) of noble metal (ratio of palladium to rhodium 5:1), which at the filter volume of 1.6761 L corresponds to a total oxide loading of 167.6 g/filter and a total noble metal amount of 2.13 g/filter. The ratio of ceramic oxides to noble metals is the same in the washcoat of both zones and constant over the entire filter. The total oxide and noble metal amount is equally divided into the two zones during the coating, as a result of which an oxide amount of 83.55 g and a noble metal amount of 1.07 g are ideally applied in each zone coating step.

(18) Pattern 1 (Non-Inventive)

(19) Both washcoat zones a) and b) of Pattern 1 (FIG. 4a) were produced using the same coating method, the ceramic suspension first being introduced into the filter in excess by applying a pressure difference (pressing from below). The excess of oxides is removed by a pressure difference reversal, i.e., the renewed application of a pressure difference (suction from below) opposite the first.

(20) First, zone a) is coated from the bottom from end face A. For this purpose, the suspension has a solids content of around 33% and is pressed into the substrate until 60% of the substrate length is filled with washcoat from bottom to top. The excess washcoat is removed from the filter with a short suction impulse counter to the coating direction (approx. 330 mbar negative pressure, 1.5 sec). After drying and calcining, the filter is coated from the bottom from the front side B in order to produce zone b). The coating is effected analogously to the coating of zone a), only the coating parameters differ slightly (solids concentration of around 35%, suction pulse negative pressure around 210 mbar, suction pulse duration around 0.5 sec., increase in the suction pulse within 0.2 sec to the maximum). The filter is dried and calcined.

(21) Pattern 2 and Pattern 3 (Inventive)

(22) Pattern 2 (FIG. 4b) and Pattern 3 (FIG. 4c) were produced according to one procedure, the coating of zone a) and zone b) each having a different coating process.

(23) First, zone a) is coated from the bottom from end face A. For this purpose, the suspension has a solids content of around 34% and is pressed into the substrate until 60% of the substrate length is filled with washcoat from bottom to top. The excess washcoat is removed from the filter with a short suction pulse (around 330 mbar negative pressure, 1.5 sec suction pulse duration, rise in the suction pulse within 0.2 sec to maximum). After drying and calcining, the filter is coated from the top from the front side B in order to produce zone b). For this purpose, a measured washcoat quantity (solids content about 44%) is added from above to the front side B and a short suction pulse (250 mbar negative pressure, 3 sec) is applied in order to distribute the washcoat in the filter. The filter is dried and calcined.

(24) Sample 4 (Non-Inventive)

(25) Both washcoat zones a) and b) of Pattern 4 (FIG. 4d) were produced using the same coating method, the ceramic suspension being applied from above to the wall flow filter and brought into the filter by applying a pressure difference (suction from below) (non-inventive).

(26) First, zone a) is coated from the front face A from above. For this purpose, the suspension has a solids content of 43-45% and is added in a metered amount from above to the front side A. A pressure difference in the form of a short suction pulse (250 mbar negative pressure, 1 sec) is applied in order to distribute the washcoat in the filter. After drying and calcining, the filter is coated from the top from the front side B in order to produce zone b). The coating parameters for this are analogous to those for coating zone a). The filter is dried and calcined.

(27) Characterization

(28) The effectiveness of a catalytically active filter is determined by the interaction of the functional groups of catalytic performance, filtration efficiency and exhaust back pressure (backpressure) which essentially result from the distribution of the catalytic material and the permeability of the washcoat layers. The distribution and quantity of the catalytically active material in the flow direction of the filter is determined via a measurement of the BET surface (DIN 66132—latest version on filing date), and the permeability is determined by measuring the back pressure on filter samples of Patterns 1 to 4.

(29) Analysis of the gradient of washcoat distribution and permeability:

(30) The samples for the analyses with respect to the determination of the gradients (determination of the distribution of the catalytic material in the axial longitudinal direction) were prepared after coating and calcination as follows: Cutting off the filter plugs on both sides (filter shortened by 2×10 mm) Dividing the residue in the longitudinal direction to 5 equal-length parts (filter disks) To determine the BET gradient, the 5 disks were ground and analyzed.

(31) For the permeability measurement, a block of 10 mm×10 mm×20 mm (width×depth×height) was sawed out of each disk in the center. Each second channel was clogged alternately to produce a small minifilter. For this minifilter, the pressure loss is measured at an air flow of 6 l/min. The pressure loss is initially set to be proportional to the permeability.

(32) FIG. 5 shows by way of example the differences in washcoat loading (gradients in the disks) between the five filter sections with respect to the BET surface, which result from the use of the different methods a) coating with excess washcoat and changing direction of the pressure difference and b) coating without and only with a low level of washcoat excess without an alternating pressure difference for the combination of the on-wall zones. The suspensions used had identical particle size distributions and identical Pd to oxide ratios. Due to the different methods, the suspensions had different viscosities and different solids concentrations. The gradient was always standardized with the value of the left disk.

(33) All 3 variants (Pattern 1, Pattern 2/3, Pattern 4) have in total the same amount of washcoat as loading. The 5 disks per variant was shown (standardized) relative to the loading of the first disks on the left. Thus, the left disk always has 100%.

(34) In the coating according to the method according to the invention (coating with excess washcoat and changing direction of the pressure difference according to method steps i) to iii)), an increasing gradient results in the coating direction, while a uniform distribution of the catalytic material without gradient results in the coating without pressure difference reversal. The effect of the coating methods on the permeability is shown in FIG. 6 by way of example for Pattern 2.

(35) FIG. 6 shows the course of the permeability by way of example for the combination of two on-wall zones, wherein the left zone was produced with a washcoat excess and a reversal of the pressure difference during the coating according to method steps i) to ii) of the first preferred embodiment of the method, while the right zone was produced without excess washcoat and without reversal of the pressure difference. Both zones contain the same amount of oxides and both cover 60% of the length of the filter.

(36) To determine the permeability, the plugs of the filter were first removed. The remainder was divided into 5 equally long disks approximately 26 mm in length. Small cuboids with a base area of 10 mm×10 mm and a height of 26 mm were in turn produced from the disks. The channels were thus provided with plugs, so that 5 small filter bodies were produced. A pressure difference volume flow curve was now determined for the small filters and the permeability was calculated via the Darcy equation. The left zone was used to standardize the permeability of the five small filters.

(37) The first disk of the zone, hereinafter referred to as region A, which was produced with excess washcoat and a reversal of the pressure difference during coating, has a permeability of 4 to 20 times higher in the first 15 to 50 mm than the zone in the following mm. The length L was measured from the end face after removal of the plugs, which had the first contact with the washcoat in the case of coating with excess washcoat and a reversal of the pressure difference. The zone which was produced without excess washcoat and without reversal of the pressure difference during coating has a permeability which corresponds to only 5% to 25% of the permeability of region A with the same particle size distribution of the oxides in the washcoat and the same amount of oxide in the zone. The same applies to the region of the zone which has been produced with a washcoat excess and a reversal of the pressure difference during the coating, which is further away from the end face than the region A.

(38) Table 1 shows the distribution of the oxides and the resulting permeability in a zone produced from below after process steps i) to iii) (the length measurement starts after the plug). The range 0-26 mm was used for standardization to 100%

(39) TABLE-US-00002 TABLE 1 0 mm-26 mm 26 mm-52 mm Oxide loading 100% 132% Permeability 100%  12%

(40) In comparison, table 2 shows the different permeabilities of the coatings from top and bottom (the length measurement starts behind the plug). The range 0-26 mm of the coating from below was set at 100% for standardization.

(41) TABLE-US-00003 TABLE 2 Permeability in the first 26 mm behind the plug Coating from bottom 100% Coating from the top  16%
Catalytic characterization of the products:

(42) In the foregoing, after Patterns 1 to 4 have been characterized with respect to the distribution of the catalytic material and the permeability, the catalytic efficiency, filtration efficiency, and exhaust back pressure of the four different patterns are subsequently determined.

(43) The particle filter Patterns 1 to 4 were co-subjected to engine test bench aging. This aging process consists of an overrun cut-off aging process with an exhaust gas temperature of 950° C. before the catalyst inlet (maximum bed temperature of 1030° C.). The aging time was 19 hours (see Motortechnische Zeitschrift, 1994, 55, 214-218).

(44) The catalytically active particle filters were then tested in the aged state at an engine test bench in the so-called “light-off test”, in the “lambda sweep test” and in the “OSC test”. In the light-off test, the light-off behavior is determined in the case of a stoichiometric exhaust gas composition with a constant average air ratio λ (λ=0.999 with ±3.4% amplitude).

(45) Table 3 below contains the temperatures T70 of Patterns 1 to 4, at which 70% of the considered components are respectively converted.

(46) TABLE-US-00004 TABLE 3 T70 HC T70 CO T70 NOx # stoichiometric stoichiometric stoichiometric 1 403 431 431 2 401 424 429 3 396 413 417 4 392 406 409

(47) The dynamic conversion behavior of the particle filters was determined in a lambda sweep test in a range from λ=0.99-1.01 at a constant temperature of 510° C. The amplitude of λ in this case was ±3.4%. Table 2 shows the conversion of Patterns 1 to 4 at the intersection of the CO and NOx conversion curves, along with the associated HC conversion of the aged particle filters.

(48) TABLE-US-00005 TABLE 4 CO/NOx conversion at the HC conversion at the CO/NOx # intersection intersection 1 95% 96% 2 95% 96% 3 96% 97% 4 96% 96%

(49) The particle filter Pattern 2 shows a slight improvement in light-off behavior compared to Pattern 1 in the aged state. The particle filters of Patterns 3 and 4 show a marked improvement in light-off behavior and dynamic CO/NOx conversion in the aged state compared with Pattern 1.

(50) In order to calculate the oxygen storage capacity of the particle filters in mg/L, the particle filter was placed between two lambda probes and the time offset of the two sensor signals was measured during a jump test (OSC test) with lambda jumps of λ=0.96-1.04 (Autoabgaskatalysatoren, Grundlagen—Herstellung—Entwicklung—Recycling—Okologie, Christian Hagelüken, 2nd edition, 2005, p. 62). Table 5 shows the results of the OSC tests of Patterns 1 to 4.

(51) TABLE-US-00006 TABLE 5 # OSC [mg/L] 1 159 2 166 3 199 4 196

(52) Patterns 3 and 4 show a markedly increased oxygen storage capacity after aging compared to Pattern 1.

(53) The particle filters in Patterns 1 to 4 were compared at a cold blow test bench with respect to the exhaust back pressure.

(54) Table 6 below shows pressure loss data which were determined at an air temperature of 21° C. and a volume flow rate of 300 m.sup.3/h.

(55) TABLE-US-00007 TABLE 6 bp @ 300 m3/h # bp @ 300 m3/h (relative to #1) 1 28.8 mbar — 2 43.2 mbar +50% 3 40.5 mbar +41% 4 64.9 mbar +125% 

(56) The combination of two layers, each produced by a coating method according to steps i)-ii) (Pattern 1), has the lowest pressure loss. The combination of two layers, each produced by a coating process in the absence of steps i)-ii) (Pattern 4), results in an enormous increase in pressure compared to Pattern 1. The two patterns in which the coating processes for Zone a) and Zone b) differ have an acceptable increase in the pressure loss with respect to Pattern 1, but have a significantly lower pressure loss compared to Pattern 4.

(57) The particle filters described were investigated for their fresh filtration efficiency on the engine test bench in the real exhaust gas of an engine operating on average with stoichiometric air/fuel mixture. A globally standardized test procedure for determining exhaust emissions, or WLTP (worldwide harmonized light vehicles test procedure) for short, was used here. The driving cycle used was WLTC Class 3. The particle filters were installed 30 cm downstream of a conventional three-way catalyst which was the same for all particle filters measured. In order to be able to detect particulate emissions during testing, the particle counters were installed upstream of the three-way catalyst and downstream of the particle filter. Table 7 shows the results of the filtration efficiency measurement.

(58) TABLE-US-00008 TABLE 7 # FE [%] 1 77 2 86 3 82 4 91

(59) Pattern 1 has the lowest filtration efficiency, in which both zones were each produced by the same coating method as in steps i)-ii). In contrast, Pattern 4 in which both zones were also produced by the same coating method, but which differed from that for Pattern 1 by excluding steps i) ii), has the highest filtration efficiency. Although the two Patterns 2 and 3, in which the coating processes for zone a) and zone b) differ and were produced in accordance with steps i) to iii), have a lower filtration efficiency than Pattern 4, they result in a significant increase in filtration efficiency compared to Pattern 1.

(60) FIG. 1 shows by way of example a summary of the effect of different combinations of different on-wall washcoat layers with a view to the effect in the exhaust gas flow. The combination of two layers with a high permeability, each produced via a coating method according to the invention according to method steps i) to iii) by applying a pressure difference and a set pressure difference reversal (Pattern 1), has the lowest pressure loss, but is otherwise weaker in filtration efficiency, light-off temperature and OSC than the other Patterns 2 to 4, The combination of two layers with low permeability, which were each produced by applying a pressure difference via a coating method not according to the invention (Pattern 4), shows very good values with regard to light-off temperature, OSC and filtration efficiency, but results in an enormous increase in pressure. The best combination of all features (pressure loss, filtration efficiency, light-off temperature and OSC) is shown surprisingly by the combination of a layer in the input cell (inflow side of the filter in the exhaust gas) with low permeability, which was produced according to claim 6 by applying a pressure difference with a layer in the output cell (outflow side of the filter in the exhaust gas) with a high permeability/which was produced via a coating method according to claim 1 method steps i) to iii). The optimum setting of the two zone lengths depends on the requirements of the respective motor. The quality criteria of the coated filter can thus be adjusted via the zone length as with a shift regulator.