Transition metal/zeolite SCR catalysts

Abstract

A method of converting nitrogen oxides in a gas to nitrogen by contacting the nitrogen oxides with a nitrogenous reducing agent in the presence of a zeolite catalyst containing at least one transition metal, wherein the zeolite is a small pore zeolite containing a maximum ring size of eight tetrahedral atoms, wherein the at least one transition metal is selected from the group consisting of Cr, Mn, Fe, Co, Ce, Ni, Cu, Zn, Ga, Mo, Ru, Rh, Pd, Ag, In, Sn, Re, Ir and Pt.

Claims

1. A catalyst composition for treating exhaust gas comprising an aluminosilicate molecular sieve having a silica-to-alumina ratio of about 8 to about 150, having a framework selected from AEI and AFX, and containing from 0.1 to 10 wt % of a mixture of Cu and Fe, based on the total weight of the molecular sieve, wherein the catalyst is effective to promote the reaction of NH.sub.3 with NO.sub.x to form nitrogen and water, selectively.

2. The catalyst composition of claim 1, wherein said framework is AEI.

3. The catalyst composition of claim 1, wherein said framework is AFX.

4. The catalyst composition of claim 1, further comprising at least one binder selected from alumina, silica, non-zeolite silica-alumina, natural clay, TiO.sub.2, ZrO.sub.2, and SnO.sub.2.

5. The catalyst composition of daim 1, wherein said catalyst composition contains from 0.5 to 5 wt % of said mixture of Cu and Fe.

6. The catalyst composition of dam 6, wherein said catalyst k a washcoat coated on a substrate selected from a metal flow-through substrate, a ceramic flow-through substrate, a wall-flow filter, a sintered metal filter, and a partial filter.

Description

(1) In order that the invention may be more fully understood, reference is made to the following Examples by way of illustration only and with reference to the accompanying drawings, in which:

(2) FIG. 1 is a graph showing NO.sub.x conversion (at a gas hourly space velocity of 30,000 hr.sup.−1) comparing transition metal-containing aluminosilicate catalysts with a transition metal-containing aluminophosphate/small pore zeolite catalyst after relatively moderate lean hydrothermal ageing performed on a laboratory reactor;

(3) FIG. 2 is a graph showing N.sub.2O formation in the test shown in FIG. 1;

(4) FIG. 3 is a graph showing NO.sub.x conversion (at a gas hourly space velocity of 100,000 hr.sup.−1) comparing Cu/Beta zeolite and Cu/SAPO-34 catalysts with a transition metal-containing aluminophosphate/small pore zeolite catalyst after relatively moderate lean hydrothermal ageing performed on a laboratory reactor;

(5) FIG. 4 is a graph showing NO.sub.x conversion (at a gas hourly space velocity of 30,000 hr.sup.−1) comparing transition metal-containing aluminosilicate catalysts with a transition metal-containing aluminophosphate/small pore zeolite catalyst after relatively severe lean hydrothermal ageing performed on a laboratory reactor;

(6) FIG. 5 is a graph showing NO.sub.x conversion for fresh Cu/Zeolite catalysts;

(7) FIG. 6 is a graph showing NO.sub.x conversion for aged Cu/Zeolite catalysts;

(8) FIG. 7 is a graph showing N.sub.2O formation for fresh Cu/Zeolite catalysts of FIG. 5;

(9) FIG. 8 is a graph showing N.sub.2O formation for aged Cu/Zeolite catalysts of FIG. 6;

(10) FIG. 9 is a graph showing the effect of adding HC species to Cu/zeolite catalysts during NH.sub.3 SCR at 300° C.;

(11) FIG. 10 is a graph showing hydrocarbon breakthrough following addition of hydrocarbon species to Cu/zeolite catalysts during NH.sub.3 SCR at 300° C.;

(12) FIG. 11 is a graph showing the adsorption profiles of n-octane at 150° C. flowing through the Cu zeolite catalysts;

(13) FIG. 12 is a graph of the temperature programmed desorption (TPD) of HC species to Cu/zeolite catalysts after HC adsorption at 150° C.;

(14) FIG. 13 is a graph similar to FIG. 6 comparing NO.sub.x conversion activity for aged Cu/Sigma-1, Cu-SAPO-34, Cu/SSZ-13 and Cu/Beta;

(15) FIG. 14 is a graph similar to FIG. 8 comparing N.sub.2O formation for the aged Cu/zeolite catalysts of FIG. 13;

(16) FIG. 15 is a graph similar to FIG. 13 comparing NO.sub.x conversion activity for aged Cu/ZSM-34, Cu/SAPO-34, Cu/SSZ-13 and Cu/Beta catalysts;

(17) FIG. 16 is a graph comparing the NO.sub.x conversion activity of fresh and aged Cu-SAPO-34 and Cu/SSZ-13 catalysts;

(18) FIG. 17 is a graph comparing the NO.sub.x conversion activity of fresh samples of Cu/SAPO-34 with a Cu/naturally occurring chabazite type material;

(19) FIG. 18 is a bar chart comparing the NO.sub.x conversion activity of fresh Cu/SAPO-34 with that of two fresh Cu/naturally occurring chabazite type materials at two temperature data points;

(20) FIG. 19 is a bar chart comparing the NO.sub.x conversion activity of aged Cu/Beta, Cu/SAPO-34, Fe/SAPO-34 and Fe/SSZ-13 catalysts at two temperature data points;

(21) FIG. 20 is a bar chart comparing the hydrocarbon inhibition effect of introducing n-octane into a feed gas for fresh Fe/Beta and Fe/SSZ-13 catalysts;

(22) FIG. 21 is a graph showing hydrocarbon breakthrough following the introduction of n-octane in the experiment of FIG. 20;

(23) FIG. 22 is a bar chart comparing the effect on NO.sub.x conversion activity for a fresh Fe/SSZ-13 catalyst of using 100% NO as a component of the feed gas with using 1:1 NO:NO.sub.2;

(24) FIG. 23 is a schematic diagram of an embodiment of an exhaust system according to the present invention.

(25) FIG. 23 is a schematic diagram of an embodiment of an exhaust system according to the present invention, wherein diesel engine 12 comprises an exhaust system 10 according to the present invention comprising an exhaust line 14 for conveying an exhaust gas from the engine to atmosphere via tailpipe 15. In the flow path of the exhaust gas is disposed a platinum or platinum/palladium NO oxidation catalyst 16 coated on a ceramic flow-through substrate monolith. Located downstream of oxidation catalyst 16 in the exhaust system is a ceramic wall-flow filter 18.

(26) An iron/small pore zeolite SCR catalyst 20 also coated on a ceramic flow-through substrate monolith is disposed downstream of the wall-flow filter 18. An NH.sub.3 oxidation clean-up or slip catalyst 21 is coated on a downstream end of the SCR catalyst monolith substrate. Alternatively, the NH.sub.3 slip catalyst can be coated on a separate substrate located downstream of the SCR catalyst. Means (injector 22) is provided for introducing nitrogenous reductant fluid (urea 26) from reservoir 24 into exhaust gas carried in the exhaust line 14. Injector 22 is controlled using valve 28, which valve is in turn controlled by electronic control unit 30 (valve control represented by dotted line). Electronic control unit 30 receives closed loop feedback control input from a NO.sub.x sensor 32 located downstream of the SCR catalyst.

(27) In use, the oxidation catalyst 16 passively oxidises NO to NO.sub.2, particulate matter is trapped on filter 18 and is combusted in NO.sub.2. NO.sub.x emitted from the filter is reduced on the SCR catalyst 20 in the presence of ammonia derived from urea injected via injector 22. It is also understood that mixtures of NO and NO.sub.2 in the total NO.sub.x content of the exhaust gas entering the SCR catalyst (about 1:1) are desirable for NO.sub.x reduction on a SCR catalyst as they are more readily reduced to N.sub.2. The NH.sub.3 slip catalyst 21 oxidises NH.sub.3 that would otherwise be exhausted to atmosphere. A similar arrangement is described in WO 99/39809.

EXAMPLES

Example 1—Method of Making Fresh 5 wt % Fe/BetaBeta or SAPO-34 or 3 wt % SSZ-13 Zeolite Catalyst

(28) Commercially available Beta zeolite, SAPO-34 or SSZ-13 was NH.sub.4.sup.+ ion exchanged in a solution of NH.sub.4NO.sub.3, then filtered. The resulting material was added to an aqueous solution of Fe(NO.sub.3).sub.3 with stirring. The slurry was filtered, then washed and dried. The procedure can be repeated to achieve a desired metal loading. The final product was calcined.

Example 2—Method of Making Fresh 3 wt % Cu/Zeolites

(29) Commercially available SAPO-34, SSZ-13, Sigma-1, ZSM-34, Nu-3, ZSM-5 and Beta zeolites were NH.sub.4.sup.+ ion exchanged in a solution of NH.sub.4NO.sub.3, then filtered. The resulting materials were added to an aqueous solution of Cu(NO.sub.3).sub.2 with stirring. The slurry was filtered, then washed and dried. The procedure can be repeated to achieve a desired metal loading. The final product was calcined.

Example 3—Lean Hydrothermal Ageing

(30) The catalysts obtained by means of Examples 1 and 2 were lean hydrothermally aged at 750° C. for 24 hours in 4.5% H.sub.2O/air mixture.

Example 4—Severe Lean Hydrothermal Ageing

(31) The catalysts obtained by means of Examples 1 and 2 were severely lean hydrothermally aged at 900° C. for 1 hour in 4.5% H.sub.2O/air mixture.

Example 5—Extended Severe Lean Hydrothermal Ageing

(32) The catalysts obtained by means of Examples 1 and 2 were severely lean hydrothermally aged at 900° C. for a period of 3 hours in 4.5% H.sub.2O/air mixture.

Example 6—Test Conditions

(33) Separate samples of Fe/BetaBeta prepared according to Example 1 and Cu/BetaBeta, Cu/ZSM-5 and Cu/SAPO-34 prepared according to Example 2 were aged according to Examples 3 and 4 and tested in a laboratory apparatus using the following gas mixture: 350 ppm NO, 350 ppm NH.sub.3, 14% O.sub.2, 4.5% H.sub.2O, 4.5% CO.sub.2, N.sub.2 balance. The results are shown in FIGS. 1 to 4 inclusive.

(34) Tests were also conducted on Cu/BetaBeta, Cu/ZSM-5, Cu/SAPO-34 and Cu/Nu-3 prepared according to Example 2 and aged according to Example 3 and tested in a laboratory apparatus using the same gas mixture as described above, except in that 12% O.sub.2 was used. The results are shown in FIGS. 5 to 8 inclusive.

Example 7—n-Octane Adsorption Test Conditions

(35) With the catalyst loaded in a laboratory apparatus, 1000 ppm (as C1 equivalents) propene, n-octane or toluene was injected during NH.sub.3 SCR at 300° C. (350 ppm NO, 350 ppm NH.sub.3, 12% O.sub.2, 4.5% H.sub.2O, 4.5% CO.sub.2, balance N.sub.2). Hydrocarbon desorption was measured by ramping the temperature at 10° C./minute in 12% O.sub.2, 4.5% H.sub.2O, 4.5% CO.sub.2, balance N.sub.2.

Example 8—Results for Experiments Shown in FIGS. 1 to 4 Inclusive

(36) FIG. 1 compares the NO.sub.x reduction efficiencies of a Cu/SAPO-34 catalyst against a series of aluminosilicate zeolite supported transition metal catalysts (Cu/ZSM-5, Cu/Beta and Fe/Beta) after a mild aging. The result clearly demonstrates that Cu/SAPO-34 has improved low temperature activity for SCR of NO.sub.x with NH.sub.3.

(37) FIG. 2 compares the N.sub.2O formation over the catalysts. It is clear that the Cu/SAPO-34 catalyst produced lower levels of N.sub.2O compared to the other two Cu-containing catalysts. The Fe-containing catalyst also exhibits low N.sub.2O formation, but as shown in FIG. 1, the Fe catalyst is less active at lower temperatures.

(38) FIG. 3 compares the NO.sub.x reduction efficiencies of a Cu/SAPO-34 catalyst against a Cu/Beta catalyst tested at a higher gas hourly space velocity. The Cu/SAPO-34 catalyst is significantly more active than the Cu-Beta catalyst at low reaction temperatures.

(39) FIG. 4 shows the NO.sub.x reduction efficiencies of a Cu/SAPO-34 catalyst and a series of aluminosilicate zeolite supported transition metal catalysts (Cu/ZSM-5, Cu/Beta, and Fe/Beta) after severe lean hydrothermal aging. The result clearly demonstrates that the Cu/SAPO-34 catalyst has superior hydrothermal stability.

Example 9—Results for Experiments Shown in FIGS. 5 to 12 Inclusive

(40) NH.sub.3 SCR activity of fresh (i.e. un-aged) Cu supported on the small pore zeolites SAPO-34 and Nu-3 was compared to that of Cu supported on larger pore zeolites in FIG. 5. The corresponding activity for the same catalysts aged under severe lean hydrothermal conditions is shown in FIG. 6. Comparison of the fresh and aged activity profiles demonstrates that hydrothermal stability is only achieved for aluminosilicate zeolites when the Cu is supported on a small pore zeolite.

(41) The N.sub.2O formation measured for the fresh and aged catalysts is shown in FIGS. 7 and 8, respectively. The results clearly show that N.sub.2O formation is significantly reduced by means of supporting Cu on zeolites that do not have large pores.

(42) FIG. 9 compares the effect of HC on Cu/zeolite catalysts where SAPO-34 and Nu-3 are used as examples of small pore zeolite materials. For comparison, ZSM-5 and Beta zeolite are used as examples of a medium and large pore zeolite, respectively. Samples were exposed to different HC species (propene, n-octane and toluene) during NH.sub.3 SCR reaction at 300° C. FIG. 10 shows the corresponding HC breakthrough following HC addition.

(43) FIG. 11 shows the adsorption profiles of n-octane at 150° C. flowing through different Cu/zeolite catalysts. HC breakthrough is observed almost immediately with Cu supported on the small pore zeolites SAPO-34 and Nu-3, whereas significant HC uptake is observed with Cu on Beta zeolite and ZSM-5. FIG. 12 shows the subsequent HC desorption profile as a function of increasing temperature and confirms that large amounts of HC are stored when Cu is supported on the larger pore zeolites, whereas very little HC is stored when small pore zeolites are employed.

Example 10—Results for Experiments Shown in FIGS. 13 and 14

(44) Cu/SSZ-13, Cu/SAPO-34, Cu/Sigma-1 and Cu/Beta prepared according to Example 2 were aged in the manner described in Example 4 and tested according to Example 6. The results are shown in FIG. 13, from which it can be seen that the NO.sub.x conversion activity of each of the severely lean hydrothermally aged Cu/SSZ-13, Cu/SAPO-34 and Cu/Sigma-1 samples is significantly better than that of the corresponding large-pore zeolite, Cu/Beta. Moreover, from FIG. 14 it can be seen that Cu/Beta generates significantly more N.sub.2O than the Cu/small-pore zeolite catalysts.

Example 11—Results for Experiments Shown in FIG. 15

(45) Cu/ZSM-34, Cu/SAPO-34, Cu/SSZ-13 and Cu/Beta prepared according to Example 2 were aged in the manner described in Example 3 and tested according to Example 6. The results are shown in FIG. 15, from which it can be seen that the NO.sub.x conversion activity of each of the lean hydrothermally aged Cu/SSZ-13, Cu/SAPO-34 and Cu/ZSM-34 samples is significantly better than that of the corresponding large-pore zeolite, Cu/Beta.

Example 12—Results for Experiments Shown in FIG. 16

(46) Fresh samples of Cu/SSZ-13 and Cu/SAPO-34 were prepared according to Example 2, samples of which were aged in the manner described in Example 5. Fresh (i.e. un-aged) and aged samples were tested according to Example 6 and the results are shown in FIG. 16, from which it can be seen that the NO.sub.x conversion activity of Cu/SSZ-13 is maintained even after extended severe lean hydrothermal ageing.

Example 13—Results for Experiments Shown in FIGS. 17 and 18

(47) Cu/SAPO-34 and a Cu/naturally occurring chabazite type material having a SAR of about 4 were prepared according to Example 2 and the fresh materials were tested according to Example 6. The results are shown in FIG. 17, from which it can be seen that the NO.sub.x conversion activity of the naturally occurring Cu/chabazite is significantly lower than Cu/SAPO-34. FIG. 18 is a bar chart comparing the NO.sub.x conversion activity of two fresh Cu/naturally occurring chabazite type materials prepared according to Example 2 at two temperature data points (200° C. and 300° C.), a first chabazite material having a SAR of about 4 and a second chabazite material of SAR about 7. It can be seen that whilst the NO.sub.x conversion activity for the SAR 7 chabazite is better than for the SAR 4 chabazite material, the activity of the SAR 7 chabazite material is still significantly lower than the fresh Cu/SAPO-34.

Example 14—Results for Experiments Shown in FIG. 19

(48) Cu/SAPO-34 and Cu/Beta were prepared according to Example 2. Fe/SAPO-34 and Fe/SSZ-13 were prepared according to Example 1. The samples were aged according to Example 4 and the aged samples were tested according to Example 6. The NO.sub.x activity at the 350° C. and 450° C. data points is shown in FIG. 19, from which it can be seen that the Cu/SAPO-34, Fe/SAPO-34 and Fe/SSZ-13 samples exhibit comparable or better performance than the Cu/Beta reference.

Example 15—Results for Experiments Shown in FIGS. 20 and 21

(49) Fe/SSZ-13 and Fe/Beta prepared according to Example 1 were tested fresh as described in Example 7, wherein n-octane (to replicate the effects of unburned diesel fuel in a exhaust gas) was introduced at 8 minutes into the test. The results shown in FIG. 20 compare the NO.sub.x conversion activity at 8 minutes into the test, but before n-octane was introduced into the feed gas (HC−) and 8 minutes after n-octane was introduced into the feed gas (HC+). It can be seen that the Fe/Beta activity dramatically reduces following n-octane introduction compared with Fe/SSZ-13. We believe that this effect results from coking of the catalyst.

(50) The hypothesis that coking of the Fe/Beta catalyst is responsible for the dramatic reduction of NOR conversion activity is reinforced by the results shown in FIG. 21, wherein C1 hydrocarbon is detected downstream of the Fe/SSZ-13 catalyst almost immediately after n-octane is introduced into the feed gas at 8 minutes. By comparison, a significantly lower quantity of C1 hydrocarbon is observed in the effluent for the Fe/Beta sample. Since there is significantly less C1 hydrocarbon present in the effluent for the Fe/Beta sample, and the n-octane must have gone somewhere, the results suggest that it has become coked on the Fe/Beta catalyst, contributing to the loss in NOR conversion activity.

Example 16—Results for Experiments Shown in FIG. 22

(51) Fe/SSZ-13 prepared according to Example 1 was tested fresh, i.e. without ageing, in the manner described in Example 6. The test was then repeated using identical conditions, except in that the 350 ppm NO was replaced with a mixture of 175 ppm NO and 175 ppm NO.sub.2, i.e. 350 ppm total NOR. The results from both tests are shown in FIG. 22, from which the significant improvement obtainable from increasing the NO.sub.2 content of NOR in the feed gas to 1:1 can be seen. In practice, the NO:NO.sub.2 ratio can be adjusted by oxidising NO in an exhaust gas, e.g. of a diesel engine, using a suitable oxidation catalyst located upstream of the NH.sub.3—SCR catalyst.