Title:
CONTINUOUS DIESEL SOOT CONTROL WITH MINIMAL BACK PRESSURE PENATLY USING CONVENTIONAL FLOW SUBSTRATES AND ACTIVE DIRECT SOOT OXIDATION CATALYST DISPOSED THEREON
Kind Code:
A1


Abstract:
There is disclosed high cell density or tortuous/turbulent flow through monolithic catalyst devices for the direct catalytic, and (semi) continuous oxidation of diesel particulate matter. The catalysts relate to OIC/OS materials having a stable cubic crystal structure, and most especially to promoted OIC/OS wherein the promotion is achieved by the post-synthetic introduction of non-precious metals via a basic (alkaline) exchange process. The catalyst may additionally be promoted by the introduction of Precious Group Metals.



Inventors:
Southward, Barry W. L. (Frankfurt am main, DE)
Nunan, John G. (Tulsa, OK, US)
Application Number:
12/409212
Publication Date:
04/01/2010
Filing Date:
03/23/2009
Primary Class:
Other Classes:
60/301
International Classes:
F01N3/20; F01N3/10
View Patent Images:
Related US Applications:
20090173060EXHAUST GAS CONTROL APPARATUSJuly, 2009Lortet et al.
20080256924Ultra-compact, high performance aerovortical rocket thrusterOctober, 2008Pederson et al.
20100058754Wave energy machineMarch, 2010Fong et al.
20060124116Clean gas injectorJune, 2006Bui
20090145112Method for Regenerating Nitrogen Oxide Storage CatalystsJune, 2009Goebel et al.
20080141643Systems and processes for reducing NOx emissionsJune, 2008Varatharajan et al.
20080289331MASTER CYLINDER COMPRISING A HELICAL RESUPPLY GROOVENovember, 2008Gaffe et al.
20060260314Method and system integrating combined cycle power plant with a solar rankine power plantNovember, 2006Kincaid et al.
20090132153CONTROLLING EXHAUST GAS RECIRCULATION IN A TURBOCHARGED COMPRESSION-IGNITION ENGINE SYSTEMMay, 2009Shutty et al.
20100050622Hydropneumatic accumulator with flexible porous fillerMarch, 2010Stroganov
20090013945CONTROL OF TURBOCHARGER IMBALANCEJanuary, 2009Buckland et al.



Primary Examiner:
DARJI, PRITESH D
Attorney, Agent or Firm:
SMITH, GAMBRELL & RUSSELL (ATLANTA, GA, US)
Claims:
1. A catalyst system for the direct catalytic oxidation of particulate matter in the off-gas of an internal combustion engine wherein the system comprises a standard flow through monolith device, upon which is coated an active oxidation catalyst formulation for the direct, low temperature oxidation of aforementioned particulate matter, with the active catalyst containing an active redox oxide disposed therein.

2. The catalyst system of claim 1, wherein the monolith is a flow through monolith with >900 cells per square inch.

3. The catalyst system of claim 1, wherein the monolith is a flow through monolith with >600 cells per square inch.

4. The catalyst system of claim 1, wherein the monolith is a flow through monolith with >400 cells per square inch.

5. The catalyst system of claim 1, wherein the monolith is a metal monolith capable of introducing turbulent flow in the exhaust stream.

6. The catalyst system of claim 1, wherein the monolith is a metal or ceramic foam presenting a flow path of highly tortuous nature.

7. The catalyst system of claim 1, wherein the catalyst system is a refractory oxide.

8. The catalyst system of claim 1, wherein the catalyst system contains cerium.

9. The catalyst system of claim 1, wherein the oxide is a cerium oxide in the form of a solid solution of cerium and zirconium oxide (Ce—Zr oxide).

10. The redox active oxide of claim 1, wherein the oxide is a cerium oxide in the from of a Ce—Zr oxide solid solution that is substantially phase pure cubic fluorite solid solution (as determined by conventional XRD method) with oxygen ion conducting properties and comprises a. up to about 95% zirconium b. up to about 95% cerium c. up to about 20% of a stabiliser selected from the group consisting of rare earths, yttrium and mixtures thereof.

11. The catalyst system of claim 1, wherein the catalyst system is a substantially phase pure cubic fluorite solid solution additionally modified by the introduction of one or more base metal dopant species selected from the group consisting of a transition metal, an alkali metal, an alkaline earth metal and a group IIIb metal.

12. The catalyst system of claim 11, wherein the redox oxide is a base metal doped cerium containing cubic fluorite solid solution produced by contacting redox active material with a precursor solution of dissolved cations under conditions of high ph/low hydronium ion (H3O+)/low proton (H+) content.

13. The catalyst system of claim 12, wherein the base metal is introduced into the redox active oxide by means of an ammonium hydroxide/ammoniacal complex of the metal cation.

14. The catalyst system of claim 12, wherein the base metal is introduced into the redox oxide by means of an organic amine complex of the metal cation.

15. The catalyst system of claim 12, wherein the base metal is introduced into the redox oxide by means of a hydroxide compound of the metal cation.

16. The catalyst system of claim 12, wherein the concentration of metal species introduced is about 0.01 weight % to about 10 weight %.

17. The catalyst system of claim 16, wherein the concentration of metal species introduced is most preferably 0.1 wt % to about 2.5 wt %

18. The catalyst system of claim 12, wherein the base metal doped solid solution contains metal at high levels of dispersion such that phase analysis by conventional XRD methods retains a substantially phase pure cubic fluorite phase (>95%), with bulk metal oxide dopant phase being recorded at <5% and dopant metal oxide particle size, as determined by line-broadening/Scherrer equation method, is about 30 A to about 100 A.

19. The catalyst system of claim 12, wherein the base metal doped solid solution contains metal at high levels of dispersion such that phase analysis by XRD reveals the promoted material maintains at least 95% cubic fluorite phase after hydrothermal oxidising aging at 1100° C.

20. The catalyst system of claim 12, wherein the base metal doped solid solution contains metal at high levels of dispersion such that phase analysis by XRD reveals the promoted material maintains at least 99% cubic fluorite phase after hydrothermal oxidizing aging at 1100° C.

21. A device for the direct catalytic oxidation of soot comprising the catalyst system of claim 1, and a housing wherein the temperature of required for soot oxidation is about 100 to about 650° C.

22. A device for the direct catalytic oxidation of soot comprising the catalyst system of claim 1, and a housing wherein the temperature of required for soot oxidation is about 200 to about 400° C.

23. A device for the direct catalytic oxidation of soot comprising the catalyst system of claim 1, and a housing wherein continuous soot oxidation occurs for temperatures of about 100 to about 650° C.

24. A catalytic system for the direct catalytic oxidation of soot according to claim 1, wherein the catalyst system is free of a platinum group metal.

25. The catalyst system for the direct catalytic oxidation of soot according to claim 1, further comprising a platinum group metal.

26. The catalyst system for the direct catalytic oxidation of soot according to claim 25, wherein the platinum group metal is selected from the group consisting of platinum, palladium, rhodium and mixtures thereof.

27. The catalyst system for the direct catalytic oxidation of soot according to claim 25, further comprising a catalytically active washcoat disposed upon the monolith as a single layer washcoat which additionally contains Al2O3, modified Al2O3, SiO2, ZrO2, or combinations thereof or other suitable refractory oxide as an additional support or binding agent.

28. The catalyst system for the direct catalytic oxidation of soot according to claim 25, further comprising a catalytically active washcoat disposed upon the monolith in two or more layers with a first layer containing substantially Al2O3, modified Al2O3, SiO2, ZrO2, combinations thereof or other suitable refractory oxide as a support or binding agent and a second layer comprising the active oxidation catalyst formulation, including a base metal doped mixed oxide.

29. A method of treating exhaust gas comprising passing an exhaust gas over the catalyst system of claim 1.

Description:

REFERENCE TO RELATED APPLICATIONS

This application claims the benefit of provisional application 61/308,879, filed Mar. 27, 2008, and is a continuation-in-part of application Ser. No. 12/240,170 filed Sep. 29, 2008, and application Ser. Nos. 12/363,310 and 12/363,329, both filed Jan. 30, 2009, all of which are relied on and incorporated herein by reference.

INTRODUCTION AND BACKGROUND

Over the last thirty years increasingly stringent legislative limits have been introduced to regulate the emissions from both petrol (gasoline) and diesel internal combustion engines. See Regulation (EC) No 715/2007 of the European Parliament and of the Council, 20 Jun. 2007, Official Journal of the European Union L 171/1, see also Twigg, Applied Catalysis B, vol. 70 p 2-25 and R. M. Heck, R. J. Farrauto Applied Catalysis A vol. 221, (2001), p 443-457 and references therein. In the case of diesel/compression ignition engines this has led to the implementation of the Diesel Oxidation Catalyst (DOC), Diesel NOx Trap/NOx Storage Catalyst (DNT/NSC) and Selective Catalytic Reduction catalyst (SCR) to address gaseous emissions of CO, HC (DOC) and nitrogen oxides (NOx). However, in addition to the gaseous components, the diesel exhaust stream also contains entrained solids, commonly referred to as particulate matter or soot. This carbon-based material is a byproduct of incomplete combustion and arises due to heterogeneity of the air-fuel mixture within the cylinder and presents a unique and specific challenge with regards to its control and conversion into environmentally benign products. Thus, while previously it has been possible to meet all legal requirements for exhaust emissions of particulates via engine-related control measures only (SAE Paper 2007-01-0234, Pfeiffer et al.), the stringent targets embodied in, for example, Euro 5 or Euro 6 (Regulation (EC) No 715/2007 of the European Parliament and of the Council, 20 June 2007, Official Journal of the European Union L 171/1) necessitates the introduction of the Diesel Particulate Filter (DPF), aka the ‘Wall-Flow Filter’ to enable specific remediation of soot.

The DPF typically comprises an inert porous ceramic e.g. silicon carbide, cordierite etc. monolith substrate which may be additionally wash-coated with an active catalytic formulation to facilitate the chemistries required of the device e.g. soot combustion, (secondary) emission control, NOx abatement, etc. The wash-coat formulation itself is typically a heterogeneous-phase catalyst and may contain particles of highly active precious group metal (PGM) dispersed and stabilized on a refractory oxide support or supports; e.g. alumina. The DPF may additionally contain an Oxygen Storage (OS) component to enhance the regeneration function of the filter.

The DPF achieves high filtration efficiency of particulates as a result of the physical filtration achieved by forcing the exhaust flow through the porous wall of filter. However, over time this results in a build up of stored material, commonly referred to as a filter-cake, within the filter which results in an ever increasing back pressure penalty, arising from the work required to force the gas flow through an increasingly dense flow restriction. This flow restriction leads to an unacceptable decrease in engine performance and hence, the filter-cake must be combusted in order to ‘regenerate’ the filter to a near pristine condition such that it is able to again store the carbonaceous particulates with minimal back pressure penalty. However, at this time a fully passive and continuous soot regeneration technology has not been demonstrated on a vehicle and hence the complete regeneration of the filter requires an “active” or forced regeneration strategy; see e.g. U.S. Pat. No. 7,441,403; U.S. Pat. No. 7,313,913. These active strategies are reliant upon the manipulation of the gross reaction conditions of the exhaust in order to achieve filter regeneration. Hence, the regeneration of the particulate filter described above may be achieved by the use of auxiliary devices. For example, an air fuel nozzle and an ignition device can be used and operated, when desired, to heat the exhaust gases and the particulate filter to the temperature required for the homogeneous combustion of the trapped particulate matter. In this manner, the trapped soot may be burned from the filter surfaces to permit a continuous flow of the exhaust gases. Alternatively, an electric heater may be employed to generate the heat to initiate the combustion cycle; e.g. U.S. Pat. No. 7,469,532. More commonly however, the filter is regenerated by a so-called “post-injection” cycle in which secondary fuel is introduced, either by late cylinder injection or via dedicated fuel injection unit in the exhaust train, and the hydrocarbons thus entrained in the exhaust flow are combusted over an oxidation catalyst situated prior to the DPF to generate a transient thermal ‘bloom’ within the filter which initiates the conversion of the soot into environmentally benign products (CO2, H2O); e.g. see SAE paper 2008-01-0481 and references therein.

However, the use of wall-flow filters to achieve efficiency in particulate removal presents some immediate issues. Firstly due to the wall flow mechanism of filtration, the DPF introduces a significant back-pressure on the engine. Moreover, the addition of a washcoat to the bare filter increases this back pressure and the sieving action employed to trap soot results in yet a further and continuous increase in back pressure. As indicated previously any increases in backpressure are at the expense of engine efficiency and lead to an ever increasing fuel economy penalty, due to the wasted work pushing exhaust gas through the soot filter cake, washcoat formulation and filter. Thus, significant efforts have been expended in the development of mechanically and thermally robust DPFs with high filtration efficiency but decreased back-pressure penalty and in the development of active washcoat formulations capable of high conversions at minimal wash-coat loads in attempts to minimize the back-pressure/fuel economy issues otherwise evident.

Additionally, there remain outstanding questions regarding the regeneration cycle employed by the DPF. Such traditional ‘active’ cycles are all energy intensive and result in a substantial and unattractive fuel penalty; i.e. an additional and ongoing operational cost. Thus, the use of sacrificial hydrocarbon species in the active regeneration cycle imposes as high as a 5% decrease in fuel economy. Moreover, the implementation of an active emissions control strategy requires complex and accurate engine management protocols to avoid incomplete regeneration and/or untreated emissions; e.g. U.S. Pat. No. 7,412,822. In addition, soot combustion initiated in this manner results in a phenomenon known as ‘oil dilution’ which can both adversely affect engine operation and result in ash deposition (inorganic salts) within the filter which impact the back pressure, soot capacity and catalytic performance of the filter; e.g. U.S. Pat. No. 7,433,776. Finally, it should be noted that soot combustion initiated in this manner proceeds in a more homogeneous; i.e. non-catalytic manner and can be uncontrolled. This in turn can result in localized exothermic ‘hotspots’ of extreme temperature (T>1000° C.) which can easily damage the catalytic efficiency of a formulation (PGM sintering, PGM de-alloying, surface area and porosity collapse of the support oxide). In the worst case scenario, catastrophic uncontrolled combustion of soot can destroy the DPF monolith through thermal degradation or even melting the monolith.

Hence, many attempts have been made to address or limit the extent of the issues related to the active regeneration strategy. Such efforts are exemplified by attempts to introduce more passive regeneration strategies based upon the use of the redox chemistry of advanced OS materials, e.g. US 2005/0282698 A1. In these studies it was shown that decreases in the temperature required for soot oxidation may be achieved by utilisation of active oxygen species derived from a redox active washcoat OS material. The OS materials used in the DPF are typically based upon CeO2 or other redox oxide and are employed to ‘buffer’ the catalyst from local variations in the air/fuel ratio during regeneration or other transient event. They do this by ‘releasing’ active oxygen from their 3-D structure in a rapid and reproducible manner under oxygen-depleted transients, ‘regenerating’ this lost oxygen by adsorption from the gaseous phase when oxygen-rich conditions arise. This activity is attributed to the redox activity of CeO2 via the 2Ce4+→2Ce3++[O2−] reaction. This high availability of oxygen is critical for the promotion of generic oxidation/reduction chemistries e.g. CO/NO chemistry for the petrol (gasoline) three-way catalyst, or more recently for the direct catalytic oxidation of particulate, matter (soot) in the CDPF e.g. US 2005/0282698 A1, SAE 2008-01-0481 and references therein. This work is one of many studies examining the chemistry, synthesis, modification and optimisation of Ce—Zr based OS materials. For example, the use of Ceria-Zirconia materials doped with lower valent ions for emission control applications have been extensively studied e.g. U.S. Pat. No. 6,468,941, U.S. Pat. No. 6,585,944. These studies demonstrate that lower valent dopant ions such as Rare Earth metals, e.g. Y, La, Nd, Pr, etc., Transition metals e.g. Fe, Co, Cu etc. or Alkaline Earth metals e.g. Sr, Ca and Mg can all have a beneficial impact upon oxygen ion conductivity This is proposed to arise from the formation of oxygen vacancies within the preferred cubic fluorite lattice of the solid solution which lowers the energy barrier to oxygen ion transport from the crystal bulk to the surface thereby enhancing the ability of the solid solution to buffer the air fuel transients occurring in the exhaust stream of a typical petrol (gasoline) three-way catalyst application.

Additionally it has been shown (U.S. Pat. No. 6,468,941 and U.S. Pat. No. 6,585,944) that the use of specific examples of the above dopants can provide full stabilization of the preferred cubic fluorite lattice structure for ceria-zirconia solid solutions, with Y being identified as having particular benefit hereto. The presence of the preferred cubic fluorite structure has been found to correlate with the most facile redox chemistry for Ce4+→Ce3+, from both the surface and bulk of the crystal, thus dramatically increasing the oxygen storage and release capacity, as compared to bulk CeO2. This benefit is especially pronounced as the material undergoes crystal growth/sintering due to the hydrothermal extremes present in typical exhaust environments. The incorporation of especially Y and to a lesser extent La and Pr have also been demonstrated to limit or, in certain cases, circumvent the disproportionation of the single cubic phase Ceria-Zirconia into a composite consisting of more Ce-rich cubic phases and more Zr-rich tetragonal phases, a process which results in marked decrease in redox function, surface area etc. of the solid solution.

Finally, U.S. Pat. No. 6,468,941, U.S. Pat. No. 6,585,944, U.S. patent application Ser. No. 12/363,310 and U.S. patent application Ser. No. 12/363,329 (both applications being incorporated herein by reference) teach the potential for employing base, i.e. non-precious group (Pt, Pd, Rh, Au etc.) dopant metals into or with the cubic fluorite lattice of the solid solution as an alternative means to promote the redox chemistry of cerium, with Fe, Ni, Co, Cu, Ag, Mn, Bi and mixtures of these elements being identified as of particular interest. Hence while typical non-promoted OS materials typically exhibit a redox maximum, as determined by H2 Temperature Programmed Reduction (TPR), at ca. 600° C., the inclusion of base metals within the lattice can decrease this temperature by >200° C. at a fraction of the cost incurred by the use of precious metals.

However, while these base metals can be beneficially incorporated in the CeZrOx lattice and this incorporation can significantly promote low temperature redox function for fresh materials, the addition of these elements can also decrease fresh and aged phase purity and significantly decrease hydrothermal durability (promote crystal sintering and material densification), leading to losses in aged performance cf. base compositions without additional base metal. In addition, during conventional aging cycles reactions may occur between the gas phase and the CeZr material which can result in extraction of these additional base elements from the cubic fluorite lattice. This, in turn, can result in formation of separate bulk phase(s) with low intrinsic catalytic activity or in a worst case scenario, phases which directly interact with the OS or other catalyst component resulting in a direct or indirect poisoning of the catalyst. Hence, until recently, particular synthetic care was required to enable the incorporation of promotant lower valent ions into the cubic fluorite structure while ensuring both the electrical neutrality and phase preservation. Thus, as shown in U.S. application Ser. No. 12/363,310, the synthesis of an OS material containing a specific low valent base metal promoter (Ag) ‘doped’ into a cubic fluorite structure with ca. 40% Ce resulted in phase disproportionation into Ce-rich and Ce-poor domains, with a marked decrease in redox performance. This contrasted with a newly developed basic exchange process which was able to provide an equivalent composition with high activity and hydrothermal durability for use in the DPF.

Unfortunately, despite the large number of attempts to employ advanced OS materials in either passive or active regeneration methodologies in vehicular applications these have previously met with limited success. Extensive studies of the chemistry occurring in these systems have demonstrated that the activity of the OS-based catalyst is dependent upon high ‘Contact Efficiency’ between the OS material and the soot; e.g. see, Applied Catalysis B. Environmental 8, 57, (1996). Subsequent studies, described in SAE paper 2008-01-0481 and U.S. application Ser. Nos. 12/363,310 and 12/363,329 have now identified that the loss of contact efficiency between the OS and soot may arise from specific chemistries involving the significant NO engine emissions typical of pre-EuroV legislation engines. This process has been denoted as ‘de-coupling’ of the OS and soot and is the result of the reaction of engine out NO over oxidized PGM to produce NO2 which combusts the soot in the immediate environment of the catalyst producing CO+NO. The NO byproduct of this process is further ‘recycled’ to NO2 and the soot combustion re-initiated, again removing only that soot which immediately contacts the catalyst. This cycle is the basis of U.S. Pat. No. 4,902,487 and was previously believed to be the major reaction providing low temperature soot combustion/regeneration. However, this mechanism appears only effective at removing low concentrations of soot and indeed only that proportion of soot in direct contact with the catalyst. Thus, this mechanism effectively ‘de-couples’ the catalyst and soot and dramatically decreases the effectiveness of the OS-mediated regeneration method and may in fact be considered to be a reactive poison which effectively ‘deactivates’ the ‘true’ OS mediated low temperature, passive, soot regeneration reaction required for optimum soot emission control. Fortunately, however, the design of the next generation ion exchanged OS materials has been found to be effective at both circumventing this ‘de-coupling’ process and also in promotion of the redox characteristics of the OS and hence demonstrated robust performance benefits with respect to soot regeneration on all both the engine dynamometer and in vehicle trials on wash-coated DPFs (see U.S. application Ser. No. 12/363,329).

Based upon the aforementioned requirements and challenges, it is apparent that the conventional approach of using the wall-flow DPF to ensure highly effective trapping and subsequent combustion of particulates presents many challenging technical obstacles. Accordingly, what is needed in the art are improved materials and/or methods for the control and conversion of particulate matter whilst offering both a reduction in fuel penalty for regeneration but also decreased complexity with regards to initiation and control of any regeneration cycle compared to the traditional DPF and conventional active regeneration strategy. Herein we propose the use of an in-line soot combustion device and catalyst with minimal/significantly decreased back-pressure penalty for the (semi) continuous and (semi) passive combustion of retained particulate matter at significantly lower temperatures than those required for the conventional wall-flow filter.

SUMMARY OF THE INVENTION

A significant advance in the development of a method and apparatus for the (semi) continuous, direct catalytic, oxidation of diesel particulate matter may be realised by the combination of base metal modified Oxygen Storage (OS) materials with a conventional flow substrate. The substrate is selected from a range of ceramic or metallic technologies upon which the active washcoat is disposed. Such substrates can be metallic parts, ceramic or metal foams. The substrate is further characterised by presenting a high number of channels or cells per unit area or by the ability to introduce turbulent flow due to the construction of its internal flow channels. The particular combination of the base metal modified OS direct soot oxidation catalyst with the flow through monolith provides a synergy which enables high conversion of particulate matter without the backpressure penalty introduced by the conventional DPF. Specifically, the synergy is believed to arise from the ability of the active OS to combust soot at lower temperatures which in turn is facilitated by the decreased thermal mass of the conventional substrate, with the latter still providing sufficient geometric surface area for soot deposition and reaction. This provides for the large improvements in lower temperature activity and is in marked contrast to the conventional wall flow DPF wherein large thermal mass of the substrate, particularly for SiC DPF, inhibits initiation and especially propagation of soot combustion. Thus, this combination of technologies provides a means for the effective conversion of particulate matter under conditions more typical of the standard driving cycle i.e. soot combustion without recourse to high temperature active regeneration cycles and the various penalties and other issues associated thereto.

The doped OS materials herein are based upon ZrO2/CeO2 solid solutions containing a substantially phase pure cubic fluorite structure and are produced by the specific ion exchange of base i.e. non-precious group metals. The range of appropriate materials and full details regarding execution of the ion exchange are described in U.S. application Ser. Nos. 12/363,310 and 12/363,329. The mode of ion exchange essentially involves the introduction of active metal/cations into the solid solution under chemically basic, i.e. conditions of high pH, that is say high OH/low hydronium (H3O+) or proton (H+) content. As demonstrated in the aforementioned work, the resultant materials demonstrate high activity and hydrothermal durability in contrast to any promotion realized by conventional impregnation of an acidic metal e.g. metal nitrate, where formation of bulk oxide phases in fresh materials and rapid sintering of such oxide phases, with resultant deactivation, is the norm. The proposed exchange of the H+ species, present at Ce3+ defect sites within the Ce—ZrOx lattice, by metal ions enables the incorporation and stabilization of specific mono-valent e.g. K+, di-valent e.g. Cu2+, tri-valent e.g. Fe3+ and higher valence ions at high dispersion within the oxide matrix. The choice of base metals thus incorporated is based upon oxides known to be active for reactions of especial interest or catalytic importance. Metals of specific catalytic significance include Ag, Cu, Co, Mn, Fe, alkali metals, alkaline earth metals or transitions metals, or other metal or metalloid known to form a stable nitrate which can undergo subsequent decomposition and reduction N2 under conditions within the conventional operational window of the vehicle exhaust. The term “transition metal” refers to the 38 elements in Groups 3 to 12 of the Periodic Table of Elements.

The use of high cell density/turbulent flow through monoliths is also required to provide sufficient interaction and subsequent reaction between the entrained soot particles within the exhaust flow and the active catalytic coating. The term high cell density is consistent with preformed flow through monolith substrates with a large (≧600) number of individual cells of flow channels per square inch. It is proposed that this high cell density firstly introduces turbulence at the inlet to maximise possible soot collisions with the active wash-coated walls of the monolith. Secondly, the high cell density restricts the flow path through the monolith, again increasing the potential for particulate collisions and retention/reaction on the active wash-coat, but without the large backpressure penalties associated with the conventional DPF. Moreover the use of the flow-through substrate removes existing constraints regarding total washcoat loading, or the use of layered technologies with specific functionalities, e.g. soot combustion catalyst in one layer (overcoat) and SCR catalyst in a second layer (undercoat), equally it enables the use of an undercoat rich in Al2O3 to provide high washcoat adhesion, but with low intrinsic catalytic function, onto which a second pass containing all required OS, PGM and NOx trap etc. active components may be dispersed. In this second example, the overcoat would under normal conditions present lower adhesion and would conventionally be diluted with binder, e.g. Al2O3, however, the incorporation of binder results in a decrease in activity due to dilution of the active phase, hence the layered design is preferred. This layering ensures the surface coating that would interact/react with the soot as it passed through the flow-through substrate would exclusively consist of active material and would therefore maximize catalytic action. The enabling of higher washcoat loads when using the flow through monolith also provides the capability of employing higher concentrations of active materials to be coated on the substrate thereby further enhancing the performance and hydrothermal durability of the technology without the catastrophic back pressure penalty such an approach would present using the conventional DPF. Hence by use of the flow through substrate washcoat load could be increased from 10 g/l to 180 g/l or higher concomitantly increasing the effective geometric surface area for catalyst to soot contact to again increase in combustion efficiency.

In addition by use of the flow though monolith the textural characteristics of the washcoat e.g. particle size, roughness etc. may be optimised for activity rather than merely to minimize back pressure penalty. Conventional formulations for DPFs typically target a D50 (diameter of particle at 50%) value of 5 microns or less to enable ‘in-wall’ coating, i.e. coating of the internal porosity of the substrate without formation of a discrete washcoat layer on the surface of monolith, in order to minimize backpressure penalty. Such a particle size distribution is typically achieved by aggressive milling of the raw materials used in the washcoat. However, the use of this ‘hyper-milling’ to obtain the very small particles for in-wall coating has been found to be extremely destructive to the activity, stability and surface areas of the OS and alumina components employed in typical formulations. As a result such a process can adversely affect the rate of release and total Oxygen Storage capacity of the OS. In addition the hyper-milling can result in cation extraction and phase disproportionation for the OS with further poisoning of any PGM function arising from the deposition of extracted cations. In contrast, the use of a washcoat with high textural/roughness characteristics has previously been identified as beneficial in three-way applications (e.g. see SAE 2005-01-1111) and may enhance initial flow turbulence and thus increase the probability of catalyst to soot contact. The retention of texture due to the absence of aggressive milling can also be expected to increase the probability of primary soot moieties effectively contacting the OS material. The extent of intimate contact has been shown to correlate directly with direct soot combustion (see Applied Catalysis B. Environmental 8, 57, 1996, U.S. application Ser. No. 12/363,329, SAE 2008-01-0481). Moreover, the integrity of the active formulation with respect to phase, OS function or PGM functionalities is always of prime importance especially herein since it has been demonstrated that the energy produced by the combustion of HC, CO or the SOF (soluble organic fraction) present in soot matter have been identified as a means of initiation and propagation of the combustion of the remaining soot (akin to striking a match or a primer, see SAE 2008-01-0481).

Benefits and features of the present invention include:

a) Provision of a hydrothermally robust direct soot catalyst system, active at temperatures relevant to diesel vehicle operation for (semi) continuous, direct catalytic, oxidation of soot;

b) Particulate control system without requirement for DPF substrate thereby removing associated substrate cost, back pressure constraints, canning and space requirements and ancillary systems associated with conventional DPF;

c) Provision of an active catalyst providing full oxidation function without recourse to complex conventional active regeneration cycles with associated fuel penalty, filter cake formation, potential for catastrophic uncontrolled regeneration, oil dilution, ash deposition or other issue associated with conventional DPF;

d) Flexibility of coated part design with respect to washcoat load, particle size/texture and hence the ability to optimize washcoat based upon performance and durability requirements and not merely backpressure constraints;

e) Ability to employ multilayer technologies with specific functionalised layers to providing additional catalytic properties and functions from a single monolith and to potentially achieve further chemical synergies and performance advantages previously impossible when employing the conventional DPF.

f) Synergistic operation between the active washcoat and high cell density substrate to facilitate rapid oxidation of soot and soluble organic fraction to thereby circumvent the potential for ‘face plugging’, a phenomenon associated with the use of conventional high CPSI monoliths with conventional catalyst formulations.

This strategy clearly contrasts to those employed in conventional DPF systems. For the conventional design catalytic functionality is typically more limited i.e. control of CO, HC from primary or secondary emissions ([SAE Paper 2007-01-0234, Pfeiffer et al.), NH3—SCR of NOx (US 2008/202107-A), etc. Moreover, the design constraints for conventional formulations are significant and are typically based upon the primary balance required between filtration efficiency and maximum system backpressure.

In order to meet the design targets for this synergistic operation of catalyst and monolith there are several key performance requirements that must be met. Firstly, there is a requirement for increased ceria reducibility at lower temperatures than is conventionally obtained with binary, tertiary or even quaternary Ce—Zr—REOx systems. This reducibility is critical to achieve the low temperature O-ion donation from the catalyst to the soot which has been proposed as being a key reaction step (SAE 2008-01-0481; U.S. application Ser. Nos. 12/363,310 and 12/363,329; Appl. Catal. B vol. 17, 1998, p 205, Appl. Catal. B vol. 75, 2007 p 189, Catal. Today 121, 2007, p 237, Appl. Catal. B vol. 80, 2008, p 248]. Hence examinations of the use of CeOx or CeZrOx containing oxide solid solutions for soot oxidation have been widespread. However conventional CeZrOx solid solutions, as typically employed in three-way catalysts, typically exhibit a redox maximum, as determined by H2 Temperature Programmed Reduction (TPR) at ca. 600° C. This imposes the requirement for high exhaust gas/reaction temperatures in the application in order for the OS material to provide the maximum “buffering” or oxygen donation benefit. Moreover, this requirement for high temperature to access the active lattice oxygen is a barrier to the implementation of CeZrOx for lower temperature direct soot oxidation. In order to address this temperature issue OS materials are typically “promoted” by the addition of a Precious Group Metal (PGM) component, e.g. Pt, Pd or Rh. However, promotion by these metals contributes a large additional cost to the price of the emission control system. Moreover the addition of PGM, especially Pt, promotes the ‘classical’ chemistry of NOx-mediated soot oxidation as described in U.S. Pat. No. 4,902,487. However, it has now been found and clearly demonstrated that NOx mediated soot oxidation is only effective at removing low concentrations of soot and indeed only that proportion of soot in direct contact with the catalyst and may be considered to be an effective catalyst poison for direct OS mediated soot oxidation (SAE 2008-01-0481, U.S. application Ser. Nos. 12/363,310 and 12/363,329). Thus, what is required in the art is a method to promote the oxygen ion conductivity of the CeOx/CeZrOx-based oxide material, but without use of expensive PGM and without the undesirable consequence of increasing the NOx oxidation chemistry of the catalyst.

A second limitation, again typical OS materials used to date, is a limitation with regard to their total Oxygen Storage Capacity, that is to say the amount of available oxygen as measured by TPR is typically lower than that expected from consideration of the total Ce IV content of the OS material. Many data available to date are consistent with as little as only ca. 50% of the total Ce IV available undergoing reduction. At this time it is uncertain whether this is due to a fundamental issue, or due to limitations with the current synthetic method(s) employed in the manufacture of the OS material leading to a mixed Ce IV/Ce III valency or whether a combination of additional chemical, structural or textural limitations are responsible.

Finally, typical OS materials provide only limited, if any, additional synergies to the emission control system. As described elsewhere, ideal material components provide additional integrated chemical mechanisms to further enhance emissions control, e.g. NOx scavenging and reduction to N2.

Hence, while OS materials are key components in realising highly active and materials present significant limitations to development of the next generation of exhaust catalyst that will be required to comply with newer and ever more stringent emission targets. What is required is a new class of OS materials that are active at lower temperatures, especially the Cold Start portion of vehicular applications to promote catalytic function. These OS materials should also display high hydrothermal durability and be tolerant to potential exhaust poisons in order to enable their use in the wide range of demanding exhaust environments.

The above-described and other features will be appreciated and understood by those skilled in the art from the following detailed description, drawings, and appended claims.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention will be further understood with reference to the accompanying drawings, wherein:

FIG. 1—shows a schematic of the synthetic gas bench (SGB) reactor in which the concept trials were executed;

FIG. 2—shows the impact of reactive gas mixture during loading on subsequent soot combustion;

FIG. 3—outlines the various gas compositions employed during the loading and regeneration trials;

FIG. 4—compares the back pressure (hereafter B.P.) response for a 400 CPSI (cells per square inch) flow through monolith during a three hour (10800 s) soot loading cycle as a function of either gas environment during loading or temperature and gas environment;

FIG. 5—impact of reaction conditions during loading (ex FIG. 4) on subsequent soot combustion cycle via TPO;

FIG. 6—compares the O2 concentrations at the reactor outlet during the TPO cycles described in FIG. 5;

FIG. 7—impact of monolith cell density on B.P. response during soot loading;

FIG. 8—impact of monolith cell density on combustion of retained soot;

FIG. 9—impact of cell density on O2 consumption for combustion of retained soot;

FIG. 10—impact of soot loading temperature for a 900 CPSI flow through monolith;

FIG. 11—displays the B.P. response and O2 consumption traces associated with TPO cycles described in FIG. 10;

FIG. 12—shows an example of a temperature programmed soot loading using a 900 CPSI monolith;

FIG. 13—shows a temperature programmed reaction experiment performed after the temperature programmed reaction soot loading in FIG. 12;

FIG. 14—shows the TPO results for the 900 CPSI monolith after a soot loading with reactive gas and temperature ramp, as per FIG. 12;

FIG. 15—the performance of the coated 900 CPSI monolith is examined in a temperature programmed reactive gas soot loading;

FIG. 16—shows a TPO performed subsequent to the loading cycle of FIG. 15;

FIG. 17—the effect of GHSV on B.P., CO2 evolution and O2 consumption during a soot loading in reactive gas with a simultaneous ramp from 100 to 200° C. for a 900 CPSI part;

FIG. 18a—TPO of samples ex FIG. 17—B.P. response and CO2 evolution; and

FIG. 18b—TPO of samples ex FIG. 17—NO and CO2 evolution and O2 consumption.

DETAILED DESCRIPTION OF INVENTION

FIG. 1 shows a schematic of the synthetic gas bench (SGB) reactor in which the concept trials were executed. Prior to a normal experiment, a monolith core (using a flow through washcoated monolith with 400 or 900 cell per square inch-CPSI) and quartz sleeve packed with quartz wool were placed in the stainless steel reactor as shown. During the course of the subsequent experiments the temperature, pressure drop and O2 content of the reactor were monitored using the respective probes positioned as shown in FIG. 1. Representative sampling of the gaseous reaction byproducts was performed by on-line mass spectrometry with appropriate corrections for m/z overlaps.

The test protocol employed in these studies typically consisted of two phases:

Phase 1) Soot loading cycle: In this portion of the experiment Printex U soot analogue (obtained from Evonik Degussa) is introduced into the reactor via the use of fluidised bed system. The fluidized bed unit contains the soot material and a flow of N2 is based through the base of the bed to establish the fluid condition and thus entrain suspended solid material with the gas flow. The N2/soot flow is then mixed with the reactive gas leg and passes through the reactor, where soot deposition in the monolith may occur. The rate of soot delivery is 0.2 g/hour under typical loading conditions. In order to retain any soot passing through the flow through monolith, i.e. determine soot ‘slip’ or low filtration efficiency, a bed of quartz wool was packed in the outlet position of the reactor.

Phase 2) Regeneration: The sample is purged with dry N2 and then heated in N2/O2 (as a TPO or Temperature Programmed Oxidation) or reactive gas mixture (as described in FIG. 3) to 750° C. and the reactor conditions e.g. back pressure, O2 content, temperature and off gas monitored. Note that any soot trapped in the quartz wool is also combusted during the regeneration but only at high temperatures via the conventional homogeneous combustion pathway, this turn enables a determination of soot ‘slip’ through the monolith and thus determine the impact of cell density on trapping efficiency. An important note is made herein in that prior to any performance testing the various monolith cores were stabilised with respect to performance/aged, this being achieved by an in-situ thermal treatment at 750° C. for four hours.

Examples of regeneration tests for a 900 CPSI monolith coated with the active catalyst described in Examples 1 (method of manufacture of Ag—OS component) and 2 (process for the production of the final coated monolith) are shown in FIG. 2. Herein we examine the impact of the presence of O2 during soot loading on the subsequent combustion (catalytic vs homogeneous) of the soot retained in the catalyst and secondary quartz trap. The performance is broadly similar and shows the presence of two discrete combustion events, one at ca. 250° C., ascribed to direct catalytic soot combustion arising from soot in good contact with the active washcoat and a second event with two features at between 600-700° C. from filter cake combustion and combustion of soot retained in the quartz wool filter. This result mirrors those in SAE 2008-01-0481 and re-confirms the critical importance of catalyst-soot contact for direct oxidation to occur. A difference may be seen in O2 consumption, as determined by the O2 sensor, which suggests that the presence of O2 during the loading is slightly beneficial, possibly through O2 chemisorption on the surface of the soot during loading.

FIG. 3 outlines the various gas compositions employed during the loading and regeneration trials. Hence, in subsequent figures any reference to Reactive gas, for example, refers to a gas composition containing N2, O2, CO, NO and propene in the concentrations listed in line 3.

FIG. 4 compares the back pressure (hereafter B.P.) response for a 400 CPSI (cells per square inch) flow through monolith during a three hour (10800 s) soot loading cycle as a function of either gas environment during loading or temperature and gas environment. The data reflect a clear difference between soot loading at 200° C. under N2/O2 and reactive gas mixture. In the former case there is a continuous increase in the B.P. response of the system (monolith plus quartz wool filter bed), consistent with a systematic deposition and accumulation of soot. In contrast the loading cycle at 200° C. under reactive gases shows a markedly lower rate of B.P. increase during the accumulation cycle. This is consistent with a large decrease in the concentration of soot matter accumulated within the system over time, from which it may be inferred that there is consumption, i.e. oxidation of soot during the loading cycle. Comparison of CO2 evolution data during the loading cycle did show significantly higher CO2 for the reactive gas loading, although given the simultaneous oxidation of CO and propene this data is regarded as a partial corroboration only and subsequent TPO (FIG. 5) is considered more definitive. The trend of decreased B.P. increase is even more evident for loading cycles in reactive gas at 250 and 300° C. Thus, at 250° C. there is only a small increase in the B.P. over the cycle while at 300° C. the B.P. can be seen to actually decrease after the initial loading period. Again both samples showed high levels of CO2 production during loading, consistent with continuous, direct catalytic, oxidation of soot.

Subsequent TPO, shown in FIG. 5, is consistent with the B.P. response trends seen during soot loading (FIG. 4). Herein TPO after the loading cycle at 200° C. in N2/O2 results in a CO2 evolution profile with three features, a small oxidation feature at between 250-350° C., ascribed to catalytic combustion of soot and two large CO2 features at 640° C., due to filter cake combustion, and at >700° C. ascribed to the combustion of soot ‘slip’ i.e. soot that passed through the monolith and was trapped in the quartz wool ‘filter’ toward the outlet of the reactor. Since this quartz wool is located outside of the main heated zone of the furnace any soot trapped herein is only combusted at high temperatures and hence provides a simple diagnostic as to the extent of soot ‘slip’. Thus, in this instance, it can be seen that at lower temperatures, and in the absence of general combustion chemistry, there is a large ‘slip’ of soot through the conventional 400 CPSI part. This soot is accumulated and results in the large B.P. increase seen in FIG. 4. This response may be contrasted with the reactive gas loading at 200° C. In this instance there are again three main CO2 evolution features, catalytic combustion at ca, 300° C., filter cake combustion at 650° C. and ‘slip’ combustion at ca. 710° C. However, the total CO2 production is decreased to a large extent, especially for the highest temperature ‘slip’ event, consistent with increased continuous soot oxidation during the loading. Moreover, the CO2 due to catalytic combustion is significantly increased and filter cake CO2 decreased, reflecting a significant enhancement of catalytic function under simulated exhaust conditions. These trends are further evident in the loading cycles at 250 and 300° C. Both show further decreases in total CO2 production i.e. retained soot and especially decreases in CO2 due to soot ‘slip’. Hence, a comparison of the 300° C. reactive gas loading to the 200° C. N2/O2 loading shows a decrease in CO2 of >80%, i.e. >80% of soot loaded during the cycle is combusted via a continuous, direct catalytic, soot oxidation process.

FIG. 6 compares the O2 concentrations at the reactor outlet during the TPO cycles described in FIG. 5. The data reflects the same trends noted above with decreased O2 consumption being recorded for reactive gas soot loading cycles and for soot loading cycles at 250 and 300° C. For the latter two cycles, there is also the appearance of a feature at ca. 475° C., which does not correlate to any specific CO2 evolution feature. This peak is ascribed to the desorption of NO/NO2 from the catalyst and will be examined in more detail in later figures (see FIGS. 9, 11, 13, 14, 15, 16 and 18b). Note, due to the positioning of the O2 sensor at the outlet of the monolith there is no O2 consumption recorded for the high temperature soot ‘slip’ event.

The impact of monolith cell density, 900 CPSI vs 400 CPSI, on the B.P. response during soot loading is recorded in FIG. 7. Comparison of the loading cycles at 200° C. show general similarities for the two substrates, albeit that the 900 CPSI part shows a slightly higher rate of B.P. increase during the loading cycle, consistent with the expected large impact of soot accumulation in the narrower channels of this substrate.

A comparison of the subsequent TPO reactions after the loading cycle of FIG. 7 is shown in FIG. 8. The data show a clear change in the effectiveness of the technology as a function of cell density. Hence in contrast the previous data for the 400 CPSI part, the 900 CPSI substrate shows a dramatic improvement in soot filtration efficiency, with only very small CO2 evolution features seen for both filter cake and ‘slip’ combustion events. Moreover, the sample also exhibits an increased efficiency with the direct catalytic oxidation feature, hence peak CO2 production from direct catalytic oxidation is now observed at ca. 240° C. versus ca 300-310° C. for the 400 CPSI monolith. Thus by use of the high cell density monolith and active washcoat it is possible to achieve high filtration efficiency, >95% based upon the total CO2 production at T>500° C. versus the 400 CPSI monolith, and also continuous, low temperature, direct catalytic soot oxidation.

Comparable differences in performance with respect to O2 consumption are observed in FIG. 9 for the 900 CPSI vs the 400 CPSI monoliths. Ex 900 CPSI O2 consumption is predominantly seen for T<300° C., with no significant O2 consumption at T>600° C. The converse is seen for the 400 CPSI with a major O2 consumption being recorded at ca. 610-620° C., from filter cake oxidation. Interestingly, all three samples again show an additional feature at ca. 475° C. as per FIG. 6, associated with NOx evolution from the washcoat.

The impact of loading on temperature on subsequent TPO of accumulated soot is the 900 CPSI monolith result in large decreases in accumulated soot. Hence, while loading at 100° C. gives a peak CO2 yield of ca. 8200 counts/s, there is only 6000 c/s and 1,000 c/s for loading cycles at 150 and 200° C. respectively. Moreover, for the loading cycles at 150 and 200° C. there is no evidence for filter cake formation, based upon the absence of any higher temperature CO2 production peak. Indeed integration of the total CO2 evolution from the ex 200 cycle on the 900 CPSI monolith versus the 200° C. N2/O2 cycle on the 400 CPSI monolith indicates >99% of all soot introduced during the loading cycle undergoes direct catalytic combustion, thereby offering the potential for usage of the technology in a ‘real’ life application.

FIG. 11 displays the B.P. response and O2 consumption traces associated with the TPO cycles described in FIG. 10. In all cases the data sets are consistent with the observed CO2 production profiles. Hence, in all cases, CO2 evolution/residual soot combustion is associated with O2 consumption and with a net decrease in B.P. as the monolith channels are cleaned of the restrictive soot particles. The extent of O2 consumption follows the net CO2 production i.e. 100>150>200° C. Again, all samples the secondary NOx related feature at 475° C. The B.P. responses also appear to reflect the conditions of soot loading with the ‘relaxation’ response being sharpest for the 200° C. cycle, then 150° C. and finally 100° C. loading cycle, again consistent with the residual soot retention for the various tests.

In order to better mimic driving conditions we performed soot loading cycles under conditions of dynamic temperature changes. Hence, FIG. 12 shows an example of a temperature programmed soot loading using a 900 CPSI monolith. In this test there was simultaneous soot loading cycle in full reactive gas mixture with heating of the sample from 100° C. to 200° C. The data shows the expected CO (and propene) light-off curves, which were again found to be coincident with soot combustion, as reflected in the peak then decay seen for CO2 production and O2 consumption traces. In this experiment, as in all tests performed during this study, there was no production of CO during the oxidation of soot (determined by analysis of the corrected mass spectrometer peak at m/z 12 in which one may account for the background and dynamic contributions of mass fragmentation from CO and CO2). The continuous combustion of soot also helps to account for the overshoot seen in the bed thermocouple, which was found to be ca. 245° C. versus the set point of 200° C.

FIG. 13 shows a temperature programmed reaction experiment performed after the temperature programmed reaction soot loading in FIG. 12. The protocol for this test entailed cooling the sample in-situ to 100° C. in flowing N2, after the soot loading cycle was completed, upon stabilisation at 100° C., the full reactive gas mixture was then reintroduced, and the sample heated to 750° C., per standard method. The data shows the expected light-off of CO (propene also undergoes light-off but the signal is omitted for clarity with CO, NO and NO2 traces) as evidenced by the responses in CO, CO2 and also the O2 sensor. Interestingly, there is again a peak of CO2 production at ca. 225° C., and then a decrease, this feature is ascribed to the combustion of the residual soot retained on the part. During this combustion event there is no significant change in the B.P. of the sample, suggesting the retained soot is at such a low level as to not result in any meaningful contribution to system backpressure. Finally, there is a very small CO2 evolution at 475° C., this latter feature is coincident with the apparent O2 consumption event noted in previous tests (see FIGS. 9 and 11) but also with a NOx (NO and NO2) desorption event. This event is attributed to the intrinsic NOx scavenging and release properties of the Ag—OS material, as described in SAE 2008-01-0481, application Ser. Nos. 12/363,310 and 12/363,329. Thus, during the loading cycle and in the subsequent temperature programmed reaction, any NO2 that is generated which would normally result in ‘de-coupling’ of catalyst-soot contact, is trapped on the highly dispersed Ag centres and retained to high temperatures where it is released in the plume observed. The plume of desorbed NOx then may react with any traces of soot remaining on the part, particularly any species that are spatially distant from the catalyst surface i.e. with ‘poor’ contact.

To further examine the impact of temperature during a reactive soot loading cycle additional tests were performed. FIG. 14 shows the TPO results for the 900 CPSI monolith after a soot loading with reactive gas and temperature ramp, as per FIG. 12. Employing the TPO protocol rather than the full reactive gas mix simplifies the chemistries and result traces. Thus, in the TPO protocol there are no light-off features but rather a series of peaks due to the various phenomena occurring over the catalyst versus temperature. Firstly there is a CO2 production peak, attributed to the combustion of residual retained soot. This peak is centred at 300° C., ca. 75° C. higher than in the temperature programmed reaction case. This reflects the important contribution of the exothermicity of CO and HC light-off in facilitating lower temperature soot oxidation. Hence, in the reactive gas temperature ramp, as the CO and HC begin to combust they generate a thermal bloom with the monolith which is sufficient to overcome the activation energy barrier for the initiation of soot oxidation. Then once the combustion of soot is initiated, a further exotherm is generated and the resulting thermal ‘cascade’ is sufficient to result in very high soot conversion rates, this process is the related of the method for lower temperature soot oxidation as described in US 2005/0282698 A1. The soot oxidation event in this instance is correlated to a very small B.P. ‘relaxation’. The sample is also seen to desorb water, this release being associated with desorption of combustion by-products from HC oxidation. Finally, at 475° C. one again sees the NOx desorption/apparent O2 consumption event associated with the Ag—OS scavenging function, however in this instance there does not appear to be any significant associated CO2 production from the combustion of trace soot in poor contact. However, what is clear is that under the loading conditions employed there is consistently high activity towards direct soot oxidation resulting in very low levels of residual soot remaining on the 900 CPSI monolith, again confirming the potential for the approach for continuous, direct catalytic, soot oxidation.

In FIG. 15 the performance of the coated 900 CPSI monolith is examined in a temperature programmed reactive gas soot loading. In this instance the maximum temperature employed was 500° C. (ramping from 50° C.). In the case the CO (and HC) light-off traces are very clearly represented, as is the associated O2 consumption. Again the CO2 evolution trace shows an increase to a peak at ca. 250° C. before decreasing to a steady state value. This is again consistent with the active catalytic combustion of retained soot. Hence, in this and all other regards the performance trends replicate previous findings, including the NOx scavenging/apparent O2 consumption at ca. 475° C.

FIG. 16 shows a TPO performed subsequent to the loading cycle of FIG. 15. Herein there are no significant reaction or desorption events evident. In particular, there is no additional CO2 production, no high temperature soot ‘slip’ phenomenon, i.e. the data is consistent with complete conversion of any soot loaded during the loading cycle, further confirming the high effectiveness of the technology.

Next the impact of Gas Hourly Space Velocity on performance was examined. Hence, FIG. 17 contrasts reactive gas loading cycles, with temperature ramp (100-200° C.) under the standard GHSV of 15000 h−1 versus a GHSV of 25000 h−1 (versus monolith volume). It is emphasised at this point that the soot delivery rate in both tests as determined by the flow rate through the fluidised bed was constant in both cases and the increase in GHSV was achieved by increasing the flow rates of the various gases within the reactive gas manifold. Analysis of the subsequent data from both tests show comparable response with response to gas phase chemistry, with CO (and HC) light-off being unaffected, as evidenced by the comparable CO2 responses. There is an offset in the O2 sensor values, possibly due to the total increased flow employed in the high GHSV test, but again the dynamic responses are identical. After completion of light-off there is some difference in the CO2 responses, with the low GHSV test showing higher net CO2. Coincident to this, the B.P. response of the high GHSV test shows a steady increase, this is ascribed to a combination of the higher net static back pressure observed due to the higher flow rate but also to an increase in the net rate of soot accumulation during the test. This raises questions regarding location of soot, i.e. is the B.P. increase due to soot ‘slip’ or is the soot still retained on the part, and also what is the maximum effective rate of soot deposition that may be employed and still achieve high continuous soot combustion rates.

In FIG. 18a/b these questions are answered. Herein the subsequent TPO cycles loading cycles. The data show only CO2 production at lower temperatures with a peak at ca. 300° C. Hence even under the conditions of higher flow there was no soot slip i.e. all soot introduced was retained with the monolith. The decrease in rate of soot oxidation is thus ascribed to an exothermal effect with the increasing flow rate through the part during loading resulting in a net ‘dilution’ of the exotherm cascade which is believed to be critical for the propagation of soot burn. However, as indicated the ‘excess’ soot generated due to this process is merely retained unreacted on the part, thus in the subsequent TPO the soot oxidation follows the same profile as the lower GHSV case and there is simply an increase in the net CO2 production. This is also reflected in the B.P. responses with the sample loaded under high GHSV showing a more rapid and larger B.P. ‘relaxation’ than the sample loaded under lower GHSV. Similarly the NOx evolution response is larger for the high GHSV sample, reflecting the higher mass fraction of NOx exposure during the test. This in turn results in the small differences in apparent O2 consumption, as recorded by the O2 sensor. Thus, to conclude while there is an impact of GHSV on activity, the impact is not catastrophic and the monolith retains its ability to either combust or trap all soot at lower temperatures and then to facilitate its combustion at temperatures still within the normal operating window of a conventional vehicle, i.e. the system still provides effective soot filtration and combustion without recourse to conventional active regeneration strategy.

DETAILED DESCRIPTION OF THE INVENTION

The present invention relates to a method and apparatus for the continuous/semi-continuous direct catalytic, oxidation of diesel particulate matter by the combination of base metal modified Oxygen Storage (OS) materials in association with turbulent flow/high cell density flow through monoliths. The particular combination of the base metal modified OS direct soot oxidation catalyst with the flow through monolith provides a synergy which enables high conversion of particulate matter without the backpressure penalty introduced by the conventional DPF. It is believed that the synergy arises from the ability of the active OS material to combust soot at lower temperatures than in conventional systems, which in turn is facilitated by the decreased thermal mass of the conventional substrate, with the latter still providing sufficient geometric surface area for soot deposition and reaction. Large improvements in lower temperature activity can be obtained in marked contrast to the conventional wall flow DPF wherein large thermal mass of the substrate, particularly for SiC DPF, inhibits initiation and especially propagation of soot combustion.

The present invention represents a significant advance in the development of a method and apparatus for the (semi) continuous, direct catalytic, oxidation of diesel particulate matter may be realized by the combination of base metal modified Oxygen Storage (OS) materials with a conventional flow substrate. The substrate is selected from a range of ceramic or metallic technologies upon which the active washcoat is disposed. Such substrates can be metallic parts, ceramic or metal foams.

More particularly and in a further aspect, the present invention relates to a synergistic combination of a catalyst and a substrate for the filtration and continuous, direct catalytic, oxidation of diesel particulate matter at low temperatures. The catalyst comprises catalytically active precious metal (Pt, Pd, Rh or combinations thereof), a host cerium-based solid solution which is a substantially phase pure cubic fluorite (as determined by x-ray diffraction method) of the CeZrOx type which is well known in the art and a refractory oxide support, e.g. (γ)Al2O3, ZrO2 or other known oxide support. The CeZrOx is further modified by the incorporation of an active base metal, e.g. Ag, Cu etc. as disclosed in application Ser. No. 12/363,310. The catalyst further comprises a monolith substrate, of conventional design, wherein the monolith is an inert ceramic or metal substrate upon which the active catalyst formulation/washcoat is disposed. The monolith substrate is further characterised by a high cell density, i.e. a large number of active channels per unit area, for effective synergy a value of >600 cells per square inch. In the case of a metallic substrate the active washcoat may be applied to the perforated, punched and embossed metal foils (e.g. TS, LS, PE and MX type systems; see for example U.S. Pat. No. 6,689,327) with beneficial effect.

The combined active washcoat and monolith system may be applied to the challenge of particulate emission control catalysts for diesel (or other fuel lean) or potential gasoline (stoichiometric) application. The particular example described herein is for the application of these materials in the area of continuous, direct catalytic oxidation of diesel particulate matter upon its interaction with the high cell density substrate. These benefits arise in this application due to the aforementioned synergies arising from the high cell density monolith and the new generation of modified OS materials. The latter has been previously been demonstrated as having benefits in affecting either lower temperature regeneration/oxidation of soot or an increased regeneration efficiency at a lower temperature as compared to non-modified OS materials (application Ser. Nos. 12/363,310 and 12/363,329). Now in combination with a conventional flow monolith of appropriate architecture, it becomes possible to realise a completely passive particulate control catalyst.

It should be further noted that the terms “first”, “second” and the like herein do not denote any order of importance, but rather are used to distinguish one element from another, and the terms “a” and “an” herein do not denote a limitation of quantity, but rather denote the presence of at least one of the referenced items. Furthermore, all ranges disclosed herein are inclusive and combinable (e.g., ranges of “up to about 25 weight percent (wt. %), with about 5 wt. % to about 20 wt. % desired, and about 10 wt. % to about 15 wt. % more desired” is inclusive of the endpoints and all intermediate values of the ranges, e.g. “about 5 wt. % to about 25 wt. %, about 5 wt. % to about 15 wt. %” etc.

The details regarding the synthesis, characterization and preferred compositions, structures, dopant levels etc for the Cerium-containing mixed oxide/solid solution material are detailed in Ser. Nos. 12/363,310 and 12/363,329. Preferably, the solid solution contains a cationic lattice with a single-phase, as determined by standard x-ray diffraction method. More preferably this single-phase is a cubic structure, with a cubic fluorite structure being most preferred. Additionally, it is noted that the doping process may be performed without formation of an additional bulk phase, as determined by XRD. In various embodiments, the OS material may include those OS materials disclosed in U.S. Pat. Nos. 6,585,944; 6,468,941; 6,387,338 and 6,605,264 which are herein incorporated by reference in their entirety. However, the flexibility of the basic exchange provides for the modification of all current known cerium oxide and Ce—Zr-based solid solution materials to be thusly modified and enhanced.

As indicated, the OS materials modified by the doping method shall preferably be characterized by a substantially cubic fluorite structure, as determined by conventional XRD methods. The percentage of the OS material having the cubic structure, both prior and post exchange, is preferably greater than about 95%, with greater than about 99% typical, and essentially 100% cubic structure generally obtained (i.e. an immeasurable amount of tetragonal phase based upon current measurement technology). The exchanged OS material is further characterized in that it possess large improvements in durable redox activity with respect to facile oxygen storage and increased release capacity e.g. as determined by H2 Temperature Programmed Reduction (TPR) method. Thus, for Cu exchanged solid solutions, for example, the reduction of Ce+Cu is observed to occur at a temperature of about 300 to about 350° C. lower than would occur in the absence of the Cu dopant (see application Ser. No. 12/363,310).

In an exemplary embodiment, an active soot oxidation catalyst comprising a precious group metal or metals (Pt, Pd, Rh and combinations thereof), a base metal doped cerium-oxide containing solid solution and a refractory oxide carrier all of which together are employed as a coating, e.g., disposed on/in an inert substrate or carrier, the substrate or carrier being characterized by a high number of channels or cells per unit area or by the its ability to introduce turbulent flow due to the construction of its internal flow channels. Exhaust gas treatment devices can generally comprise housing or canister components that can be easily attached to an exhaust gas conduit and comprise a substrate for treating exhaust gases. The housing components can comprise an outer “shell”, which can be capped on either end with funnel-shaped ‘end-cones’ or flat ‘end-plates’, which can comprise ‘snorkels’ that allow for easy assembly to an exhaust conduit. Housing components can be fabricated of any materials capable of withstanding the temperatures, corrosion, and wear encountered during the operation of the exhaust gas treatment device, such as, but not limited to, ferrous metals or terrific stainless steels (e.g., martensitic, terrific, and austenitic stainless materials, and the like).

Disposed within the shell can be a retention material (“mat” or “matting”), which is capable of supporting a substrate, insulating the shell from the high operating temperatures of the substrate, providing substrate retention by applying compressive radial forces about it, and providing the substrate with impact protection. The matting is typically concentrically disposed around the substrate forming a substrate/mat sub-assembly.

Various materials can be employed for the matting and the insulation. These materials can exist in the form of a mat, fibres, preforms, or the like, and comprise materials such as, but not limited to, intumescent materials (e.g., a material that comprises vermiculite component, i.e., a component that expands upon the application of heat), non-intumescent materials, ceramic materials (e.g., ceramic fibers), organic binders, inorganic binders, and the like, as well as combinations comprising at least one of the foregoing materials. Non-intumescent materials include materials such as those sold under the trademarks “NEXTEL” and “INTERAM 1101HT” by the “3M” Company, Minneapolis, Minn., or those sold under the trademark, “FIBERFRAX” and “CC-MAX” by the Unifrax Co., Niagara Falls, N.Y., and the like. Intumescent materials include materials sold under the trademark “INTERAM” by the “3M” Company, Minneapolis, Minn., as well as those intumescent materials which are also sold under the aforementioned “FIBERFRAX” trademark.

Housings as described above are well known and understood by those skilled in the art.

The substrates or carrier employed in this invention can comprise any material designed for use in a spark ignition or diesel engine environment having the following characteristics in addition to the high cell density/turbulent flow requirement stated previously: (1) capability of operating at temperatures up to about 600° C. and up to about 1,000° C. for some applications, depending upon the device's location within the exhaust system (e.g., manifold mounted, close coupled, or underfloor) and the type of system (e.g., gasoline or diesel); (2) capability of withstanding exposure to hydrocarbons, nitrogen oxides, carbon monoxide, particulate matter e.g. soot and the like, CO2, and/or sulfur; and (3) have sufficient surface area and structural integrity to support a catalyst, if desired. These materials should be inert under the conditions imposed on them when in use. Some possible materials include cordierite, silicon carbide, metal, metal oxides; e.g. alumina and the like, glasses and the like, and mixtures comprising at least one of the foregoing materials. Some suitable inert ceramic materials include ‘Honey Ceram’, commercially available from NGK-Locke, Inc, Southfield, Mich., and ‘Celcor’, commercially available from Corning, Inc., Corning, N.Y. These materials can be in the form of foils, perform, mat, fibrous material, monoliths e.g. a honeycomb structure, and the like, other porous structures e.g., porous glasses, sponges, foams, pellets, particles, molecular sieves, and the like (depending upon the device), and combinations comprising at least one of the foregoing materials and forms, e.g., metallic foils, open pore alumina sponges and porous ultra-low expansion glasses. Furthermore, these substrates can be coated with oxides and/or hexaaluminates, e.g. stainless steel foil coated with a hexa-aluminate scale.

Although the substrate can have any size or geometry, within the previously defined limits, the size and geometry are preferably chosen to optimize surface area in the given exhaust gas emission control device design parameters. Typically, the substrate has a honeycomb geometry, with the combs through-channel having any multi-sided or rounded shape, with substantially square, triangular, pentagonal, hexagonal, heptagonal, or octagonal or similar geometries preferred due to ease of manufacturing and increased surface area.

The exhaust gas treatment devices can be assembled utilizing various methods. Three such methods are the stuffing, clamshell, and tourniquet assembly methods. The stuffing method generally comprises pre-assembling the matting around the substrate and pushing, or stuffing, the assembly into the shell through a stuffing cone. The stuffing cone serves as an assembly tool that is capable of attaching to one end of the shell. Where attached, the shell and stuffing cone have the same cross-sectional geometry, and along the stuffing cone's length, the cross-sectional geometry gradually tapers to a larger cross-sectional geometry. Through this larger end, the substrate/mat sub-assembly can be advanced which compresses the matting around the substrate as the assembly advances through the stuffing cone's taper and is eventually pushed into the shell.

Exhaust gas treatment devices comprising the doped solid solutions can be employed in exhaust gas treatment systems to provide both an active soot combustion catalyst but also a NOx adsorption function, and thus specifically reduce a concentration of undesirable constituents in the exhaust gas stream. For example, as discussed above, an exemplary catalyst system can be formed utilizing the doped OS as a catalyst component, wherein the catalyst system is disposed on a substrate, which is then disposed within a housing. Disposing the substrate to an exhaust gas stream can then provide at least a NOx storage function, and desirably even reduce the concentration of at least one undesirable constituent contained therein.

According to one embodiment of the present invention, the catalyst does not conform the standard architecture of a CDPF or Diesel NOx Particulate Trap and hence does not comprise a porous substrate having alternating channels. Rather the preferred configuration of the catalyst is as a conventional ‘flow through’ monolith, of high unit cell count per unit area, upon which is disposed the active catalyst washcoat. The combination of the active washcoat with the high internal surface area and turbulent deposition mechanism is sufficient to facilitate retention and continuous particulate oxidation under conventional operating temperatures and flows of a diesel/compression ignition vehicle.

EXAMPLES

The benefits obtained by the active washcoat employing the doped Cerium containing oxide and high cell density monolith are clearly evident in FIGS. 4-18b, wherein the benefit of enhanced redox performance of the doped OS, in combination with an appropriate substrate result in high rates of direct soot combustion under conditions appropriate for vehicular application. It should be stressed that the redox promotion obtained by base metal doping is observed for both a range of cationic dopants and a range of OS compositions, and the data included herein for the 2Ag—OS is merely a representative example.

The data herein reflect a systematic study of the various parameters considered relevant to achieve the desired aim of continuous, low temperature, direct catalytic soot oxidation. The impacts of these parameters on performance are summarized, with reference to specific case data as follows:

a) Reactivity of soot: The reactivity of soot e.g. soluble organic fragment has been shown to play a large role in determining the reactivity of soot and thus the effective performance of soot oxidation catalysts (Atmos Env, vol. 15 (1), 1981, 91-94, SAE paper 2008-01-0481, App Catal B, vol. 75 (1-2), 2007, p 11-16, etc.). Indeed, comparison of conventional soot TGA shows an increase in Tmax (temperature for maximum rate of soot combustion) of ca. 50° C. for Printex U cf. ‘real’ diesel soot collected from a vehicle (SAE 2008-01-0481). Thus, in this study Printex U soot analogue was employed to specifically remove this variable from any discussion. Hence, all particulate matter combusted during these tests may be considered to be equivalent in reactivity and thus present no inherent bias in any given data set. Moreover, the oxidation of the Printex material may be considered to be a ‘worst case’ scenario i.e. its oxidation is representative of combustion of very ‘dry’, refractory carbonaceous material high in graphitic content and low in SOF. Thus, the promising data herein reflect a true performance advantage of interest to real world applications.

b) Gas environment during soot accumulation: There is a clear impact of reactive gas chemistry on catalyst performance both during the loading cycle as shown in FIGS. 2, 4, 5 and 6 but also the nature of the gas atmosphere can be seen to impact the regeneration, as is evident from contrasting TPO versus temperature programmed reaction burn out protocols (FIGS. 5, 8, 10, 13, 14, 16 and 18a/b). This impact is attributed to a combination of heat transfer and catalyst activation. One heat transfer component arises due to the external heating of the active catalyst arising from the combustion of the significant levels of fuel components within the reactive gas mixture, principally CO and HC. This energy is retained within the washcoat, resulting in the hotter than expected bed temperatures observed, and thus helping to overcome the activation energy barrier to catalytic soot oxidation. A second combined heat transfer and catalyst activation component is arises from the activation of the redox oxide arising from its participation in the CO oxidation process. It has been shown that the doped cerium oxides are effective oxidation catalysts, even in the absence of PGM, and can facilitate CO oxidation at low temperatures (DP-316440). In doing so the catalyst O ion transport function is activated, and energy released at the active site of CO oxidation. The subsequent re-oxidation of the depleted oxygen results in a further exotherm, distributed throughout the entire structure of the OS, in a sense further priming the OS to initiate soot oxidation. This mechanism forms part of the basis of US 2005/0282698 A1, wherein a fuller explanation may be found.

c) Static temperature effect during soot accumulation: Obviously thermal energy/temperature is required to overcome the activation energy barrier for catalytic soot combustion. Hence, with increasing inlet temperature, there is a concurrent increase in rate of catalytic oxidation and hence decreases in soot slip and mass of retained soot on the monolith irrespective of all other factors (FIGS. 4-18b).

d) The role of cell density of the monolith: This is a key factor for the invention, with the use of higher cell density/increased cell count per unit area, resulting in a large enhancement of catalyst performance (FIGS. 7-9). Thus by merely substituting the 400 CPSI monolith with, the 900 CPSI one results in a dramatic improvement in soot filtration efficiency (>95% based upon the total CO2 at T>500° C. vs 400 CPSI), the ability to circumvent soot ‘slip’ through the monolith i.e. no high temperature CO2 production due to soot passing through the monolith and being retained in the quartz wool filter, and also a small decrease in temperature required for soot combustion (which is attributed to the higher effectiveness of the oxidation washcoat). The impact of cell density also has a positive synergy with temperature with larger net performance gains being observed for the 900 CPSI monolith for higher temperature than for the 400 CPSI system.

e) The role of NOx on catalyst performance: The ability of the metal doped OS material to scavenge NOx at low temperatures and release the retained species at higher temperatures (see FIGS. 13, 14, 15 and 18b) is of particular importance. This capability effectively disables the ‘de-coupling’ mechanism of related to NO2 mediated soot oxidation, which has been shown to destroy the intimate contact between catalyst and soot required for direct catalyzed soot oxidation (see SAE paper 2008-01-0481, U.S. patent application Ser. No. 12/363,329). Indeed by use of the metal doped OS material it appears that one can, in selected cases, also employ the NOx desorption plume beneficially to remove trace particulate matter that is in poor contact i.e. spatially discrete/removed from the active catalyst e.g. see FIG. 13. However, it must be stressed this is not the primary catalytic process responsible for the high activity seen at low temperatures but is rather an additional minor benefit.

g) Dynamic temperature effect during soot accumulation: Comparison of static temperature soot loading and regeneration cycles (FIGS. 4, 5, 7, 8 and 11) versus loading and regeneration cycles with dynamic temperature change i.e. temperature ramp (FIGS. 12, 13, 14 and 15) illustrate a further manifestation of energy with the process. Thus it may be seen that increases in temperature during the soot accumulation process, combined with the specific exotherm associated Light-off of CO and HC, result in a further improvement in performance compared to a simple static temperature loading. This is ascribed to the combination of exothermal effects, which we have dubbed thermal bloom propagation, described previously and explained in more detail in US 2005/0282698 A1.

h) The effect of GHSV and flow velocity: Clearly, the residence time of the particulate matter within the monolith is an important factor. Thus, the longer the particulate reside within the monolith channels, the greater is the probability of interaction with the active washcoat coated on the walls and, hence, the higher probability of retention and reaction. Moreover, since the particulates are entrained by the flow i.e. derive their kinetic energy from Brownian collisions, at higher flow rates, the velocity of the particulates are higher. This both decreases residence time within the monolith but also provides a force driving the following regime to be more laminar and less turbulent, thereby decreasing the possibility of particulate to wall interaction. These hypotheses are consistent with data in FIGS. 17, 18a and 18b. In addition at higher flow velocities there is an increase in energy transport out of the monolith, i.e. the local exotherms are diminished due to increased Brownian collisions transferring kinetic energy to molecules leaving the flow channel. Hence, at higher GHSV the rate of soot oxidation is somewhat decreased and results in increased formation of retained soot species. However, under the conditions examined, the increased flow was not enough to result in soot ‘slip’, prevent extensive direct catalytic oxidation or indeed prevent complete regeneration in the subsequent burn out cycle. Finally, it should be noted that during the burn out cycle the temperatures required for combustion of the retained soot mass fraction was still significantly lower than the >600° C. employed in conventional DPF active regeneration. Indeed the temperatures required were still only of the order of 300-330° C., i.e. temperatures easily within the normal operational window of a diesel vehicle. Hence, the concept of direct catalytic soot oxidation is applicable to vehicular application.

EXAMPLES

The procedure for making 100 grams of 2% Ag(NH3)2 OS, as employed in the test technology is as follows:

1. Weigh 100 g of OS, correct for moisture content (ca. 1.5% water).

2. Weigh 3.15 g of silver nitrate crystals. One must compensate for the percentage of metal in the nitrate salt or solution used. Silver nitrate is 63.52% silver.

3. Dissolve silver nitrate in 50 g deionised water. The amount of water used is determined by the water adsorption capacity of the mixed oxide used. This is generally between 0.5 and 0.5 g water per gram mixed oxide.

4. Add concentrated NH4OHaq (30% ammonia) to the silver nitrate solution, dropwise, until a clear silver diamine solution is obtained. Solution will first turn brown-black, then clear upon excess addition of ammonium hydroxide.

5. Add silver diamine solution to mixed oxide powder. Mix thoroughly to produce homogeneous and even-colored moist powder.

6. Allow powder to rest at room temperature for one hour.

7. Dry in oven at ˜110° C. for ca. two hours or until dry.

8. Calcine in furnace at 540° C. for four hours in air.

OS=40% CeO2; 50% ZrO2/HfO2; 5% La2O3; 5% Pr6O11

The procedure for producing the active washcoat and producing the 400 and 900 CPSI parts tested in this study is as follows: Slowly add alumina with milling to a d50 of 7 microns (±1), d90=20-25 and 100% pass<60 microns. Adjust pH to 3.0-3.5 and specific gravity to allow one pass coating then coat monolith in one pass and calcine at temperatures ≧540 C for ≧one hour. Next slurry required 2Ag—OS in DI water, mill at the natural pH of the material to a d50 of 2±0.3, d90 of <10 microns and 100 pass<30 microns. Prevent pH decreasing below 4 by addition of base. Next pre-mix pt nitrate and pd nitrate solutions for 15 minutes. To this mixture add dilute sugar solution and mix for a minimum of 30 minutes; add to Ag—OS slurry within 60 minutes of initial mixing to avoid precipitation of metal. Add PGM sugar solution dropwise to Ag—OS slurry vortex. Prior to addition slurry must be at a pH of 5.5-6.0 and during metal addition, prevent slurry from going to pH values below 4.0 with the judicious use of base. Stir two hours to allow full chemisorption. Adjust pH and specific gravity to allow one pass coating then coat monolith in 1 pass and calcine at temperatures ≧540° C. for ≧one hour.