WO2005077498A1 - Sulfur oxide adsorbents and emissions control - Google Patents

Abstract

High capacity sulfur oxide absorbents utilizing manganese-based octahedral molecular sieve (Mn-OMS) materials are disclosed. An emissions reduction system for a combustion exhaust includes a scrubber (24) containing these high capacity sulfur oxide absorbents located upstream from a NOx trap (26) or particulate filter.

Description

SULFUR OXIDE ADSORBENTS AND EMISSIONS CONTROL

RELATED APPLICATION DATA This application claims the benefit of U.S. Application Ser. No. 10/771,866, filed February 4, 2004, and this application claims the benefit of U.S.

Application Ser. No. , filed February 3, 2005, both of which are incorporated herein by reference.

GOVERNMENT RIGHTS This invention was made with Government support under Contract Number

DE-AC06-76RLO1830 awarded by the U.S. Department of Energy. The Government has certain rights in the invention.

TECHNICAL FIELD The present invention is generally related to pollution control, and more particularly, but not exclusively, is directed to high capacity sulfur oxide adsorbents and uses therefore in emissions control.

BACKGROUND The combustion waste gases (i.e. the exhaust) of thermal power plants, factories, on-road vehicles, diesel generators, and the like contain SO_x and NO_x. State and federal regulations limit the permissible amounts of these emissions because they create environment problems, such as acid rain. Accordingly, there is a continual need for improvements in the cost effective and efficient control of these emissions. One mechanism for limiting NO_x and SO_x emissions is to remove or scrub the pollutants from the exhaust gas using an absorption bed, trap or similar device. Because many NO_x traps have been found to be poisoned by the presence of SO_x, it is important to remove as much SO_x from the exhaust gas as possible. However, as compared to the large volume of studies on ?NO_x reduction, sulfur oxide removal using solid adsorbents is an area in need of scientific advancement. For example, certain types of materials have been identified as possible solid absorbents for use in a SO_x adsorption bed or traps, for example calcium oxide and alkalized alumina (Na/Al₂O₃ or K/Al₂O₃), copper-based adsorbents, e.g. Cu/Al₂O , promoted metal oxides, e.g. TiO₂, Al₂O₃, ZrO₂, promoted cerium oxide (La- or Cu- doped CeO₂), and supported cobalt (Co/Al₂O₃). Unfortunately, over the temperature range of about 250°C to 475°C, these materials typically have a relatively low absorption capacity. For example, their total adsorption capacity of SO₂ is typically less than about 10 wt% based on the weight of the absorbent, and their breakthrough absorption capacity can be substantially lower, depending on operating conditions. As it is combustion in this temperature range that leads to a significant portion of the total SO_x emissions, a greater adsorption capacity at these temperatures is needed. One approach to increasing the absorption capacity of SO_x absorption beds is to provide an oxidation catalyst upstream or admixed with the bed so as to convert most of the SO₂ to SO₃, since SO₃ is generally more readily adsorbed than SO₂ due to its formation of stable surface sulfates. However, the cost of recovery of the oxidation catalyst (frequency a precious metal) and the relatively poor conversion efficiency of SO₂ to SO₃ at temperatures below about 300°C limits the effectiveness of this approach as well. Accordingly, there is a need for solid SO_x absorbents with high absorption capacity at lower temperatures and which reduce or eliminate the need for separate oxidation catalysts. In one aspect the present invention addresses this need and provides a major improvement to SO_x absorption for emissions control.

SUMMARY The present invention provides systems and techniques for SO_x emission control. While the actual nature of the invention covered herein can only be determined with reference to the claims appended hereto, certain aspects of the invention that are characteristic of the embodiments disclosed herein are described briefly as follows. In one form, the invention concerns materials for absorbing, trapping, or otherwise eliminating oxides of sulfur from gases, for example those sulfur oxides present in exhaust gases of internal combustion engines. The materials are mixed oxides having a framework of metal cations M each surrounded by 6 oxygen atoms wherein the octahedra (MO₆) thus ormed are connected together by edges and vertices generating a structure that produces channels in at least one direction in space. The sides of these channels are formed by linking the octahedra (MO₆), which connect together by the edges, these sides connecting themselves together via the vertices of the octahedra. Thus, the width of the channels can vary depending on whether the sides are composed of 2, 3 and or 4 octahedra (MO₆), which in turn depends on the mode of preparation. This type of material is known by its acronym OMS, Octahedral Molecular Sieve. In accordance with one form of the invention, the materials are selected so that they have a structure that generates channels either with a square cross section composed of, for example, one octahedra by one octahedra (OMS l x l), two octahedra by two octahedra (OMS 2 x 2) or three octahedra by three octahedra (OMS 3 x 3), or with a rectangular cross section composed of, for example, two octahedra by three octahedra (OMS 2 x 3). Thus, certain of the materials will have a pyrolusite (OMS l x l), hollandite (OMS 2 x 2), romanechite (OMS 2 x 3) or todorokite (OMS 3 x 3) type structure. Other OMS structures, such as 1x3, 1x4 or 2x4, 3x4 or 4x4, are also contemplated, though the 2x2 structure has been found to be particularly effective in certain applications. The OMS materials of the present invention are preferably manganese based (Mn-OMS), which means that a major portion of the metal cation M is manganese (Mn). An element is in the majority when it satisfies the following formula: (n_maj/∑n_M)>l N, where n_Mis the number of atoms of element (M) and N is the number of different elements (M) composing the framework, n_maj being the highest number of atoms of element (M). Most preferably, over 50% of the elements M are manganese for example at least 75% or at least 90% of the element M by mole. Preferably the manganese has an oxidation number between +2 and +4. The balance of element M can include one or more elements from groups HEB to mA in the periodic table such as Zn²⁺, Co²⁺, Ni²⁺, Fe²⁺, Al³⁺, Ga³⁺, Fe³⁺, Ti³⁺, In³⁺, Cr³⁺, Si⁴⁺, Ge⁴⁺, Ti⁴⁺, Sn⁴⁺, and Sb⁵⁺ and combinations thereof. The material used in this invention has a characteristic structure of high surface area and may be capable of oxidizing SO₂ to SO₃ and converting SO₃ to a sulfate. In certain cases, but not all cases, another cation, such as H^", NH , Li⁺, Na⁺, Ag⁺, K⁺, Rb⁺, Tl⁺, Cs⁺, Mg²⁺, Ca²⁺, Sr^2"", Ba²⁺, Ra²⁺, Cu²⁺, Pb²⁺, locates in the channels in the OMS structures. The absorbing phase of materials with type OMS 2*2, OMS 2*3 and OMS 3*3 of one aspect of the invention has a three-dimensional structure that generates channels in at least one direction in space, is composed of octahedra (MO₆) and comprises: at least one element (M) selected from the group formed by elements from groups πiB, IVB, VB, NEB, NUB, VIII, IB, IIB, HIA of the periodic table and germanium, each element M being coordinated with 6 oxygen atoms, and located at the center of the oxygen octahedra, wherein a major portion of M is manganese; and at least one element (B) selected from the group formed by the alkali elements IA (such as K+), the alkaline-earth elements HA, the rare earths IIIB, transition metals (such as Ag+) or elements from groups IIIA and INA, element B generally being located in channels in the oxide structure. More particularly, elements M are selected from scandium, titanium, zirconium, vanadium, niobium, chromium, molybdenum, tungsten, manganese, iron, cobalt, nickel, copper, zinc, aluminum, gallium, and mixtures thereof. The average charge (oxidation number) carried by the cation or cations M from groups IHB to HIA is preferably about +3.5 to +4. Preferably, at least about 50% of the elements (M) in the material are manganese, titanium, chromium, aluminum, zinc, copper, zirconium, iron, cobalt, and/or nickel. More preferably, over 50% of the elements (M) are manganese, chromium, copper, iron, titanium and/or zirconium. In one form, manganese composes at least about 50% of the M element by mole, for example at least 75% or 90% of the M element. Other elements M from groups MB to ID?A, including the transition metals, can be added in minor quantities as dopants. Preferably, the elements from groups IIIB to HIA added in minor quantities are selected from scandium, titanium, zirconium, vanadium, niobium, chromium, molybdenum, tungsten, manganese, iron, cobalt, nickel, copper, zinc, aluminum, gallium, and mixtures thereof. Elements (B) belong to the group formed by the alkali elements IA, alkaline-earth elements UA, rare earth elements IIIB, transition metals (such as elements from group IB, i.e. silver) and elements from groups IIIA and IV A. They are located in the channels of the material. Preferably, metal B is selected from the group formed by potassium, silver, sodium, magnesium, barium, strontium, iron, copper, zinc, aluminum, rubidium and calcium and mixtures thereof. In a preferred embodiment, the material is of the formula X_aMn₈Oι₆ wherein X is an alkali metal, an alkaline earth metal, or a transition metal, and a is between 0.5 and 2.0. In still further preferred forms the material is cryptomelane, silver hollandite, or a combination thereof. A number of different methods exist for preparing these materials (see references 3 and 4 below, for example). They may be synthesized by mixing and grinding solid inorganic precursors of metal oxides (metals M and B), followed by calcining. The materials can also be obtained by heating solutions of precursor salts to reflux, drying and calcining, by precipitating precursor salts by the sol-gel method, or by hydrothermal synthesis which consists of heating an aqueous solution containing the elements constituting the final material under autogenous pressure. The materials obtained from these syntheses can be modified by ion exchange or isomorphous substitution. For example, it has been found that highly crystallized silver hollandite can be obtained by the ion exchange of cryptomelane in a silver salt melt. Preferably, this silver salt melt is substantially pure, i.e. at least about 95% the liquefied silver salt. This ion exchange should occur at a temperature above the melting temperature of the silver salt but below the decomposition temperature of either the cryptomelane or the silver salt. A typical temperature range for the ion exchange will be between 200°C and 800°C. At atmospheric pressure, a typical duration will be at least about 1 hour. Suitable silver salts include the nitrates, sulfate, chlorates, bromides, chlorides, fluorides, and iodides of silver and organic acid silver salts. After ion exchange, the excess salt is removed by washing with a suitable solvent to produce substantially pure silver hollandite. Previously, silver hollandite was formed by the thermal decomposition of AgMnO₄ and Ag₂O in a 1 : 1 molar ratio at 970 C under 5 kbar oxygen over 7 days (see references (11) and (12) below (Chang & Jansen, "Agl.8Mn8O16: Square Planar Coordinated At+ Ions in the Channels of a Novel Hollandite Variant," Agnew. Chem. Int. Ed. Engl. 23 No. 11, pp. 906-907, 1984 and Chang and Jansen, "Der Erste Silberhollandit," Revwe de Chimie Minerale, Vol 23(1), pp. 48-54, 1986). Synthesis of silver hollandite via the present ion exchange process is much simple and facilitates the production of large amounts for industry applications. Additionally, in certain embodiments, the silver hollandite can be of high purity with small crystal size and high surface area. Silver hollandite can be useful as a low temperature SO₂ absorbent and as a low temperature CO, NO, and hydrocarbon oxidation catalyst. Other metals may optionally be introduced into the OMS structure using any of the methods known to the skilled person: excess impregnation, dry impregnation, ion exchange, etc. The material of the invention generally has a specific surface area in the range 1 to 300 m²/g, preferably in the range 2 to 300 m²/g, and more preferably in the range 30 to 250 m²/g. The adsorption kinetics are better when the specific surface area is high, i.e., in the range 10 m²/g such as in the range 30 to 250 m²/g. The sorbent phase can be in the form of a powder, beads, pellets or extrudates; they can also be deposited or directly prepared on monolithic supports of ceramic or metal. To increase the dispersion of the materials and thus to increase their absorption capacity, the materials can be deposited on porous supports with a high specific surface area before being formed (extrusion, coating . . . ). These supports are generally selected from the group formed by the following compounds: alumina (alpha, beta, delta, gamma, khi, or theta alumina), silicas (SiO₂), silica-aluminas, zeolites, titanium oxide (TiO₂), zirconium oxide (ZrO ), magnesium oxide (MgO), divided carbides, for example silicon carbides (SiC), used alone or as a mixture. Mixed oxides or solid solutions comprising at least two of the above oxides can be added. For many uses, such as in connection with a vehicle exhaust, it is usually preferable to use rigid supports (monoliths) with a large open porosity (more than 70%) to limit pressure drops that may cause high gas flow rates, and in particular high exhaust gas space velocities. These pressure drops are deleterious to proper functioning of the engine and contribute to reducing the efficiency of an internal combustion engine (gasoline or diesel). Further, the exhaust system is subjected to vibrations and to substantial mechanical and thermal shocks, so catalysts in the form of beads, pellets or extrudates run the risk of deterioration due to wear or fracturing. Two techniques can be used to prepare the catalysts of the invention on monolithic ceramic or metal supports (or substrates). The first technique comprises direct deposition on the monolithic support, using a wash coating technique which is known to the skilled person, to coat the adsorbing phase prepared using the operating procedure described, for example, in reference (4) below. (S. L. Suib, C-L O'Young, "Synthesis of Porous Materials", M. L. Occelli, H. Kessler, eds, M. Dekker, Inc., p. 215, 1997). The adsorbent phase can be coated just after the co-precipitation step, hydrothermal synthesis step or heating under reflux step, the final calcining step being carried out on the phase deposited on the monolith, or the monolith can be coated after the material has been prepared in its final state, i.e., after the final calcining step. The second technique comprises depositing the inorganic oxide on the monolithic support and then calcining the monolith between 500°C and 1100°C so that the specific surface area of this oxide is in the range 20 to 150 m²/g, then coating the monolithic substrate covered with the inorganic oxide with the adsorbent phase obtained after the steps described in the reference (4). Monolithic supports that can be used include: ceramics, where the principal elements can be alumina, zirconia, cordierite, mullite, silica, alumino-silicates or a combination of several of these compounds; a silicon carbide and/or nitride; an aluminium titanate; and/or a metal, generally obtained from iron, chromium or aluminium alloys optionally doped with nickel, cobalt, cerium or yttrium. The structure of a ceramic supports can be that of a honeycomb, or they are in the form of a foam or fibers. Metal supports can be produced by winding corrugated strips or by stacking corrugated sheets to constitute a honeycomb structure with straight or zigzag channels which may or may not communicate with each other. They can also be produced from metal fibers or wires which are interlocked, woven or braided. With supports of metal comprising aluminum in their composition, it is recommended that they are pre-treated at high temperature (for example between 700°C and 1100°C) to develop a micro-layer of refractory alumina on the surface. This superficial micro-layer, with a porosity and specific surface area which is higher than that of the original metal, encourages adhesion of the active phase and protects the remainder of the support against corrosion. The quantity of sorbent phase deposited or prepared directly on a ceramic or metallic support (or substrate) is generally in the range 20 to 300 g per liter of said support, advantageously in the range 50 to 200 g per liter. The materials of the invention can thus adsorb oxides of sulfur present in the gases, in particular exhaust gases. These materials are capable of adsorbing SO_x at a temperature which is generally in the range 50°C to 650°C, preferably in the range 100°C to 600°C, more preferably in the range 150°C to 550°C. For diesel engines in automobiles, an intended application, the temperature of the exhaust gas may be in the range 150°C to 500°C and rarely exceeds 600°C. The materials used in the process of the invention are thus suitable for sorbing oxides of sulfur present in the exhaust gases of stationary engines or, particularly, automotive diesel engines or spark ignition (lean burn) engines, but also in the gases from gas turbines operating with gas or liquid fuels. These exhaust gases typically contain oxides of sulfur in the range of a few tens to a few thousands of parts per million (ppm) and can contain comparable amounts of reducing compounds (CO, H₂, hydrocarbons) and nitrogen oxides. These exhaust gases might also contain larger quantities of oxygen (1% to close to 20% by volume) and steam, though the present sorbents can be effective in oxygen free environments as well. The sorbent material of the invention can be used with HS Vs (hourly space velocity, corresponding to the ratio of the volume of the monolith to the gas flow rate) for the exhaust gas generally in the range 500 to 150,000 h^"1, for example in the range 5,000 to 100,000 h^"1. In has been found that absorption of SO_x leads to a noticeable color change in the sorbent. Accordingly, in one variation, a SO_x trap is provided by a quantity of the sorbent contained in a housing having a window. The color of the sorbent material is periodically monitored through the window with the need to replace or recharge the trap indicated by the color change. In this or other refinements, a spent SO_x trap can be regenerated by appropriate reflux synthesis so as to reuse the sorbent support and the housing. The SO_x sorbent material of the present invention can be used anywhere SO_x needs to be absorbed. It has been found that significant advantages can be realized in the overall control of emissions from a combustion exhaust by locating the SO_x sorbent material upstream from a particulate filter or NOx trap. In one aspect of the invention, the SOx sorbent material is used in connection with "regenerable" NOx traps without the need to bypass or otherwise protect the SOx sorbent during regeneration of the NOx trap. As discussed more fully in reference (10) below (R. Burch, "Knowledge and Know-How in Emission Control for Mobile Applications," p. 308-326, 2004), regenerable NOx traps are constructed to capture NOx (for example as nitrate) during normal operation (lean conditions) and then to release the nitrogen (for example as N2) during a brief fuel-rich reduction step (rich conditions). These rich conditions typically last less than about 1 minute, and the lean conditions are typically at least about 5 times the duration of the rich conditions. However, when the NOx trap and the SOx sorbent are arranged in series, the SOx sorbent would also be subject to this lean/rich cycling, unless measures are employed to limit the fuel rich reduction conditions to the NOx trap. These measures, such as a bypass, can be difficult to implement. Advantageously, no such measures need to be employed with the present sorbent materials. For in certain embodiments, the present sorbent material is able to survive lean/rich cycling and still retain substantial sorbent capacity. According to another aspect of the invention, SO_x adsorption is provided by a manganese oxide material which has inherent oxidizing capability, so that SO₂ can be oxidized and absorbed without use of a separate and costly oxidation catalyst, and which has a high total absorption capacity, for example greater than about 40% by weight, thereby providing economical and efficient emissions control. Additionally, in other aspects of the invention, silver hollandite can be useful as a low temperature SO₂ absorbent and as a low temperature CO, NO, and hydrocarbon oxidation catalyst. According to another aspect of the invention, one or more sorbents of the invention, such as cryptomelane and silver Hollandite, can be combined in the same SOx trap, or alternatively, or additionally, can be placed in a series configuration in the emissions stream. In one aspect, for example, a SOx trap comprising silver hollandite can be placed upstream of a SOx trap comprising cryptomelane.

BRIEF DESCRIPTION OF THE FIGURES Although the characteristic features of this invention will be particularly pointed out in the claims, the invention itself, and the manner in which it may be made and used, may be better understood by referring to the following description taken in connection with the accompanying figures forming a part thereof. FIG. 1 is exemplary plots of the absorption of SO₂ on 2x2 Mn-OMS materials and on MnO . FIGS. 2a and 2b are exemplary scanning electron microscopy images of a 2x2 Mn-OMS material before and after SO absorption, respectively. FIGS. 3a and 3b are exemplary x-ray diffraction patterns of a 2x2 Mn-

OMS material before and after SO₂ adsorption, respectively. FIG. 4 is exemplary plots of absorption of SO₂ on a Mn-OMS material at different gas feed temperatures. FIG. 5 is exemplary plots of adsorption of SO on a Mn-OMS material at different gas feed rates. FIG. 6 is exemplary plots of adsorption of SO on a Mn-OMS material at different concentrations of SO₂ in the feed gas. FIG. 7 is exemplary plots of absorption of SO₂ on a Mn-OMS material at different feed gas compositions. FIG. 8 is exemplary plots of absorption of SO₂ on 2x3 Mn-OMS materials at different gas feed rates. FIG. 9 is exemplary plots of absorption of SO₂ on a 2x4 Mn-OMS material. FIG. 10 is a schematic illustration of an emissions control system implemented on a vehicle producing combustion exhaust according to an embodiment of the invention. FIG. 11 is a perspective view of a SO_x filter having a monitoring window and regeneration ports according to an embodiment of the invention. FIG. 12 is an exemplary ?XRD Pattern of Ag-Hollandite from Synthesis Method A. (your figure 2) FIG. 13 is an exemplary TEM Image of Ag-Hollandite from Synthesis

Method A. FIG. 14 is an exemplary Selected Area Electron Diffraction Pattern of Ag- Hollandite from Method A. FIG. 15 is an exemplary ?XRD Pattern of Ag-Hollandite from Method B. FIG. 16 is an exemplary TEM Image of Ag-Hollandite from Method B. FIG. 17 is an exemplary TEM Image of Ag-Hollandite from Method B. FIG. 18 is an EDS of Spot A in FIG. 17. FIG. 19 is an EDS of Spot B in FIG. 17. FIG. 20 is an exemplary plot of the CO oxidation properties of Ag- Hollandite from method B and cryptomelane over a range of temperatures, (your fig. 10) FIG. 21 is an exemplary plot if the C₃H^ oxidation properties of Ag- Hollandite from method B and cryptomelane over a range of temperatures, (your fig- 11) FIG. 22 is an exemplary plot of the NO oxidation properties of Ag- Hollandite from method B and cryptomelane over a range of temperatures, (your fig. 12) FIG. 23 is an exemplary plot of the de-NOx properties of Ag-Hollandite from method B and cryptomelane over a range of temperatures, (your fig. 13)

DESCRIPTION OF THE ILLUSTRATED EMBODIMENT For the purposes of promoting an understanding of the principles of the invention, reference will now be made to the embodiments illustrated in the drawings and specific language will be used to describe the same. It will nevertheless be understood that no limitation of the scope of the invention is hereby intended. Alterations and further modifications in the illustrated devices, and such further applications of the principles of the invention as illustrated herein are contemplated as would normally occur to one skilled in the art to which the invention relates. In one form, the present invention employs a manganese based octahedral molecular sieve as a high capacity sulfur oxide solid absorbent. As compared to other adsorbents studied for the removal of SO₂ from waste gases, this material provides surprising high capacity and efficiency. In a preferred form, this material is referred to as Mn-OMS 2x2. The basic structure of the materials employed in the Examples that follow consists of Mn0₆ octahedra joined at edges to form a 2 x 2 hollandite tunnel structure with a pore size of about 0.46 nm. (1) For cryptomelane, a counter- cation, K⁺, is present within the tunne 1 structure for charge compensation. Mn can assume an oxidation state of 4+, 3+, or 2+, and the average Mn oxidation state can be controlled within a certain range during synthesis. Generally, this material has a high surface area (~ 80 m²/g), and high redox reaction activity. (2) Without intending to be bound by any theory of operation, the present invention is based on carrying out the following reaction: S0₂ + KχMn₈Oι₆ → MnSO₄ + K₂O (1)

SO₂ is oxidized to S0₃ by Mn⁴⁺ and Mn³⁺, and ?Mn⁴⁺ and Mn³⁺ are simultaneously reduced to Mn²⁺ (MnO). The SO₃ produced then reacts with Mn²⁺ to form MnSO₄. As explained herein, tunnel structure cryptomelane was found to be a high capacity sulfur dioxide adsorbent. Its SO₂ capacity from 250°C to 475°C is more than ten times higher than that of conventional SO₂ adsorbents. Its maximum SO₂ capacity can be as high as about 74 wt%. The dominant mechanism for SO₂ absorption is believed to be that SO₂ is oxidized by Mn⁴⁺ and Mn³⁺ to SO₃, with the SO₃ reacting with the co-produced Mn²⁺ to form MnSO . It has been found that this reaction is primarily controlled by the mass diffusion of SO₂ through the adsorbent, and that it can surprisingly effectively occur in an oxygen-free environment. In addition, the visibly significant color change of cryptomelane from black to yellow after SO₂ absorption can be used as a convenient indicator for the adsorbent replacement. Cryptomelane for SO₂ absorption can be synthesized either from a mixture of KMnO₄ and MnSO₄ or a mixture of MnSO₄ and KOH solution. After SO₂ absorption, MnSO__t is formed, which can subsequently be dissolved in water and used as raw material for a subsequent cryptomelane synthesis. To regenerate the SO₂ absorption trap, therefore, only KOH and O₂ is needed because the adsorbent support (such as a monolith) and the MnSO₄ can be re-used. This highly efficient SO₂ adsorbent can be used for removal of SO₂ generated from thermal power plants, factories, and on-road vehicles. It can be especially effective for removal of SO_x that is present in the emissions of diesel trucks, in order to protect downstream emissions control devices such as particulate filters and NOx traps that are poisoned by SO_x. Turning now to FIG. 10, a vehicle 20 implementing a simplified emissions control system according to the present invention is depicted. Vehicle 20 has an engine 22 fluidly connected to upstream and a downstream emissions control devices 24 and 26 respectively. Devices 24 and 26 perform different emissions control functions, and while they could be combined into a single device, as described more fully below, certain problems are avoided by the provision of separate devices. The exhaust 21 from the engine 22 is first fed to the upstream device 24. The transfer of the exhaust, and all other fluid transfer operation can be in any conventional fashion, such as the exhaust piping of a conventional automobile, and may include intermediate fluid processing operations, such as catalytic conversion, mixing with other gases, or recycling of exhaust to the engine. The upstream device 24 is a SO_x scrubber, whose function is to remove any sulfur oxides from the exhaust gas 21 and to prevent their passage via channel 25 to the downstream device 26 and eventually the exhaust 28 to the atmosphere. The SO_x scrubber functions to remove most if not all of any sulfur oxides in the gaseous exhaust 21. The SO_x scrubber contains a solid SO_x sorbent as described herein, preferably one supported on a monolith or similar support. The sorbent removes the SO_x from the passing gas stream, for example by permanent or reversible sorption (adsorption or absorption) trapping, filtering, or chemical reaction therewith, and thereby prevents SO_x from entering the downstream device 26. The downstream device 26 provides a different emissions control function than the upstream device 24. As illustrated, the downstream device is a NOχ scrubber or particulate filter and any conventional scrubber or filter can be employed. As many conventional NOχ traps and/or particulate filters are fouled or poised by the presence of SO_x, the provision of upstream device 24 inventively reduces or eliminates this possibility by providing an inlet stream to device 26 that is substantially SO_x free. For example, it is contemplated that device 24 will function to cause fluid at 25 to have less than 1% of the SO_x concentration in the exhaust 21, more preferably less than about 0.1%. Turning now to FIG. 11, an exemplary SO_x scrubber 30, which can be employed for device 24 in the FIG. 10 system, is depicted. Scrubber 30 includes a housing 32 having a fluid inlet 34, a fluid outlet 36 and a fluid flow path therebetween. The housing contains a SO_x sorbent in the flow path so as to facilitate the removal of SO_x from the fluid as it passes through the SO_x scrubber. Housing 32 also contains a window 38 providing visual access to the sorbent contained therein. As the sorbent contained in the housing 32 absorbs the SO_x, it will undergo a noticeable color change, with the sorbent nearer the inlet 34 becoming saturated (and thus changing color) sooner than the sorbent near the outlet 36. The resulting transition between different colored portions of the sorbent provides an indication on the extent that the sorbent packing has becoming spent. Accordingly, a series of indicator marks 39 are provided on the window 38 or on the housing 32 adjacent the window 38 for measuring the remaining sorption capacity of the scrubber 30. For example, during routine maintenance of a machine on which scrubber 30 is implemented, vehicle 20 for example, the window 38 can be checked to determine whether replacement of the scrubber 30 is necessary. When replacement is needed, i.e. the sorbent is saturated and entirely changed colors, the scrubber 30 can simply be removed from the exhaust stream and replaced. In another form of the invention, once removed, the spent sorbent and/or the scrubber 30 can be reused. For example when the spent sorbent is converted to MnSO₄, this MnSO can be used as a starting material to reform the Mn-OMS material on the support. This reforming can be accomplished by removing the spent sorbent and its support (such as a monolith) from the housing 32. After processing and appropriate calcination the spent sorbent is returned to its OMS structure and is ready to absorb additional SO_x. Alternatively, the necessary reagents for reforming the spent sorbent, for example KOH and O₂, can be circulated through the housing 32 without removing the spent sorbent. The inlet and outlet ports 34 and 36 can be used as the reagent inlet and outlet ports to recharge the sorbent in this fashion when the scrubber 30 is removed from the exhaust stream. However, as illustrated, scrubber 30 includes optional dedicated inlet and outlet ports 44 and 42 for this purpose. Ports 42, 44 permit recharging without removal from the exhaust, or they may be used in conjunction with offline recharging via inlet and outlet 34, 36. Reference will now be made to examples illustrating specific features of inventive embodiments. It is to be understood, however, that these examples are provided for illustration and that no limitation to the scope of the invention is intended thereby. Further, certain observations, hypotheses, and theories of operation are presented in light of these examples in order to further understanding, but these are likewise not intended to limit the scope of the invention. EXAMPLES EXAMPLE 1 SAMPLE PREPARATIONS AND TEST CONDITIONS OMS 2X2 2x2 manganese based octahedral molecular sieve (tunnel structure cryptomelane) was prepared using the methods developed by DeGuzman, et al. (3) A typical synthesis was carried out as follows: 11.78 g KMnO₄ in 200 ml of water was added to a solution of 23.2 g MnSO₄- 4H O in 60 ml of water and 6 ml of concentrated HNO₃. The solution was refluxed at 100°C for 24 h, and the product was washed and dried at 120°C. Hydrothermal reaction in Teflon bottles at 90°C, instead of the reflux method, was also used for the synthesis. An alternative synthesis method for cryptomelane was purging O₂ through mixture of MnSO and KOH solution, followed by calcination at 600°C. (3) A typical preparation was: a solution of 15.7 g KOH in cold 100 ml of water was added to a solution of 14.9 g of MnSO₄Η₂O in 100 ml of water. Oxygen gas was bubbled (about 10 L/min) through the solution for 4 hours. The product was washed with water and calcined in air for 20 h. The dried materials were sieved to provide 40 - 80 mesh particles for the SO₂ absorption tests, which was carried out in a temperature controlled reactor with a Sulfur Chemiluminescent Detector (SCD) analysis system. Unless otherwise stated, the absorption testing conditions were 0.5 gram 40-80 mesh absorbent particles, 100 standard cubic centimeters per minute (seem) exhaust air flow with 250 parts per million (ppm) SO₂, 75% N₂, 12% O₂, and 13% CO₂. The SO₂ absorption performance of cryptomelane material synthesized by refluxing mixture of KMnO and MnSO₄ solutions was also systemically tested under different temperature, gas hour space velocity (GHSV), SO₂ concentrations, and feed gas compositions, and the results are summarized in Table 1. Before each SO₂ absorption measurement, the material was heated at 500°C for 2 h in flowing air. To characterize the property changes before and after SO₂ absorption, powder X-ray diffraction pattern (?X?RD), particle surface area (S A), and scanning electron microscopy (SEM) images were collected on some of the tested materials. Table 1 SO₂ absorption test conditions for cryptomelane material

* Steam was introduced by purging O₂ through flask containing temperature- controlled de-ionized water. After passing through the absorbent, steam was removed before SCD detector using ?MD Gas Dryer (from Perma Pure Inc.) which can selectively separate H₂O from gases mixture.

EXAMPLE 2 (COMPARATIVE) COMPARATIVE BREAKTHROUGH ABSORPTION CAPACITIES

For purposes of comparison, SO₂ absorption capability of several commercially available materials were tested at a temperature range from 250°C to 475°C under the testing conditions indicated above (0.5 gram 40-80 mesh absorbent particles, 100 seem exhaust air flow with 250 ppm SO₂, 75% N₂, 12% O₂, and 13% CO₂). These materials included La₂O₃ or BaO doped ZrO₂-CeO₂ mixtures (from Daiichi Kigenso Kagaku Kogyo Co., Ltd.), ZrO₂ (from RC100, Inc.), Al₂O₃ (from Engelhard, acidic), CaO (from Alfa Aesar Inc), and MnO₂ (from Erachem Comilog, Inc.) which were obtained from their respective commercial sources. Table 2 presents a summary of the SO₂ absorption capacities for certain of these SO₂ adsorbents. The SO₂ capacity was calculated based on weight of SO₂ adsorbed per gram of adsorbent when 1% of the initial SO₂ concentration was observed eluting from the absorbent bed. This is defined as the breakthrough absorption capacity. As seen in Table 2, the SO₂ breakthrough absorption capacities for these materials are generally less than 5 wt%. Of the other materials tested, the absorption capacities for the MnO₂ was selected for more direct comparison to the materials of Example 1 and are presented in the Examples below.

Table 2 SO₂ breakthrough absorption capacity of conventional SO₂ adsorbents

1. Other test conditions: 0.5 g 40-80 mesh absorbent, feed gas: 250 ppm SO₂, 75% N₂, 12% O₂, and 13% CO₂ 2. SO₂ capacity based on gram of SO₂ adsorbed per gram of catalyst at 1% SO₂ breakthrough point EXAMPLE 3 BREAKTHROUGH ABSORPTION CAPACITIES

Table 3 gives the SO₂ breakthrough and total absorption capacities of the materials synthesized according to Example 1, K_xMn₈Oι₆ A (reflux synthesis, with projected final average Mn oxidation state 3.5⁺), K_xMn₈O₁₆B (reflux synthesis, with projected final average Mn oxidation state 4⁺), Cu-doped K_xMn₈Oι₆ B (hydrothermal synthesis, with projected final average Mn oxidation state 4⁺), and K_xMn₈O₁₆ C (synthesized from MnSO₄ and KOH). The projected final Mn oxidation state (PAOS) is calculated based on the relative amount of KMnO and MnSO in the starting solution, i.e. PAOS = (moles of KMnO₄ * 7 + moles of MnSO₄ * 2)/( moles of KMnO + moles of MnSO₄ ). Breakthrough capacities were measured at 1% breakthrough as described in Example 2, and total SO₂ absorption capacity was also measured with the values given in parentheses in Table 3. For example, the breakthrough and maximum SO₂ absorption capacities for K_xMn₈O₁₆ B are 58 wt% and 68 wt% respectively. Under similar reaction conditions, these materials have significantly higher breakthrough SO₂ absorption capacity than the conventional SO₂ adsorbents given in Table 2. To facilitate comparison, the results for the commercially obtained electrolytic MnO₂ (EMD, Erachem Comilog, Inc.) discussed above are presented in Table 3.

Table 3 SO₂ absorption capacity of cryptomelane materials synthesized

*Other test conditions: 0.5 g 40-80 mesh absorbent, 325°C, feed gas: 250 ppm SO₂, 82% N₂, and 18% O₂ for K_xMn₈O₁₆ C, for others: 75% N₂, 12% O₂, and 13%) CO₂ ^aCuSO₄ was added in MnSO solution. ^bData in parentheses are maximum SO₂ absorption capacities. EXAMPLE 4 SO₂ ABSORPTION AT 325°C

FIG. 1 is a plots showing SO₂ absorption on K_xMn₈Oι₆ A, B, Cu-doped K_xMn₈Oι₆ B, and EMD MnO₂, as a function of the weight percentage of S0₂ fed at 325°C. The left axis is the percentage of SO₂ not absorbed (i.e. that passed through the bed) and corresponds to the S-shaped curves. The right axis is the wt% SO₂ absorbed and corresponds to the curves whose slope is initially 1 at low feed amounts and then tends towards slope of zero at high feed amounts. All weight percentages are relative weight of absorbent.

EXAMPLE 5 CHANGES AFTER ABSORPTION FIGS. 2 and 3 are before and after Scanning Electron Microscopy (SEM)

Images and x-ray diffraction patterns (?XRD), respectively, for the SO₂ absorption by K_xMn₈Oι₆ B at 325°C. As shown in FIG. 2, the morphology and the crystal structure of the K_xMn₈Oι₆ material significantly changes after SO₂ absorption. The surface area of this material also decreased sharply from 74 m²/g to 4.6 m²/g, and the XRD patterns indicate that the OMS structure had converted to a mixture of MnSO₄ and manganolangbeinite K₂Mn₂(SO₄)₃. A visible color change in the absorbent was also evident. It was initially black and changed to yellow after the SO₂ absorption. EXAMPLE 6 TEMPERATURE DEPENDENCE

FIG 5 is a plot of the wt% of SO₂ absorption on K_xMn₈O₁₆ B at 250°C, 325°C and 475°C under the other test conditions as indicated in Table 1 (0.5 g 40- 80 mesh absorbent, feed gas: 250 ppm SO₂, 82% N₂, 18% O_2, - 8,000 hr^"1 GHSV). Even at as low a temperature as 250°C, this material could adsorb more than 66 wt% SO₂, although absorption is not 100% and some SO₂ breakthrough was observed even initially.

EXAMPLE 7 FEED GAS FLOW RATE DEPENDENCE

FIG. 6 shows the feed gas GHSV effect on the SO₂ absorption on K_xMn₈O₁₆ B at 325°C. The breakthrough SO₂ capacity decreased from 61, to 44, and 33 wt% as the feed GHSV increased from 8K, to 30K and 60K hr^"1. As the feed GHSV increased, the total SO₂ absorption capacity decreased, but not significantly, from 74 to 64 and 63 wt%.

EXAMPLE 8 FEED GAS COMPOSITION DEPENDENCE

FIG. 7 shows that increasing SO₂ concentration in the feed gas from 50 ppm to 250 ppm almost had no effect on the SO₂ adsorption on K_xMn₈O₁₆ B. FIG. 8 shows feed gas composition effect on the SO₂ absorption on K_xMn₈Oι₆ B at 325°C. CO and NO, which also exist in the combustion waste gases, did not have any effect on the SO₂ absorption on K_xMn₈Oι₆. CO₂, at - 13% level, could slightly, if any, decrease the SO₂ breakthrough capacity (from 61 wt% to 59wt% ) and the total SO₂ capacity (from 74 to 68 wt%). Surprisingly, the OMS material K_xMn₈Oι₆ B shows a higher SO₂ breakthrough capacity of 74 wt% and total SO₂ capacity of -80 wt% in CO-NO- SO2-H₂O mixture feed gas even through the GHSV was 17K hr^"1. When steam was introduced into the system, the SCD signals were not very stable which may generate some error of the measurement. In the absent of O₂, cryptomelane material K_xMn₈Oι₆ B still has SO₂ breakthrough capacity of 41 wt% and total SO₂ capacity of 58 wt%. After SO₂ absorption test, the weight gain of the adsorbent was measured though it was not possible to collect all the adsorbent particles. Table 4 gives the weight gain data of the feed gas composition effect tests. For comparison, the total SO₂ absorption calculated from S0₂ concentration change for each test was also listed.

Table 4 Adsorbent weight gain after SO₂ absorption test in different feed gas at 325°C

Discussion of Examples 1-8 : Based on reaction (1), the maximum SO₂ capacity is believed to be controlled by oxidation state of Mn in the adsorbent. For the OMS materials K_xMn₈O₁₆ B and Cu-doped K_xMn₈Oι₆ B, the average projected oxidation state of ?M?n is 4; there are very few K⁺ cations present in the strvicture. If the small amount of K⁺ is ignored, the maximum S0₂ capacity should be ~ 73.5 wt%. The measured maximum SO₂ capacities (see Tables 3 and 4) are - 70 Vt% for these two materials, which is very close to 73.5 wt% and reflects the contribution of the K₂O byproduct. This result supports the hypothesis that reaction (1) dominates SO₂ absorption. Direct evidence is also seen in the XRD patterns before and after the SO₂ absorption. After SO₂ absorption, the OMS structure completely changed to MnSO and manganolangbeinite K₂Mn₂(SO )₃. The foirmation of K₂Mn₂(SO )₃ 9 94- means more SO₄ ^" than Mn is formed and that reaction (1) needs to be modified slightly. A postulate is that the cryptomelane material itself or the SO₂ adsorbed material (mostly MnSO₄) acts as a catalyst for the following reaction and formation of K₂Mn₂(SO₄)₃.

SO₂ + O₂ → SO₃ (2) Because the amount of K⁺ is small, the whole SO₂ absorption process is mostly dominated by manganese oxidation. It should be noted that, while the presence of O₂ in the feed gas does help increase absorption, it is not necessary, as demonstrated by the satisfactory absorption performance in the oxygen -free feed gas test (feed gas composition: 250 ppm SO₂, 12.5% N₂, and 87.5% He). After SO₂ absorption in an O₂-free environment, both MnSO₄ and K₂Mn₂(S0 )₃ were formed, suggesting that other reactions involving oxygen transfer and formation of some amorphous phases also happened in an O₂-free environment. The synthesized K_xMn₈Oι₆ A had an average projected Mn oxidation state of +3.5. The ideal formula for this material should be

(after water removal at 500°C and assuming all the counter-cations are K⁺). Then according to reaction (1), the maximum capacity is 45wt%, which exactly matches the measurement (see Table 3). If reaction (2) also exists with this material, K₂Mn₂(SO₄)₃ should form, which was not seen in an XRD pattern (not shown), and the maximum SO2 capacity should be 60 wt%. This suggests that reaction (1) predominates, and that reaction (2) occurs only to a very small extent or not at all. Based on the results from K_xMn₈Oι₆ A (hydrothermal synthesis, with projected final average Mn oxidation state 3.5), K_xMn₈Oι₆ B (reflux synthesis, with projected final average Mn oxidation state 4) and Cu-doped K_xMn₈Oι₆ B (hydrothermal synthesis, with projected final average Mn oxidation state 4), neither the choice between reflux or hydrothermal synthesis, nor doping with Cu significantly changes SO₂ break through absorption capacity or maximum absorption capacity. This suggests that there is no kinetic or thermodynamic effect attributable to doping or the synthesis mechanism. The alternative synthesis method of purging O₂ through a mixture of

MnSO₄ and KOH solution followed by calcination at 600°C did show some effect. K_xMn₈O₁₆ C, synthesized using this method, gives ~50wt% SO₂ breakthrough capacity and - 60wt% SO2 total capacity, which are lower than those for K_xMn₈O₁₆ B. Potential reasons for this disparity include 1) possible incomplete oxidation of Mn²⁺ to Mn⁴⁺, 2) the relative low surface area ( 32 m²/g vs. 75 m₂/g) and high density (~ lg/cm³ vs. 0.67 g/cm³), and 3) the product is not pure OMS as a small amount of K₂SO₄ was found to exists (see Fig. 9 for its XRD pattern). While the synthesis conditions could still be optimized to improve adsorption performance, the absorption performance is adequate to be effective and consideration of other factors renders this a desirable approach. For example, this synthesis method can significantly decrease the overall cost of the absorbent production. After SO₂ absorption, almost pure MnSO₄ is formed, which, being one of the starting materials, can be recaptured and reused. Also the density of K_xMn₈Oι₆ C is - 50% higher than that of K_xMn₈Oι₆ B, which indicates more adsorbents can be loaded and higher SO₂ capacity in a given volume can be achieved. Although electrolytic manganese dioxide (EMD) from Erachem Comilog, Inc. has Mn⁴⁺ and a surface area of - 30 m²/g, the SO₂ capacity for this material does not approach that of the OMS materials. This indicates that the OMS structure is important for high SO₂ absorption. K_xMn₈Oι₆ B was tested at a temperature 250°C, 325°C, and 475°C. At 250°C, a lower SO₂ absorption rate was observed but the maximum SO₂ capacity is almost the same as that measured at 325°C and 475°C. In comparing the results at 325°C and 475°C, minimal difference was observed except that at 475°C, after the breakthrough capacity was reached, the SCD detector background slightly increased, indicating that some SO₃ was being released even though total absorption amount was still increasing. Substantial variation in the feed gas flow rate affected the SO₂ absorption performance of K_xMn₈O₁₆ B adsorbent, though in practice this affect can be mitigated with appropriate sizing and design of the SO_x trap. For example, the SO₂ breakthrough capacity decreased about 50% when the feed GHSV increased from 8K to 60K hr^"1. This indicates the SO₂ absorption reaction is mostly controlled by SO2 mass diffusion through the adsorbent. Since the SO2 concentration in the feed gas had little or no effect on its absorption suggests that reaction (1) is 0^th order for SO₂, and it is mostly controlled by the available active sites on the adsorbent. Most components in the simulated exhaust combustion gases tested, CO, NO, CO₂, and H₂O, did not have a significant effect on the SO₂ absorption capacity of the K_xMn₈Oι₆ B adsorbent, and the absorption capacity in an oxygen free environment, while lower, was still acceptable. Therefore, it is expected that the Mn-OMS materials should be useful to remove SO₂ from gas streams in a wide variety of applications.

EXAMPLE 9 OTHER OMS STRUCTURES

SO₂ absorption capacities of other manganese oxides with tunnel structure, including Todoro?kite-type magnesium manganese oxide with channels of 3 x 3 MnO₆ units, sodium manganese oxide with channels of 2 x 4 MnO₆ units, sodium manganese oxide with channels of 2 x 3 MnO₆ units, and pyrolusite manganese oxide with l x l MnO₆ units, were studied. Pyrolusite, Mnθ₂ l x l, was obtained from Stream Chemicals. The as- received chemical was ball-milled for 1 hr to get - 1 μm particles before SO₂ absorption test. One todorokite material, OMS-1, was provided by Engelhard Corporation. Other tunnel-structured manganese oxides were prepared in the lab using the methods described in the published literature Birnessite was used as precursor for the synthesis of channel-structured manganese oxides. Birnessite-type layered manganese oxides were prepared using the methods used by Golden, et al. (5) and (6). A typical synthesis was carried out as: 250 ml 6.4 M NaOH solution was mixed with 200 ml 0.5M MnSO₄ at room temperature. Oxygen was immediately bubbled through a glass frit at a rate of 4 L/min. After 4.5 h the oxygenation was stopped and the precipitate was filtered out and washed with deionized water 4 times, and then dried in air at 100°C. About 13 g grey color birnessite product was obtained. Two sodium manganese oxides with channels of 2 x 3 MnO₆ units (Na 2 x

3 A & B) were prepared by directly calcination of birnessite in air for 12 h at 500°C, and 650°C, respectively. (7). Todorokite (magnesium manganese oxide with channels of 3 x 3 MnO₆ units) was prepared as described in (5) and (8): -3 g birnessite was added in 100 ml IM MgCl₂ solution, the mixture was shaken overnight at room temperature for Mg²⁺ ion exchange for Na⁺. The slurry was washed four times with deionized water. Then, Mg²⁺-birnessite, together with 25 ml H₂O, was autoclaved at 150°C for 48 hr. After washing with D.I water 3 times, the product was dried in air at 100°C. About 2.0 g todorokite-type tunnel structure manganese oxide (Mg 3 x 3) was obtained. Sodium manganese oxide with channels of 2 x 4 MnO₆ units was prepared using method developed by Xia et al. (9). About 5 g birnessite, together with 25 ml 2.5M NaCl solution, was autoclaved at 210°C for 48 hr. After washing with D.I water 3 times, the product was dried in air at 100°C. About 4.4 g black color product, Na 2 x 4, was obtained. The dried materials were sieved to provide 40 - 80 mesh particles for the SO₂ absorption test, which was carried out in a temperature controlled reactor with an the SCD analytical system. All these materials were tested under the same conditions (0.5 gram 40-80 mesh absorbent particles, 100 seem feed flow of 250 ppm SO₂in 82% N₂, 18% O₂) at a temperature of 325°C (See Table 5). Before each SO₂ absorption measurement, the absorbent material was heated at 500°C for 2 h in flowing air to remove residual moisture. To characterize the structure change before and after SO₂ absorption, powder X-ray diffraction pattern (?XRD) was collected on some of the tested materials.

Table 5 SO₂ absorption tests of other OMS materials

* Before each SO₂ absorption measurement, the absorbent material was heated at 500°C for 2 h in 100 seem air.

EXAMPLE 10 OTHER OMS STRUCTURES RESULTS AND CO?MPARISON The measured breakthrough capacities at selected gas flow rates for the materials prepared in Example 9 are given in Table 6 along with exemplary capacities for the 2x2 structure of Example 1.

Table 6

Save two of the materials, the SO₂ breakthrough absorption capacities for these materials are generally much higher than those of conventional SO_x absorbents (normally less than 5 wt%), establishing their usefulness as SO_x absorbents pursuant to the present invention. The lxl MnO₂ from Strem Chemicals was confirmed by ?XRD to be well crystallized pyrolusite, which consists of 1 x 1 MnO tunnels. After calcination at 500°C for 2 h in air, the structure remained stable. However the material exhibits poor absorption capacity, with a breakthrough capacity of about 0.1% and a maximum absorption capacity less than 3%. The birnessite synthesized in this work was calcined in air either at 500°C for 12 h (A), or at 650°C for 12h (B). In either case, sodium manganese oxide, Na₂Mn₅O₁₀, forms, the basic structure which consists of Mn0₆ octahedra joined at edges to form a 2 x 3 tunnel structure. (7) FIG. 8 shows the S0₂ absorption performance of both the A and B formulations of this microporous manganese oxide under different GHSV. Similar to the 2x2 cryptomelane materials, this microporous manganese oxide also has very high SO2 absorption capacity. At 3444 hr^"1 GHSV, the total SO₂ absorbed is - 70 wt%. At higher GHSV, the SO₂ absorption performance decreases, indicating the reaction is controlled by the mass diffusion of SO₂ through the absorbent. After SO₂ absorption, MnSO₄ forms. The XRD pattern of the sodium manganese oxide synthesized by hydrothermal reaction in this work matched very well with the ?XRD pattern of the sodium manganese oxide material with 2 x 4 tunnel structure published by Xia et al. (10) Likewise, as expected, after calcination at 500°C for 12 h in air, the 2 x 4 tunnel structure remains unchanged. FIG 9 gives the SO₂ absorption test result at 10755 hr^"1 GHSV. This material also has high SO₂ absorption capacity. While the todorokite magnesium manganese oxide prepared by hydrothermal reaction in this work was not well crystallized, the todorokite structure could still be identified in the XRD. However, after calcination at 500°C for 2 h in air, the 3 x 3 tunnel structure changed mostly to MgMn₂O₄. This structure change was evident in the MgMn₂θ provided by Engelhard Corporation as well. While the todorokite materials both exhibited relative high absorption capacities, this instability at moderately high temperatures is a relative disadvantage for most applications. While they could be used as absorbents, most conventional implementations require stability above the absorbents at 500°C or sometimes higher. Introduction to Examples 10 through 13:

The stability of cryptomelane was studied under oxidizing conditions, reducing conditions, and certain lean-rich cycling conditions. Cryptomelane was confirmed to be stable during oxidizing conditions and under lean-rich cycling, but that the stability under certain reducing conditions was found to be temperature and condition dependent. The studies described in Examples 10 through 13 were mostly performed using a Netzsch STA 409 TGA/DSC/MS, where TGA is Thermogravimetric Analysis, DSC is Differential Scanning Calorimetry, and MS is Mass Spectroscopy. Different gases, including air, 2% H₂ in Ar, 2% C₃H₆ (propylene) in Ar, were used for TGA-DSC analysis. To get a large amount of samples, cryptomelane materials were also treated in a tube furnace with flowing air, 2% C₃H₆ in Ar, simulated rich condition exhaust from diesel engine, simulated lean condition exhaust, and simulated lean-rich cycling exhaust from a diesel engine. The composition of these simulated exhausts are displayed in Table 7. Unless otherwise specified, the absorption tests were performed at 325 °C, using feed gas of 250 ppm SO2 in air at 8000 hr-1 gas hour space velocity (GHSV), and before each absorption test, the absorbent was pretreated in air at 500 °C for 2 hr. Additional details of these studies can be found in U.S. Provisional Application Ser. No. , filed February 3, 2005, the disclosure of which is hereby incorporated by reference to the extent not inconsistent with the present disclosure.

Table 7 Composition of simulated diesel engine exhausts used in Examples 10 through 13

EXAMPLE 10 STABILITY OF CRYPTOMELANE UNDER OXIDIZING CONDITIONS

The stability of cryptomelane after two different oxidation pretreatments was tested. The first pretreatment was air containing 10% H₂O for three (3) hours at 600 °C and 30,000 hr^"1 GHSV. The second pretreatment was the simulated lean exhaust gas (composition listed in Table 7) for one (1) hour at 500 °C. Subsequent absorption tests on 250 ppm SO₂ in air at 325 °C and 80,000 hr^"1 GHSV confirmed that the cryptomelane retained its high absorption capacity, and XRD analysis of the cryptomelane after exposure to each oxidizing stream confirmed that the cryptomelane maintained its structural configuration. EXAMPLE 11 STABILITY OF CRYPTOMELA?NE UNDER REDUCING CONDITIONS As a highly active, high valance manganese oxide, cryptomelane was expected to be somewhat unstable under reducing conditions. DSC analysis of cryptomelane exposed to a reducing stream of 2% C₃H₆ in Ar flowing at 40 ml/min while subjected to a heating rate of lOK/min showed that the cryptomelane was reduced at a temperature as low as 300 °C. As the cryptomelane was reduced, it transitioned, at least in part, into Mn₃O and MnO. When the cryptomelane was reduced at temperatures higher than 550 °C, the Mn₃O₄ was further reduced to MnO. Similar results were obtained when the reducing gas comprised 2% H₂ in Ar. The doping of Cu and Cr into cryptomelane was not seen to increase its stability under these reducing conditions. The MnO from reduced cryptomelane was found to be easily re-oxidized when heated in air, but the properties of the re-oxidized species showed some variation based on the particular reduction-oxidation cycle. For the MnO formed from reduction in 2% C₃H₆ in He at 550 °C, DSC analysis in 40 ml/min flowing air with heating rate of lOK/min showed that the MnO was re-oxidized at a temperature as low as 250 °C. After oxidation for one (1) hour at 500 °C, ?XRD analysis revealed that most of the cryptomelane crystal structure was recovered. However, the BET surface area of the cryptomelane decreased from 74 m²/g to 4.9 m²/g, and SEM images demonstrated that the needle-like crystal did not exist anymore. Subsequent SO₂ absorption tests at 325 °C of 250 ppm SO₂ in air at 80,000 hr^"1 GHSV revealed a loss of absorption capacity due to the reduction- oxidation cycle. Post absorption test XRD patterns showed that a large portion of this cryptomelane that had subjected to this reduction-oxidation cycle remained unused after the SO absorption. As discussed above, at least a portion of the reduced cryptomelane is able to recover its original crystalline structure after it is subjected to re-oxidation.

Additional tests were performed to determine phase compositions that were formed after re-oxidation occurred. These tests were performed using the following reducing gases: 2% H₂ in He and simulated rich exhaust gas (see Table 7). Table 8 summarizes the phases found after re-oxidation was performed. For all these tests, the most abundant phase formed after re-oxidation was either Mn₃O₄ or Mn₂O₃, not cryptomelane.

Table 8 Phases Formed After Reduction-Oxidation Tests of Cryptomelane

* Major phase listed first, minor phase listed in parentheses.

# Rich exhaust composition given in Table 7.

The properties of these reduction-oxidation treated cryptomelane was seen to change significantly. The rich-exhaust treated (550 °C for one hour) was found to have a significantly reduce surface area (5.1 m2/g) and SO₂ absorption capacity.

EXAMPLE 12 STABILITY OF CRYPTOMELANE UNDER LEAN-RICH CYCLING

An analysis of the results discussed in Examples 10 and 11 show that whether or not cryptomelane can recover its original crystal structure and its original SO₂ absorption performance after a reduction-oxidation cycle largely depends on the composition of the reductant and the cryptomelane treatment history. In an effort to study the effect of lean-rich cycling (NOx trap approach) on cryptomelane, cryptomelane was subjected to lean-rich cycles at 475°C for 6.5 hours (cycling at 360 seconds lean and 30 seconds rich). The exhaust gas flow was 26,000 hr^"1 GHSV and its composition was the same as is listed in Table 7. XRD patterns of cryptomelane taken both before and after the treatment cycle were almost identical. Some morphology changes were observed under SEM, and cryptomelane' s BET surface area decreased from 74 m²/g to 20 m²/g after treatment. However, the SO2 absorption performance of the lean-rich treated cryptomelane was very close to that of fresh cryptomelane. Additionally, as revealed from the ?XRD pattern of lean-rich cycled cryptomelane loaded with SO₂, a substantial amount of the cryptomelane was used for SO₂ absorption. The DSC analysis above indicated that cryptomelane can be re-oxidized at a lower temperature (250) than the temperature at which it can be reduced (300). Since the lean-rich cycling was performed at a constant temperature of 475, any cryptomelate that was reduced during the rich conditions was re-oxidized during the lean condition with the end result being that the cryptomelane structure remains unchanged. Moreover, the fact that the re-oxidation temperature is below the reduction temperature of cryptomelane provides ample flexibility in selecting the operating temperature (or temperature range) for lean-rich cycling such that any cryptomelate that is reduced during the rich conditions can be re-oxidized during the lean condition.

EXAMPLE 13 STABILITY OF SO₂-LOADED CRYPTO?MELANE

After SO₂ absorption occurs on cryptomelane, MnSO₄ and K₂Mn₂(SO₄)₃ form. As determined by TGA, these compounds are very stable in an oxidizing atmosphere. The stability of the SO₂-absorbed cryptomelane was studied using TGA-MS analysis on SO₂-loaded cryptomelane. From a sulfur stabilization point of view, the used cryptomelane was stable at temperature up to approximately 600 C in 2% H2 in He. At temperatures above 600 °C SO₂ and H₂S were released from the cryptomelane. At approximately 200 °C, some H₂O was released from the sorbent. Similar results were also observed when Cu-doped and Cr-doped SO₂ - absorbed cryptomelane was studied. Doping Cu or Cr into cryptomelane did not affect the stability of the SO₂ -loaded absorbent. Stability of the used absorbent under a rich stimulant was studied by continuously passing 2% CO and 2000 ppm C3H6 (balanced in Ar) through the used absorbent at 70,000 hr^'1 GHSV. Table 9 shows the sulfur concentration in the off gas at different temperatures. At temperatures below about 450 °C, the used absorbent was stable in rich exhaust gas, as only a small amount of sulfur was released from the absorbent. At approximately 500 °C, the used absorbent was relatively unstable and a larger amount of sulfur (100 ppm) was released from the sorbent. After 20 hours of reduction, it was found, using XRD analysis, that MnSO had been reduced to MnO and MnS. K₂Mn₂(SO₄)₃ appeared to remain stable under this extreme condition. When exposed again to air at 500 °C, the MnO and MnS were oxidized back to Mn₃O and MnSO₄. At same time, some SO₂ was released from the sorbent.

Table 9 Sulfur Concentration in Off Gas From Used KMn₈Oι₆ (MnSO ) Reduced in 2% CO / 2000 ppm C₃H₆ at 70K ¹ GHSV

Discussion of Examples 10 through 13.

Cryptomelane, a high capacity SOx sorbent, was shown to be stable under oxidizing conditions and, in general, to be easily re-oxidized after reduction. However, when subjected to reducing conditions for extended periods, the SO2 capacity upon re-oxidation can be significantly less than that for fresh cryptomelane. However, for the type of lean-rich cycling proposed for NOx trap regeneration, cryptomelane showed no noticeable change in its crystal structure or in its SO₂ absorption capacity. Further, after SO2 absorption, the used cryptomelane was shown to be stable and avoided releasing any significant amount of sulfur up to moderate temperatures. EXAMPLE 14 SILVER HOLLANDITE SYNTHESIS AND CHARACTERIZATION

Ag-hollandite was synthesized using post-ion exchange treatment of cryptomelane in AgNO₃ solution, followed by calcination in air. The cryptomelane was prepared using the method by DeGuzman, et al. (see reference (3) below). A typical post-ion exchange treatment of cryptomelane was as follows (denoted as method A): 1.0 g of cryptomelane was added to 100 mL IM AgNO₃ solution, the mixture was put into a temperature-controlled shaker and heated to 55 °C under continuous shaking. The total ion exchange duration was 24 hrs. The liquid was decanted, and the solid was dried in air at 120 °C overnight. The dried powder was calcined in air at 500 °C for 2 hrs. The yield was 1.3 g. A second method (denoted as method B) to synthesize silver hollandite from cryptomelane involved performing the ion exchange in a melt of the silver salt, rather than in solution. More specifically 2.0 g of cryptomelane were added to 20 mL of IM AgNO₃ solution, the mixture was put into a temperature-controlled furnace and heated to 150 °C for 3 hrs, and then to 250 °C for 12 hrs. After the ion exchange treatment, the extra AgNO in the mixture was washed off using deionized (D.I.) water. The yield of dried powder was 2.54 g. Table 10 displays some sample compositions that were synthesized in this study.

Table 10 Cation Molar Composition of Cryptomelane and Ag-Hollandite from ICP Analysis

Powder X-ray diffraction (?XRD) patterns were collected with a Sintag IV diffractimeter using Cu Kα radiation. Transmission electron microscope (TEM) with electron dispersion spectrum (EDS) pattern, BET surface area, ICP elemental analysis, TPR analysis, TGA-DSC, and XPS spectra were also collected on some of the tested samples. For ICP elemental analysis, about 20 mg of solid powder was dissolved in a solution of 12 ml of 2%KNO₃-30%H2θ₂ in water. The mixture was then diluted 1000 times with 2% HNO₃ before ICP chemical analysis. ?XPS measurements were performed using a Physical Electronics Quantum 2000 Scanning ESCA ?Microprobe.

SO₂ Uptake and CO, Hydrocarbon (as C₃H₆), and NO Oxidation Measurements

The test setup included a small, fixed bed quartz tube reactor, which was heated by a small clam-shell furnace. Reactant gases were metered using mass flow controllers. In different runs, different feed gases were utilized. The SO₂ analytical system comprised a HP6890 gas chromatograph equipped with a Sulfur Chemiluminescent Detector (SCD). The analytical system used was the same as that described above. The concentration of CO, C₃H₆, and CO₂ were measured using an Agilent Quad Series Micro GC. The NO, NO₂ and total NOx were measured using a 600-HCLD Digital NOx Meter (California Analytical

Instruments, Inc.). During the experimental runs, the analytical system was operated continuously, sampling the effluent every three minutes. The maximum sensitivity of the system to SO₂ (with SO₂ feed levels at 10 ppm) was approximately 50 ppb, and to CO, CO₂, C₃H₆, NO, NO₂ was approximately 5 ppm. Typical measurements employed a 0.2 g sample, pressed and sieved to 40-80 mesh. Each sample was pretreated in flowing air (100 seem) at 500 °C for two hours prior to measuring SO₂ uptake and oxidation performance.

Results and Discussion The structure of silver hollandite is different than that of traditional hollandite in that the silver cations do not occupy the centers of the cubic cages formed by MnO₆ octahedra, but rather the common faces of the cubes, coordinated with four oxygen anions at about 0.24 nm. This special structure, as well as the special property of Ag cations and Ag-Mn mixtures, may contribute to some unique catalytic and ion conductive properties. The Ag+ cation can be doped into the tunnel structure of cryptomelane by post ion-exchange. Actually, cryptomelane can be effective for selective adsorption of Ag+ at a low pH range, even in the presence of large amount of other cations. After the Ag+ ion exchange, the original cryptomelane crystal structure disappears, and a crystallized Ag-hollandite forms. Highly crystallized Ag-hollandite was obtained using synthesis method A (Ag-Hollandite A) as described above, i.e., by ion exchanging at 55°C and calcining without intermediate wash of the product with D.I. water. Figures 13-15 shows the resulting ?XRD pattern, TEM image selected area electron dispersion spectrum, respectively. The cation composition is given in Table 10. The XRD pattern (Figure 12) matched well with that of Ag-hollandite, Ag!.₈Mn₈Oι₆. Certain amounts of metallic silver also existed in the final product. Because there was no product washing before final calcination, is believed that this metallic silver came from the decomposition of the extra AgNO₃. These silver particles were so big that they were visible to the naked eye. The crystal size of

Ag-hollandite A was small. The crystals were needle-like, being approximately 30 nm width and 100-200 nm in length. This morphology is quite similar to that of the cryptomelane, except that the Ag-hollandite crystals were relatively shorter. The EDS pattern and ICP analysis both show evidence of a small amount of K+ in the crystal structure, which indicated that the ion exchange process was not substantially complete. This also suggests that the Ag-hollandite crystal structure was able to accommodate a small amount of K+. Chemical analysis of a water-washed sample of Ag-hollandite A with large silver particles manually picked out indicated that the K+ ions were potentially located inside the hollandite structure. The BET surface area of the as-synthesized material (Ag-Hollandite A) was 35 m /g, which can be mostly attributed to the small particle size. In an attempt to remove the extra metallic silver, the Ag-hollandite A final product was treated with a 2M HNO₃ solution. It was found, using ?XRD pattern analysis, that the silver phase could be removed, but that extra hollandite structure also resulted. Additionally, ICP results showed that the amount of silver in the sample was less than that in Ag-hollandite A. This indicated that the Ag+ cations inside the tunnel structure of Ag-hollandite could be easily removed and replaced by H+ cations, and that the use of an acid wash to purify the Ag-Hollandite A was not advantageous. In another attempt to remove the extra silver phase, the ion-exchanged cryptomelane was washed with D.I water before the final calcinations occurred. Although the silver phase was removed in the final product, excess hollandite structure showed up again. The ??XRD pattern of this sample was similar to that of the intermediate product for Ag-hollandite A, i.e. cryptomelane ion exchanged in IM AgNO₃ at 55 °C, washed off extra AgNO3 with D.I. water, dried at 120 C in air, no calcination at 500 °C. Because no other cations were available for ion exchange with Ag+ during washing with D.I. water, those Ag+ cations already exchanged with the K+ cations likely remained in their original structure. This also suggests that the final calcination step did not change the structure when no additional AgNO₃ was available. In other words, the difference in the final product structure and cation composition between the intermediated washed sample and the non-washed sample likely came from changes during the calcination step. Thus, to get highly crystallized Ag-hollandite in the Ag-Hollandite A synthesis method, the calcination would appear to need to be carried out without washing off the extra AgNO₃. In the Ag-Hollandite A synthesis method, the final calcination step involved heating in air from 25 to 500 °C at a rate of 10 °C /min, and then maintaining at 500 °C for 2 hr. Under this condition, the extra AgNO₃ melted first (at approximately 212 C) and then decomposed (at approximately 444 C). It was postulated that most of the ion exchange reaction between Ag+ and K+ occurred between 212 C and 444 C in the AgNO₃ melt, rather than during the calcination at 500 C. The Ag-hollandite synthesis method B was developed to test the theory that significant ion exchange occurs in the silver salt melt. Figure 15 shows the XRD pattern of the Ag-hollandite formed via method B, and Figures 16 and 17 show selected TEM images. In the TEM images (Figures 16 and 17), small silver particles (indicated as spot B) are evident coated on the Ag-hollandite surface (spot A). The EDS spectrums of Spots A and B are given in Figures 18 and 19 respectively. The EDS signal for Ag from spot A (Ag-hollandite crystal without Ag metal) is smaller than that for Mn, whereas in the EDS signal for spot B, the Ag signal is much larger than that for Mn. This indicates even in method B, there are still some Ag particles on Ag-hollandite surface. Note that the signal ratio of Ag/Mn in spot A (Ag-hollandite crystal from method B without Ag metal) is larger than that in Fig 14 (Ag-hollandite from method A). This indicates the Ag concentration in the method B synthesized Ag-hollandite is much higher than that in the previous sample (method A). The composition of Ag-Hollandite B is given in Table 10. Compared to the Ag-hollandite from method A, the Ag-hollandite B has several special properties. For example, in the XRD patterns for Ag-Hollandite A (Figure 12) a silver peak is present, but no such peak is observed in the ?XRD pattern for Ag-hollandite B (Figure 15). Additionally, the crystal size of Ag- hollandite B is slightly larger than that from method A, and was almost the same as that of the starting material, cryptomelane. Further, some small silver particles (<5 nm diameter) were found coated on the Ag-hollandite B surface, which was not seen on the previous Ag-hollandite sample. Finally, in Ag-hollandite B, the K+ concentration decreased, and the Ag+ concentration increased. Generally, the Ag-Hollandite B synthesis method yielded more pure Ag- hollandite products, and the ion-exchange between silver and potassium cations was substantially completed. The small silver particles may have resulted from the decomposition of a small amount of AgNO₃ at 250 °C during the ion exchange synthesis process. The small particle size of Ag-hollandite A may be the result of AgNO₃ decomposition, i.e. the NOx gas cuts the long crystal into small pieces. Like cryptomelane, Ag-hollandite is quite stable under oxidation conditions. When heated in air, for example its structure remains unchanged up to about 700 °C. However, under reducing conditions, its reactivity is somewhat enhanced. The reactivity of cryptomelane and Ag-hollandite A and B were studied in 10% H₂ in Ar using TPR analysis. As found, H₂ reacted with Ag-hollandite A and B at temperatures as low as 100 °C, whereas cryptomelane required temperatures higher than 250 °C to react with H₂. This unique Ag-Hollandite property has use in the low temperature oxidation of some reductants. Like the cryptomelane discussed above, silver hollandite was also found to be a good high capacity SOx absorbent, particularly so at lower temperatures. Table 11 shows the low temperature SO2 absorption performance of Ag- Hollandite. When compared to cryptomelane, Ag-hollandite shows much higher SO₂ capacity at low temperatures. Ag-hollandite from method B has SO₂ capacity of about 8.75 wt% at 150 °C, and 26.8 wt% at about 200 °C. Ag-hollandite from method A also shows a higher low-temperature SO₂ capacity (28 wt% at about 250 °C) than cryptomelane. The low temperature SO₂ capacity strongly depends on how much Ag-hollandite is present in the sorbent. As pointed out above, in synthesis method A, if extra AgNO₃ is washed off before final calcination, (i.e. no ion exchange in the silver salt melt) only a small amount of Ag-hollandite is formed. As a result, this absorbent shows a low SO₂ capacity at low temperatures. For comparison, several other absorbents, including Ca(OH)₂ (from Aldrich), Ca(OH)₂ (from thermal decomposition of Ca(OH)₂ at 600 °C 12h), and ZrO₂- Ce0₂-La₂O₃ (62, 29, 9 wt%) mixture (fromDaiichi Kigenso Kagaku Kogyo Co., Ltd.) were also tested and their SO₂ capacities are given in Table 11. Previous work has shown that these materials show relatively high SO₂ capacity at 325°C.

Table 11. SO₂ Breakthrough Capacity of Ag-Hollandites and Other Absorbents

Feed: 10 ppm SO2 in air unless specified, GHSV: 60 K hr-1

Breakthrough capacity (1O0 x g. SO₂ / g/ absorbent): defined as the point where

SO₂ out exceeded 100 ppb. ^x Feed gas: 10 ppm SO₂, 230 ppm NO, 500 ppm CO, 378 ppm C₃H₆, 15% O2, balanced in N₂, GHSV: 6O K hr-1.

The CO oxidation and C₃H₆ oxidation properties of Ag-hollandite and cryptomelane were compared over various temperatures, and the results plotted in Figures 20-21. At higher temperatures (i.e. above about 200), the conversions were similar, but for Ag-hollandite showed significantly higher conversion at lower temperatures. For example, at 100 °C, Ag-hollandite B can oxidize more than 80% of CO to CO₂, versus less than 2 % for cryptomelane. At 150 °C, Ag- hollandite B can oxidize 70% of the C₃H₆ present to CO₂, versus about 50% for cryptomelane. Ag-hollandite was also seen to have a greater low temperature sulfur tolerance relative to cryptomelane, which is reasonable because Ag- hollandite is a better low-temperature sulfur absorbent. With 8 ppm SO₂ in the feed, CO conversion over cryptomelane at 200 °C decreases from 100% to 60% over 15 hrs, whereas, the corresponding conversion over Ag-hollandite B at 150 °C remains unchanged (CO conversion stays at 100%). Ag-hollandite also can be useful in the oxidation of NO. Figure 22 shows the oxidation performance of Ag-hollandite and cryptomelane with respect to NO, i.e. oxidizing NO to N0₂ at different temperatures. At 200 °C, Ag-hollandite can oxidize 38% of the NO present to NO₂; however, at higher temperature (above 300 °C), its NO oxidation ability seems to be less than that of cryptomelane. At temperatures above about 300 °C, the NO oxidation of Ag-hollandite B is generally about 20% lower than that of cryptomelane. While this result seems inconsistent with the observation that Ag-hollandite is more reactive with reductants than cryptomelane, the inconsistency can be explained, for example, by a difference in de-NOx performance (i.e., conversion of NOx to N₂) of Ag- hollandite and cryptomelane. The de-NOx performance were measured for Ag- hollandite and cryptomelane, and the results given in Figure 23. As can be seen, Ag-hollandite show substantially greater ability to decrease the total NOx in the feed in the 200 °C to 350 °C range. Therefore, the difference in NO oxidation ability between cryptomelane and Ag-hollandite (noted with respect to Figure 22) likely comes from this difference in their de-NOx abilities, for example, some NO₂ is decomposed to N₂ and O by Ag-hollandite. Ag-hollandite can also be useful as a catalyst for CO, NO, and C₃H₆ oxidation reactions, since no structure change was observed after these reactions based on ?X?RD analysis. Also, as shown in Table 11, Ag-hollandite A has 36 wt% SO₂ capacity at 325 C using stimulant exhaust gas with CO, C₃H₆, and NO as feed gas. If CO, NO, or C₃H₆ has the ability to permanently reduce Ag-hollandite, the SO₂ capacity would have shown a larger decrease. Therefore, after reaction with these reductants, the reduced Ag-hollandite must be readily oxidized back to its original structure by the oxygen in the feed gas.

Summary of EXAMPLE 14

Ag-hollandite was successfully synthesized by the ion exchange of cryptomelane in an AgNO₃ melt. Compared to the high temperature solid state syntheses process described in references (11) and (12) below, this process is very simple and the resulting Ag-Hollandite has a smaller crystal size and a higher surface area. Ag-hollandite synthesized according to certain embodiments disclosed above, was found to be even more useful as a low temperature SO₂ absorbent than cryptomelane. Ag-hollandite can also be used as active catalyst for CO, NO, and hydrocarbon oxidation. Silver-hollandite is able to maintain this catalytic oxidation behavior even when the material is simultaneously aging by adsorption of SO2, whereas many other oxidation catalysts rapidly lose oxidation activity toward these gases as SO₂ adsorption progresses. Moreover, with CO-NO mixtures, silver-hollandite shows evidence of lean NOx conversion, i.e. partial conversion of NOx to N₂ gas, which can be advantageous in emissions control. Additional details of these studies can be found in the above referenced U.S. Provisional Application filed February 3, 20O5.

CLOSURE While the invention has been illustrated and described in detail in the drawings and foregoing description, the same is to be considered as illustrative and not restrictive in character. Only certain embodiments have been shown and described, and all changes, equivalents, and modifications that come within the spirit of the invention described herein are desired to be protected. Any experiments, experimental examples, or experimental results provided herein are intended to be illustrative of the present invention and should not be considered limiting or restrictive with regard to the invention scope. Further, any theory, mechanism of operation, proof, or finding stated herein is meant to further enhance understanding of the present invention and is not intended to limit the present invention in any way to such theory, mechanism of operation, proof, or finding. Thus, the specifics of this description and the attached drawings should not be interpreted to limit the scope of this invention to the specifics thereof. Rather, the scope of this invention should be evaluated with reference to the claims appended hereto. In reading the claims it is intended that when words such as "a", "an", "at least one", and "at least a portion" are used there is no intention to limit the claims to only one item unless specifically stated to the contrary in the claims. Further, when the language "at least a portion" and/or "a portion" is used, the claims may include a portion and/or the entire items unless specifically stated to the contrary. Likewise, where the term "input" or "output" is used in connection with a fluid processing unit, it should be understood to comprehend singular or plural and one or more fluid channels as appropriate in the context. Finally, all publications, patents, and patent applications cited in this specification are herein incorporated by reference to the extent not inconsistent with the present disclosure as if each were specifically and individually indicated to be incorporated by reference and set forth in its entirety herein, including the following scientific publications referenced in the specification above: (1) X. Chen, Y.F. Shen, S. L. Suib, and C.L. O'Young, "Characterization of Manganese Oxide Octahedral Molecular Sieve (M-OMS-2) Materials with Different Metal Cation Dopants", Chem. Mater. 2002, 14, 940-948. (2) G.G. Xia, Y.G. Yin, W.S. Willis, J.Y. Wang, and S . Suib, "Efficient Stable Catalysts for Low Temperature Carbon Monoxide Oxidation", J. Catal. 1999, 185, 91-105. (3) R.N. DeGuzman, Y.F. Shen, E.L Neth, SL. Suib, C.K. O'Young, S. Levine, and J.M. Newsam, "Synthesis and Characterization of Octahedral Molecular Sieves (OMS-2) Having the Hollandite Structure", Chem. Mater. 1994, 6815-821. (4) S. L. Suib, C-L OYoung, "Synthesis of Porous Materials", M. L.

Occelli, H. Kessler, eds, M. Dekker, Inc., p. 215, 1997. (5) D.C. Golden, C.C. Chen, and J.B. Dixon, "Synthesis of Todorokite", Science, 1986, 231, 717-719. (6) D.C. Golden, C.C. Chen, and J.B. Dixon, "Transformation of birnessite to buserite, todorokite, and manganite under mild hydrothermal treatment", Clays

Clay Miner. 1987, 35, 271-280. (7) J.P. Parant, R. Olazcuaga, M. Devalette, C. Fouassier, and P. Hagenmuller, "Sur Quelques Nouvelles Phases de Formule Na_xMnO₂ (x < 1)", /. Solid State Chem. 1971, 3, 1-11. (8) Y.F. Shen, R.P. Zerger, R.N. DeGuzman, SL. Suib, L. McCurdy, D.I.

Potter, and C.L. O'Young, "Manganese oxide octahedral molecular sieves: preparation, characterization, and applications", Science, 1993, 260, 511-515 (9) G.G. ?Xia, W. Tong, E.N. Tolentino, N.G. Duan, S.L. Brock, J.Y. Wang, S.L. Suib, and T. Ressler, " Synthesis and characterization of nanofibrous sodium manganese oxide with a 2 x 4 tunnel structure", Chem. Mater, 2001, 13, 1585-1592. (10) R. Burch, "Knowledge and Know-How in Emission Control for

Mobile Applications," Catalysis Reviews, Vol. 46, No. 3-4, pp. 271-333, 2004. (11) Fung Ming Chang & Martin Jansen, "Agl.8Mn8O16: Square Planar Coordinated At+ Ions in the Channels of a Novel Hollandite Variant," Agnew. Chem. Int. Ed. Engl. 23, No. 11, pp. 906-907, 1984. (12) Fung Ming Chang and Martin Jansen, "Der Erste Silberhollandit,"

Revue de Chimie Minerale, Vol 23(1), pp. 48-54, 1986.

Claims

WHAT IS CLAIMED IS:

1. A method for emissions control comprising: at a location upstream from a particulate filter or NOx trap, removing a substantial quantity of a sulfur oxide from a combustion exhaust by contacting the exhaust with a quantity of at least one sorbent comprising a manganese-based octahedral molecular sieve (Mn-OMS).

2. The method of claim 1 wherein the at least one sorbent has the formula X_aMn₈O₁₆ wherein X is selected from the group consisting of alkali metals, alkaline earth metals, and transition metals, and a is between 0.5 and 2.0.

3. The method of claim 2 wherein X comprises potassium or silver.

4. The method of claim 1 wherein the contacting is at a temperature of at least about 100°C.

5. The method of claim 4 wherein the Mn-OMS has a sulfur dioxide absorption capacity of at least about 40% by weight at the contacting temperature.

6. The method of claim 1 wherein the exhaust is from a motor vehicle.

7. The method of claim 6 wherein the exhaust is a diesel engine exhaust.

8. The method of claim 7 wherein the exhaust cycles between a lean and rich condition.

9. The method of claim 1 wherein MnSO is formed by the contacting, the method further comprising regenerating the sorbent after the contacting by reacting the MnSO with KOH and O₂.

10. The method of claim 1 wherein the contacting removes at least about 90% of the sulfur oxide from the exhaust.

11. The method of claim 1 wherein the Mn-OMS is formed from MO₆ octahedra connected together such that the structure generates micropores in the form of channels of AxB octahedra, wherein A and B are integers from 2 to 4.

12. The method of claim 10 wherein more than 50% of the M elements by mole are manganese.

13. The method of claim 12 wherein a cation selected from H^1", ?NH₄ ⁺, Li⁺, Na⁺, Ag⁺, K⁺, Rb⁺, Tl⁺, Cs⁺, Mg²⁺, Ca²⁺, Sr²⁺, Ba²⁺, Ra²⁺, Cu²⁺, Pb²⁺ is located is the channels.

14. The method of claim 13 wherein the cation is K⁺ or Ag+.

15. The method of claim 11 wherein up to 5% of the M elements are dopant selected from copper, chromium, iron, nickel, cobalt, zinc, aluminum, gallium, titanium, tin, lead, antimony, indium, silicon, germanium, titanium and combinations thereof.

16. The method of claim 15 wherein at least 1% of the M elements are the dopant.

17. A system for emissions control comprising: a source of a combustion exhaust stream; and first and second emissions control devices receiving the exhaust stream; wherein the first emission control device is upstream from the second emission control device; wherein the first emission control device contains at least one sorbent for removing a substantial quantity of sulfur dioxide from the exhaust stream; wherein the at least one sorbent comprises a manganese-based octahedral molecular sieve (Mn-OMS).

18. The system of claim 17 wherein the second emission control device is a particulate filter or NOx trap.

19. The system of claim 18 wherein the source of a combustion exhaust stream is the engine of a vehicle.

20. The system of claim 18 wherein the engine is a diesel engine and the exhaust stream cycles between a rich and a lean condition.

21. The system of claim 20 wherein the sorbent has a sulfur dioxide absorption capacity greater than about 40% by weight at a temperature greater than 200°C.

22. The system of claim 17 wherein the Mn-OMS is a 2x2 structure.

23. The system of claim 17 wherein the first emission control device includes a quantity of the sorbent sufficient to remove at least about 90% of the sulfur dioxide in the exhaust stream over at least 24 hours of normal operation of the source of the combustion exhaust stream.

24. The system of claim 23 wherein the sorbent has a 2x2 structure.

25. A method for removing sulfur dioxide from a gas comprisin : removing at least about 95% of the sulfur dioxide in a gaseous stream by passing the gaseous stream through a sorbent bed, wherein the sorbent bed includes a manganese-based octahedral molecular sieve (Mn-OMS) on a support.

26. The method of claim 25 wherein the gaseous stream is less than 1 molar percent oxygen.

27. The method of claim 25 wherein the gaseous stream is substantially devoid of oxygen.

28. The method of claim 25 wherein the gaseous stream is a combustion exhaust.

29. A low emission motor vehicle comprising: a combustion engine for powering the vehicle; a first emission control device containing at least one sorbent and receiving an exhaust of the engine; and a second emission control device downstream from the first emission control device for removing particulates and or NO_x from the exhaust which has passed through the first emission control device; wherein the at least one sorbent includes a quantity of a manganese-based octahedral molecular sieve (Mn-OMS) for substantially reducing the amount of

SO₂ that would otherwise enter the second emission control device.

30. The motor vehicle of claim 29 wherein the engine is a diesel engine.

31. The motor vehicle of claim 29 wherem the first emission control device includes a housing having a window for determining the color of the sorbent.

32. The motor vehicle of claim 29 wherein the sorbent has a 2x2 structure.

33. A method comprising: substantially reducing the levels of sulfur oxides in an exhaust gas by contacting the exhaust gas with a material selected from materials with structure type OMS 2x2, OMS 2x3 and OMS 3x3, formed fromM0₆ octahedra connected together such that the structure generates micropores in the form of channels, said octahedra comprising at least one element (M) selected from elements from groups mB, IVB, VB, VLB, VIIB, VIII, IB, im and IHA of the periodic table and germanium wherein at least a major portion of element (M) is manganese, said material further comprising at least one element (B) selected from the group formed by the alkaline elements, the alkaline-earth elements, the rare earth elements, the transition metals and elements from groups IHA, IVA of the periodic table.

34. The method of claim 33 wherein the average valence of the metals (M) is between about +3.5 and +4.

35. The method of claim 33 wherein at least a major portion of said element (B) is selected from potassium, silver, sodium, strontium, copper, zinc, magnesium, rubidium and calcium and mixtures of at least two of said elements.

36. The method of claim 35 wherein at least 75% of the element M is manganese.

37. An emissions control device comprising: a housing defining an inlet and an outlet; wherein a quantity of a sulfur oxide sorbent is contained in the housing; wherein the sorbent comprises a manganese-based octahedral molecular sieve (Mn-OMS); wherein the housing includes a window for monitoring the color of the sorbent.

38. The device of claim 37 wherein the inlet is coupled to an exhaust stream from the internal combustion engine of a motor vehicle.

39. A method comprising: removing sulfur oxides from the exhaust of an internal combustion engine with an emissions control device according to claim 37; and removing the emissions control device from the exhaust based on color change of the sorbent.

40. A method for reducing emissions comprising: providing at least one sorbent comprising a manganese-based octahedral molecular sieve (Mn-OMS) upstream from a NOx trap; providing a combustion exhaust that cycles between a lean and a rich condition for regenerating the NOx trap; and passing the combustion exhaust through the at least one sorbent and the NOx trap.

41. The method of claim 40 wherein the at least one sorbent has the formula X_aMn₈Oι₆ wherein X is selected from the group consisting of alkali metals, alkaline earth metals, and transition metals, and a is between 0.5 and 2.0.

42. The method of claim 41 wherein X comprises potassium or silver.

43. The method of claim 42 wherein in successive lean rich cycles, the duration of the lean condition is at least 5 times the duration of the rich condition.

44. A method comprising: producing silver hollandite by contacting cryptomelane with a silver salt melt.

45. The method of claim 44 wherein the contacting is at a temperature of between 200°C and 800°C for at least 1 hour.

46. The method of claim 44 wherein the silver salt is selected from AgNO₃, AgF, AgBr, AgCl, Agl, Ag₂SO₄, and silver organic acid salt.