Patent Application
20160136576
The present invention concerns the selective removal of nitrogen oxides (NOx) from gasses. In particular, the invention concerns highly alkali metal resistant, iron- or copper-zeolite catalysts and the use of said catalysts for removal of NOx from exhaust or flue gases, said gases comprising alkali or earth alkali metals. Such gases comprise for example flue gases arising from the burning of biomass, combined biomass and fossil fuel, and from waste incineration units. The process comprises the selective catalytic reduction (SCR) of NOx, such as nitrogen dioxide (NO2) and nitrogen oxide (NO) with ammonia (NH3) or a nitrogen containing compound selected from ammonium salts, urea or a urea derivative as reductant.
The selective catalytic reduction (SCR) of nitrogen oxides (collectively denoted NOx) with ammonia as the reducing agent is an important process in the avoidance of harmful emissions from combustion and high temperature processes (H. Bosch et al., Catal. Today, 1988, 2, 369; G. Busca et al., Appl. Catal. B, 1998, 18, 1; P. Forzatti et al., Heterog. Chem. Rev., 1996, 3, 33; S. Brandenberger et al., Catal. Rev., 2008, 50, 492).
The process is currently being used extensively to reduce the NOx from stationary sources (especially power plants) and SCR technology is also increasingly being employed in the reduction of NOx from automotive vehicles. The current industrial catalyst of choice for the SCR reaction is vanadia supported on anatase phase titania—often promoted with WO3 or MoO3. The V2O5/WO3(MoO3)—TiO2 catalyst does however have some limitations in the form of toxicity and in the form of a limited stability and selectivity at higher temperatures (S. Brandenberger supra). Conventional vanadia based catalysts have a weakness in the form of a susceptibility to poisoning by alkali metals (J. P. Chen et al., J. Catal., 1990, 125, 411; J. Due-Hansen et al., J. Catal., 2007, 251, 459; Y. Zheng et al., Ind. Eng. Chem. Res., 2004, 43, 941). This is a significant drawback in the combustion of alkali rich fuels like straw, and this problem of alkali poisoning has spread the search for alternative SCR catalysts with a greater alkali tolerance. As of now there is no such commercial catalyst for the biomass fired flue gas NOx treatment.
Because of the severe environments in which the SCR reaction is conducted, long-term deactivation has been an important practical problem. Although the causes for deactivation are many and complex, chemical deactivation is a major cause and it is directly related to the mechanism of the SCR reaction. Among chemical deactivation alkali poisoning is very severe when biomass is used as a fuel.
There is consequently still a need for developing SCR catalysts which may function well under the specific and very demanding conditions of biomass incineration, and at the same time be sufficiently robust to allow for uninterrupted performance over long time periods. Due to the toxicity issues associated with Vanadium, there is also a need for developing vanadia-free SCR catalysts.
The unique properties of zeolites could convey a good alkali tolerance to the catalyst, so Metal/Zeolite systems may have an application in NOx removal from stationary emitters. In the present work, we report two efficient and promising alkali resistant, vanadia-free catalysts for biomass fired SCR application: a Cu-zeolite and a Fe-zeolite catalyst.
U.S. Pat. No. 7,264,785 mentions a catalyst system for Internal Combustion Engines (i.e. non-stationary) comprising different metals supported on a zeolite for use in Selective Catalytic Reduction (SCR) of NOx by ammonia. The reference also mentions SCR of NOx in exhaust gases by ammonia, but only achieves about 50-65% total conversion of NOx. The role of the zeolite is to absorb humidity and to catalyze the conversion of ammonia precursors such as urea to ammonia. Other references mention zeolitic catalyst systems which are impregnated/ion exchanged with metal/metal ions, eg. U.S. Pat. No. 6,528,031 B1 which only discusses noble metals, US 2008/0127638 A1 which discusses the platinum group metals, U.S. Pat. No. 7,005,116 B2 which only discusses the use of transition metals, MORDENITE is not found to be a suitable zeolite either, U.S. Pat. No. 5,059,569 A1 which discusses Cu, V, W, Fe, Co and Mo on zeolites with a SiO2/Al2O3 ratio of 4-6, US 2007/0134146 A1 which is directed to copper-on-Y-zeolite catalysts with a typical loading of about 5% metal oxide, U.S. Pat. No. 5,260,043 A1 which only discusses Co, Ni, Fe, Cr, Rh, Mn and not Cu. US 2010/0075834 A1 discloses a preparation of metal-doped zeolites by grinding a dry mixture of a zeolite with a compound of a catalytically active metal, followed by heating the mixture in a reactor. The obtained catalyst can be used in SCR deNOx reactions. Cu, Co, Rh, Pd, Ir, Pt, Ru, Fe, Ni, and V are mentioned, only a Fe based catalyst is exemplified and not Cu. Due to the vastly different manufacturing processes, the obtained catalysts are not easily comparable.
None of these references mention Cu or Fe as the preferred catalytic metal, and none mention the selective catalytic reduction of NOx in exhaust or flue gases obtained from burning biomass. Also, no reference discusses the problem of alkali metals being present in exhaust gases released on burning biomass, which will normally lead to fast and irreversible poisoning of standard commercial SCR deNOx catalysts.
The first aspect of the present invention concerns the use of a zeolite catalyst in the selective removal of nitrogen oxides (SCR) from gases containing a significant amount of alkali metal and/or alkali earth compounds, which catalyst comprises:
Zeolites are hydrated aluminosilicates with open,wide-meshed frameworks composed of SiO4 and AlO4 tetrahedra. Mordenite is a high-silica zeolite in which the Si/Al ratio of the framework is moderately variable. Mordenite-type zeolites with a SiO2/Al2O3 ratio of 5-40 have been found to be the optimum choice for alkali resistivity.
The second aspect of the invention concerns a method for providing a SCR zeolite catalyst, comprising the steps of:
The third aspect of the invention concerns a process for the selective removal of nitrogen oxides with a nitrogen containing compound selected from ammonia, ammonium salts, urea or a urea derivative from gases resulting from the burning of biomass, combined biomass-fossil fuel, or emerging from stationary waste incineration units, which process comprises using a catalyst obtainable by the method of the second aspect of the invention.
The first aspect of the present invention concerns the use of a zeolite catalyst in the selective removal of nitrogen oxides (SCR) from gases containing a significant amount of alkali metal and/or alkali earth compounds, which catalyst comprises:
In a preferred embodiment the metal is copper. In another embodiment the metal is iron. The metal is present as metal oxides in the final catalyst.
In one embodiment the zeolite has a SiO2/Al2O3 ratio of 5-40. In another embodiment the zeolite has a SiO2/Al2O3 ratio of 5-25. In another embodiment the zeolite has a SiO2/Al2O3 ratio between 10 and 25. In another embodiment the zeolite has a SiO2/Al2O3 ratio between 5 and 10. In another embodiment the zeolite has a SiO2/Al2O3 ratio of 5-15. In preferred embodiments the zeolite has a SiO2/Al2O3 ratio selected from 40, 25, 15, 10 or 5.
In one embodiment the catalyst has a M/Al ratio of 1.60 or lower. In another embodiment the catalyst has a M/Al ratio of 1.26 or lower. In another embodiment the catalyst has a M/Al ratio of 0.78 or lower. In another embodiment the catalyst has a M/Al ratio of 0.58 or lower. In another embodiment the catalyst has a M/Al ratio of 0.44 or lower. In another embodiment the catalyst has a M/Al ratio of 0.38.
In a further embodiment the zeolite is selected from mordenite zeolites in protonated form.
In a further embodiment the zeolite is selected from HMORDENITE, and other Mordenite type zeolites. In a particularly preferred embodiment the zeolite is HMORDENITE.
In yet another embodiment the zeolite is selected from commercially available mordenite-type zeolites such as 660HOA, AR 1 (zeolite); CBV-10A, CBV-21A; CP 504-20; CPX 51; HS-690; Hydrogen Mordenite; HSH 620HOA, HSZ 600HOA; HSZ 640HOA, HSZ 640NAD; HSZ 650; HSZ 690; HSZ 690HOA; HSZ 690HOD, Izuka Lite; JRC-Z-M; LZM; LZM 8; M zeolites; MOR zeolites; NC 301; NM100S; PQ 511; SP 30; SP 30 (zeolite); Superzeo; T 81; T 81 (zeolite); TSZ 600; TZM 1013; ZM 510; ZPC 10A, Zeocat FM 8; Zeocat FM 8/25H, Zeocros CF 815A; Zeocros CF 815B; Zeolite AR; Zeolite MOR; Zeolites, MOR-type; Zeolon; Zeolon400; Zeolon 500H; Zeolon 900 and Zeolon 900Na.
The zeolite support is preferably treated with a solution of a suitable Fe- or Cu precursor to achieve a final loading of between 3 and 6% w/w of the relevant metal oxide after calcination.
Precipitating iron or copper on zeolites as disclosed in the present application has the technical effect that it allows a high loading of up between 3 and 6% w/w metal (as the corresponding metal oxide in the final catalyst) without exceeding monolayer coverage, in contrast to typical (non-zeolitic) industrial catalysts, which have approximately half the capacity of the zeolitic catalysts of the present invention. The thus obtained catalysts showed remarkable SCR activity (as measured by the rate constant for the SCR process) cf.
In a preferred embodiment the calcined zeolite catalyst contains between 3 and 6% w/w Cu (as copper oxides). In an even more preferred embodiment the calcined zeolite catalyst contains between 4 and 5% w/w Cu. In particularly preferred embodiments the calcined zeolite catalyst contains around 4% w/w Cu or around 5% w/w Cu.
In a further preferred embodiment the calcined zeolite catalyst contains between 3 and 6% w/w Fe (as iron oxides). In particularly preferred embodiments the calcined zeolite catalyst contains around 3% w/w Fe or around 5.6% w/w Fe.
In a preferred embodiment the zeolite catalyst comprises 4% w/w Cu (as copper oxides) on the HMORDENITE zeolite support.
In another embodiment the zeolite catalyst comprises 4% w/w Cu (as copper oxides) on a mordenite-type zeolite support having a SiO2/Al2O3 ratio of 5-40.
In a further preferred embodiment the zeolite catalyst comprises 3% Fe w/w (as iron oxides) on the HMORDENITE zeolite support.
In another embodiment the zeolite catalyst comprises 3% Fe w/w (as iron oxides) on a mordenite-type zeolite support having a SiO2/Al2O3 ratio of 5-40.
Ammonia is commonly used for the reduction of nitrogen oxides to nitrogen and water by the zeolitic catalysts of the invention, but solid “ammonia-like” materials like ammonium salts, urea and urea derivatives which may be converted to ammonia under the reaction conditions for the selective removal of nitrogen oxides from gases, may be economically viable and less hazardous alternatives to ammonia. Thus, in one embodiment of the invention the selective removal of nitrogen oxides takes place in the presence of an ammonium salt. In another embodiment the selective removal of nitrogen oxides takes place in the presence of urea or a urea derivative. In a preferred embodiment the selective removal of nitrogen oxides takes place in the presence of ammonia.
The catalysts of the present invention display a useful activity over a very wide temperature range which can be tuned for individual purposes by choosing the zeolite carrier and the catalytic metal appropriately. Thus, in one preferred embodiment the selective removal of nitrogen oxides takes place between 320 and 450° C., which is suitable for most traditional stationary incineration plants having been designed for traditional SCR catalysts. In another preferred embodiment the selective removal of nitrogen oxides takes place in equipment suited for higher temperatures between 450 and 550° C. In general, the Cu-based zeolite catalysts of the present invention achieve their maximum rate constant values at higher temperatures than the corresponding Fe-based zeolite catalysts. However, as can be seen from
In a preferred embodiment the invention also provides the use of a zeolite catalyst of the invention which comprises 3-6% w/w Fe or Cu (as iron or copper oxides). In a further embodiment the invention also provides the use of a zeolite catalyst of the invention wherein the zeolite has a SiO2/Al2O3 ratio of between 5 and 40. In a further embodiment the invention also provides the use of a zeolite catalyst of the invention wherein the zeolite has a SiO2/Al2O3 ratio of between 10 and 25. In another embodiment the invention also provides the use of a zeolite catalyst of the invention wherein the zeolite has a SiO2/Al2O3 ratio of between 5 and 10. In another embodiment the invention provides the use of a zeolite catalyst of the invention wherein the zeolite support is selected from HMORDENITE and other mordenite-type zeolites. In a different embodiment, the invention also provides the use of a zeolite catalyst of the invention wherein the selective removal of nitrogen oxides takes place in the presence of ammonia or urea; and at a reaction temperature from about 320° C. to about 550° C.
The second aspect of the invention concerns a method for providing a zeolite catalyst, comprising the steps of:
In a specific embodiment the invention provides a catalyst which is obtainable by the method of the second aspect of the present invention.
The Fe precursor is conveniently chosen from iron nitrate or another aqueously soluble iron compound known to the skilled person. The Cu precursor is conveniently chosen from copper nitrate or another aqueously soluble copper compound known to the skilled person.
In a further embodiment the zeolite catalysts obtained by the method of the second aspect have a light-off temperature (i.e. the temperature at which a catalytic converter achieves a 50% conversion rate) of between 425-475° C. In another embodiment the zeolite catalysts obtained by the method of the second aspect have a light-off temperature around 400° C. In general, Cu-based zeolite catalysts of the present invention have lower light-off temperatures and higher maximum rate constant values than the Fe-based analogues.
In a preferred embodiment of the invention, the zeolite catalysts obtained by the method of the second aspect of the present invention have a large surface area and a high total acidity.
The third aspect of the invention concerns a process for the selective removal of nitrogen oxides with a nitrogen containing compound selected from ammonia, ammonium salts, urea or a urea derivative from gases resulting from the burning of biomass, combined biomass-fossil fuel, or emerging from stationary waste incineration units, which process comprises using a catalyst obtainable by the method of the second aspect of the invention.
Among the examined zeolite catalysts 5% Cu on HBEA (Cu-BEA) showed the highest catalytic activity followed by 4% Cu on HMORDENITE (Cu-MOR), 5% Cu on HZSM5 (Cu-ZSM5). 5% Cu on HBEA (Cu-BEA) catalysts showed a maximum rate constant value of 2646 cm3/g·s, whereas 4% Cu on HMORDENITE (Cu-MOR) and 5% Cu on HZSM5 (Cu-ZSM5) catalysts showed maximum rate constant values of 2542 and 2315 cm3/g·s, respectively. All the examined Fe-zeolite catalysts showed maximum rate constant values of around 1600 cm3/g·s.
The 4% Cu on HMORDENITE (Cu-MOR) and 5% Cu on HZSM5 (Cu-ZSM5) catalysts showed maximum catalytic activity at around 430° C., whereas the 5% Cu on HBEA (Cu-BEA) catalyst showed maximum catalytic activity at around 500° C.
The 3% Fe on HMORDENITE (Fe-MOR) and 3.2% Fe on HZSM5 (Fe-ZSM5) catalysts showed maximum catalytic activity at around 525° C., whereas the 5.6% Fe on HBEA (Fe-BEA) catalyst showed maximum catalytic activity at around 500° C.
It was surprisingly found that the catalytic activity of the zeolite catalysts obtained by the method of the second aspect of the present invention can be maintained even when the catalyst is exposed to moisture.
Thus, in another preferred embodiment the invention concerns a process for the selective removal of nitrogen oxides with a nitrogen containing compound selected from ammonia, ammonium salts, urea or a urea derivative from gases resulting from the burning of biomass, combined biomass-fossil fuel, or emerging from stationary waste incineration units, which gases contain significant amounts of moisture, typically between 2-20% H2O or between 10-15% H2O, which process comprises using a catalyst obtainable by the method of the second aspect of the invention.
It was furthermore surprisingly found that the zeolite catalysts obtained by the method of the second aspect of the present invention show high poisoning resistivity after doping with potassium oxide (500 μmol/g) and therefore are capable of maintaining a high catalytic activity even when exposed to gases containing significant amounts of alkali metal and/or alkali earth compounds. The poisoning resistance is believed to be due to a unique combination of high surface area, acidity and micropore structure of the zeolite support. In general, the Fe-based zeolite catalysts of the present invention were more resistant to potassium poisoning than the corresponding Cu-based zeolite catalysts. Thus the Fe-MOR catalyst retained about 70% of its activity at a loading of 500 μmol/g potassium, whereas the corresponding Cu-MOR catalyst only retained about 55% of its activity at the same potassium level.
Accordingly, a further embodiment of the invention concerns a process for the selective removal of nitrogen oxides with a nitrogen containing compound selected from ammonia, ammonium salts, urea or a urea derivative from gases resulting from the burning of biomass, combined biomass-fossil fuel, or emerging from stationary waste incineration units, which gases contain significant amounts of alkali metal and/or alkali earth compounds, such as, for example, up to several hundred mg potassium per m3 gas, which process comprises using a catalyst obtainable by the method of the second aspect of the invention.
In the context of the present invention, the terms “around”, “about”, or “approximately” are used interchangeably and refer to the claimed value, and may include variations as large as ±0.1%, ±1%, or ±10%. Especially in the case of log10 intervals, the variations may be larger and include the claimed value ±50%, or 100%. The terms “around”, “about”, or “approximately” may also reflect the degree of uncertainty and/or variation that is common and/or generally accepted in the art.
According to one embodiment of the invention, the catalyst according to the invention is provided in a form that provides minimal resistance to the flue gases, such as minimal pressure loss, while still providing reliable catalytic conversion of NOx to N2. Generally, shapes, dimensions and designs of such a catalyst are known in the art.
The catalyst can for example be shaped as a monolith, extrudate, bead, plate, sheet or fibrous cloth, where the active phases can be introduced to the conformed material either by wash-coating, extrusion or spray painting, methods that are generally well-established in the art.
One embodiment of the invention concerns a process of selectively removing nitrogen oxides with ammonia from gases resulting from the burning of biomass, combined biomass-fossil fuel or emerging from waste incineration units at a temperature from about 150° C. to about 550° C., which process comprises using a catalyst obtainable by the method of the second aspect of the invention.
Commonly, for low temperature applications, such as placement of the catalyst unit in the flue gas duct after dust filtration in waste incineration plants, the temperature of the flue gas is in the range of 150-300° C. In the case of high temperature applications, such as placement of the catalyst unit at high dust positions in the flue gas duct, the temperature of the flue gas is often in the range of 340-420° C. For intermediate temperature applications, the temperature of the flue gas is in the area of about 250-370° C. The catalysts of the present invention can be placed at high dust positions in the flue gas duct due to their superior alkali metal poisoning resistivity, which allows them to catalyze the deNOx reaction with a much higher rate constant than if they were placed after a dust filter where the temperature is lower.
Commonly, one or more heat exchange units are provided in order to utilize the thermal energy of the flue gas. In one embodiment, the SCR process according to the invention takes place before a heat exchange unit. In a further embodiment, the SCR process is conducted after a heat exchange unit. In yet another embodiment, the SCR process takes place in between heat exchange units. In still another embodiment, heat controlling means are provided in order to control the temperature of the flue gas before and/or during the SCR. Thereby the efficiency of the SCR process can be controlled and/or optimized for the respective catalyst according to the invention, and its temperature profile with respect to catalytic activity. Such heat controlling means may comprise means to alter the rate of combustion, means to alter the flow of gas and the like. Generally, such means are well-known in the art.
Very often, fuels containing alkali metals as well as earth alkali will also contain significant amounts of alkali metals as well as earth alkali in the resulting flue gases upon incineration or burning. Fossil fuels, such as oil, natural gas and coal contain lower amounts of alkali metals and earth alkali metals. Waste, such as waste burned in waste incineration plants contains high levels of alkali metals as well as earth alkali metals. Biomass or biomass fuel such as straw, woodchips and wood pellets contain very high levels of alkali metals, especially K, as well as earth alkali metals. In the case of fly ash from burning straw, alkali metals and earth alkali metals can comprise as much as half of the total weight of the fly ash. Flue gases stemming from the incineration of biomass fuel typically contain about 200-1000 mg KCl/Nm3, whereas incineration of coal only leads to ppm levels of KCl.
By the use of a catalyst according to the invention, the lifetime can be increased significantly compared to conventional, non-zeolitic catalysts. In one embodiment of the invention, the life time of the catalyst is increased by a factor of at least 1.5; 1.5-3.0; 3.0-5.0; 5.0-10; or 100, compared to a similar/comparable catalyst without zeolitic support. In a further embodiment of the invention, the lifetime of the catalyst according to the invention is 2-5 times compared to a comparable catalyst without zeolitic support. Apart from economical benefits, this also provides a greater flexibility with respect to exchange and/or cleaning of the catalyst. By a larger window of opportunity for when to close the plant for exchange, cleaning, or reactivation of the catalyst, sensitive time periods may be avoided. For many applications, a shut down during summer is less expensive than during winter.
A catalyst according to the present invention can be treated and handled using conventional methods and techniques in the field. The catalyst can also be cleaned/washed and recycled.
The present invention will be better understood after reading the following non-limiting examples.
Cu/zeolite catalysts were prepared by incipient wet impregnation using appropriate amounts of copper nitrate (Aldrich, 99.99%) as a precursor and HMORDENITE (400 m2/g), HZSM5 (500 m2/g) or HBETA (680 m2/g), respectively as supports (HMORDENITE was obtained by protonation of the ammonia form CBV21A zeolite support from Zeolyst International by routine methods). Optimum Cu/zeolite catalysts were then poisoned by incipient wet impregnation with 0.05-0.15 M solution of potassium nitrate (Aldrich, 99.99%) to obtain a potassium loading of 100, 250 and 500 μmol/g. The prepared catalysts were oven dried at 120° C. for 12 h and finally calcined at 500° C. for 5 h before use.
BET surface areas of the samples were determined from nitrogen physisorption measurements on about 100 mg sample at liquid nitrogen temperature with a Micromeritics ASAP 2010 instrument. The samples were heated to 200° C. for 1 h prior to the measurement.
EPR spectra of the catalysts were recorded ex-situ with a Bruker EMX-EPR spectrometer working in the X-band (Bruker ER 041 XGG Microwave Bridge) at microwave frequencies around 9.75 GHz. The measurements were done at room temperature on samples transferred directly after calcination into an excicator. Data treatment was performed with WIN-EPR software provided by Bruker.
NH3-TPD experiments were conducted on a Micromeritics Autochem-II instrument. In a typical TPD experiment, 100 mg of dried sample was placed in a quartz tube and pretreated in flowing He at 500° C. for 2h. Then, the temperature was lowered to 100° C. and the sample was treated with anhydrous NH3 gas (Air Liquide, 5% NH3 in He). After NH3 adsorption, the sample was flushed with He (50 ml/min) for 100 min at 100° C. Finally, the TPD operation was carried out by heating the sample from 100 to 700° C. (10° C./min) under a flow of He (25 ml/min).
H2-TPR studies were also conducted on a Micromeritics Autochem-II instrument. In a typical experiment, 100 mg of oven-dried sample was placed in one arm of a U-shaped quartz tube on quartz wool. Prior to TPR, the catalyst sample was pretreated by flushing with air at 300° C. for 2 h. After pretreatment, the sample was cooled to ambient temperature thereafter the TPR analysis was carried out in a reducing mixture (50 ml/min) consisting of 5% H2 and balance Ar (Air Liquide) from ambient temperature to 900° C. (10° C./min). The hydrogen concentration in the effluent stream was monitored by a thermal conductivity detector (TCD) and the H2 consumption values were calculated from calibration experiments.
The SCR activity measurements were carried out at atmospheric pressure in a fixed-bed quartz reactor loaded with 10 mg of fractionized (180-300 μm) catalyst samples positioned between two layers of inert quartz wool. The reactant gas composition was adjusted to 1000 ppm NO, 1100 ppm NH3, 3.5% O2, 2.3% H2O and balance N2 by mixing 1% NO/N2 (±0.1% abs.), 1% NH3/N2 (0.005% abs.), O2 (99.95%) and balance N2 (≧99.999%) (Air Liquide) using Bronkhorst EL-Flow F-201C/D mass-flow controllers. The total flow rate was maintained at 500 ml/min (ambient conditions). Further studies with SO2 were not performed on these catalysts since biomass fired straw or wood chips have a very low content of sulphur. During the experiments the temperature was increased stepwise from 200 to 600° C. while the NO and NH3 concentrations were continuously monitored by a Thermo Electron's Model 17 C chemiluminiscent NH3—NOx gas analyzer. The activity was measured after attaining steady state and care was taken not to reach conversions above 90% to keep the catalyst stressed. Fresh and poisoned catalysts are compared by change in relative activity (%) of the corresponding catalysts. The Cu-zeolite catalysts are further compared with an industrial V2O5/WO3—TiO2 (VWT) catalyst (3% V2O5-9% WO3) which also was poisoned at similar concentration levels of potassium.
Cu-zeolite catalysts were prepared by the ion-exchange method. Commercial zeolite supports HZSM5 (Si/Al=15 and 500 m2/g), HMOR (Si/Al=10 and 400 m2/g) and HBEA (Si/Al=25 and 680 m2/g) were purchased from Zeolyst International. 2 g of zeolite sample was added to 1 liter of 2 mM copper nitrate (Aldrich, 99.99%) solution. The mixture was stirred for 24 h at 80° C., whereafter the solid iron-exchanged zeolite catalyst was filtered off, washed there times with 1 liter water, dried at 120° C. for 12 h and finally calcined at 500° C. for 5 h. Catalysts prepared by this method are represented as Cu-ZSM5, Cu-MOR and Cu-BEA. Surface Cu content of the Cu-ZSM5, Cu-MOR and Cu-BEA catalysts measured by EDX analysis were found to be 2.7, 2.9 and 3.4 wt. %, respectively.
The catalytic activity of Cu-zeolite catalysts from 200-600° C. is reported in
Fe-zeolite catalysts were prepared by ion-exchange methods. Commercial zeolite supports HZSM5 (500 m2/g), HMOR (400 m2/g) and HBEA (680 m2/g) were purchased from Zeolyst International. In the ion-exchange method 2 g of zeolite sample was added to 1 liter of 2 mM iron (III) nitrate nonahydrate (Aldrich, 99.99%) solution. The mixture was stirred for 24 h at 80° C., where after the solid iron-exchanged zeolite catalyst was filtered off, washed there times with 1 liter water, dried at 120° C. for 12 h and finally calcined at 500° C. for 5 h. Catalysts prepared by this method are represented as Fe-ZSM5, Fe-MOR and Fe-BEA.
Poisoning of iron-exchanged catalysts were done by incipient wet impregnation method with 0.05-0.15 M solution of potassium nitrate (Aldrich, 99.99%) to obtain a potassium loading of 100, 250 and 500 μmol/g, respectively. The prepared catalysts were oven dried at 120° C. for 12 h and finally calcined at 500° C. for 5 h before use. Catalysts doped with 500 μmol/g of potassium are represented as K—Fe-ZSM5, K—Fe-MOR and K—Fe-BEA, respectively.
BET surface area of the samples was determined from nitrogen physisorption measurements on about 100 mg sample at liquid nitrogen temperature with a Micromeritics ASAP 2010 instrument. The samples were heated to 200° C. for 1 h prior to measurement. Surface composition of Fe was estimated with an Oxford INCA EDX analyzer.
EPR spectra of the catalysts were recorded ex-situ with a Bruker EMX-EPR spectrometer, working at the X-band (Bruker ER 041 XGG Microwave Bridge) at microwave frequencies around 9.75 GHz. The measurements were done at room temperature on samples transferred directly after calcination into excicator. Data treatment was performed with WIN-EPR software provided by Bruker.
NH3-TPD experiments were conducted on a Micromeritics Autochem-II instrument. In a typical TPD experiment, about 100 mg of dried sample was placed in a quartz tube and pretreated in flowing He at 500° C. for 2h. Then, the temperature was lowered to 100° C. and the sample was treated with anhydrous NH3 gas (Air Liquide, 5% NH3 in He). After NH3 adsorption, the sample was flushed with He (50 ml/min) for 100 min at 100° C. Finally, the TPD operation was carried out by heating the sample from 100 to 950° C. (10° C./min) under a flow of He (25 ml/min).
H2-TPR studies were also conducted on a Micromeritics Autochem-II instrument. In a typical experiment, 100 mg of oven-dried sample was placed in one arm of a U-shaped quartz sample tube on a quartz wool plug. Prior to TPR, the catalyst sample was pretreated by flushing with air at 300° C. for 2 h. After pretreatment, the sample was cooled to ambient temperature and the TPR analysis carried out in a reducing mixture (50 ml/min) consisting of 5% H2 and balance Ar (Air Liquide) from ambient temperature to 1000° C. (10° C./min). The hydrogen concentration in the effluent stream was monitored by a thermal conductivity detector (TCD) and the H2 consumption values were calculated from calibration experiments.
The SCR activity measurements were carried out at atmospheric pressure in a fixed-bed quartz reactor loaded with 30 mg of fractionized (180-300 μm) catalyst samples positioned between two layers of inert quartz wool. The reactant gas composition was adjusted to 1000 ppm NO, 1100 ppm NH3, 3.5% O2, 2.3% H2O and balance N2 by mixing 1% NO/N2 (±0.1% abs.), 1% NH3/N2 (0.005% abs.), O2 (≧99.95%) and balance N2 (≧99.999%) (Air Liquide) using Bronkhorst EL-Flow F-201C/D mass-flow controllers. The total flow rate was maintained at 500 ml/min (ambient conditions). During the experiments the temperature was increased stepwise from 200 to 600° C. while the NO and NH3 concentrations were continuously monitored by a Thermo Electron Model 17C chemiluminiscent NO—NOx gas analyzer.
3wt. % Fe-zeolite catalysts were prepared by incipient wet impregnation using appropriate amounts of iron nitrate (Aldrich, 99.9%) as a precursor and HZSM5 (Si/Al=15 and 500 m2/g), HMOR (Si/Al=10 and 400 m2/g) or HBEA (Si/Al=25 and 680 m2/g), respectively as supports (Zeolyst International). The prepared catalysts were oven dried at 120° C. for 12 h and finally calcined at 550° C. for 5 h.
The catalytic activity of 3 wt. % Fe-zeolite catalysts from 200-600° C. is reported in
For both the Cu- and Fe-zeolite catalysts of the present invention, the catalytic activity is represented as the first-order rate constant (cm3/g·s), since the SCR reaction is known to be first-order with respect to NO under stoichiometric NH3 conditions [see eg. R. Q. Long et al. J. Catal. 196 (2000) 73, J. P. Chen et al. J. Catal. 125 (1990) 411, A. L. Kustov et al. Appl. Catal. B 58 (2005) 97, J. Due-Hansen et al. J. Catal. 251(2007) 459. J. Due-Hansen et al. Appl. Catal. B 66 (2006) 161].
The first-order rate constants where obtained from the conversion of NO as:
k=−(FNO/(mcat·CNO))ln(1−X) (1)
where FNO denotes the molar feed rate of NO (mol/s), mcat the catalyst mass, CNO the NO concentration (mol/cm3) in the inlet gas and X the fractional conversion of NO.
The SCR activities of the Cu/zeolite catalysts at 400° C. are reported in
HMORDENITE Si/Al ratio influence on SCR activity was reported (G. G. Park et al., Microporous and Mesoporous Materials 48 (2001) 337-343) for Mordenite type zeolite catalysts having a Si/Al ratio of 5, 10 and 20). The SCR activity of the catalyst at 200° C. was found to be almost constant (70%) for CUHM5 and CUHM10, whilst the CUHM20 catalyst only showed 35% conversion at 200° C. There thus seems to be a gradual decrease in SCR activity going from a Si/Al ratio 5 to 20.
The samples prepared by impregnation were calcined in air. After calcination, most copper existed as copper oxide and cupric ions on the outer shells of the zeolite crystals. EPR spectroscopy has been extensively used to probe the structural environment of paramagnetic copper sites in zeolites.
H2-TPR is frequently used to study the redox property of metal oxide catalysts. In
1K-doped catalysts are prepared with 500 μmol/g of potassium
Bulk copper oxide exhibits only one TPR peak attributed to direct reduction of Cu2+ ions to Cu0 [Á. Szegedi et al. Appl. Catal. A 272 (2004) 257, B. Wichterlova et al. Appl. Catal. A 103 (1993) 269, M. Richter et al. Appl. Catal. B 73 (2007) 269]. Oxide-supported copper shows an analogous reduction behavior [Á. Szegedi et al. Appl. Catal. A 272 (2004) 257]. However, in the case of zeolites a two stage reduction is usually observed (Cu2+ to Cu+ followed by Cu+ to Cu0). Patterns generally similar to ours have been reported for Cu-BEA [R. Kefirov et al. Micropor. Mesopor. Mater 116 (2008) 180] and Cu-ZSM-5 [C. Torre-Abreu et al. Catal. Today 54 (1999) 407, M. F. Ribeiro et al. Appl. Catal. B 70 (2007) 384, M. J. Jia et al. J. Mol. Catal. A 185 (2002) 151]. In general, the reduction temperature depends on the interaction of CuO with the support. We have assigned the sharp low temperature reduction peak to the reduction of highly dispersed CuO aggregates from Cu(II) ions to Cu(I), and the second shoulder peak to the reduction of the Cu(I) species that strongly interact with the exchange sites of zeolite [B. Coq et al. Appl. Catal. B 6 (1995) 271.]. However, Cu-MOR catalyst showed only a low-temperature reduction peak at 210° C. which could indicate direct reduction of Cu2+ ions to Cu0. One stage reduction of Cu2+ ions to Cu0 have also been observed on other Cu/zeolites [S. Kieger et al. J. Catal. 183 (1999) 267, W. Grünert et al. J. Phys. Chem. 98 (1994) 10832].
NH3-TPD is a frequently used method for determining the surface acidity of solid catalysts as well as acid strength distribution.
1K-doped catalysts are prepared with 500 μmol/g of potassium
Generally, pure zeolites showed two ammonia desorption regions; one due to weak acid strength and the other due to moderate acid strength. The weak acid sites were observed at lower temperatures around 200° C., while the moderate acid sites were observed between 400-500° C. The total acidity of the pure zeolites follow the order MOR>ZSM5>BEA. After impregnating Cu on zeolites two extra desorption peaks were observed between weak and moderate acid site shifting the moderate acid site desorption peak to high temperature, while that of the weak acid site remains at the same position. The extra desorption peaks are resulting from the decomposition of the copper ammonia complex at around 320° C. [A. V. Salker et al. Appl. Cat A 203 (2000) 221].
According to the results of ammonia-TPD (Table 2), the total acidity of the Cu/zeolites is: Cu-MOR>Cu-BEA>Cu-ZSM5. Introduction of Cu leads to an increase of the total amount of desorbed ammonia. The difference between the amounts of NH3 desorbed from the Cu/zeolite and the corresponding zeolite is ascribed to the formation of complexes by interaction between ammonia molecules and copper cations. If this value is normalized to the copper content it would provide information about the average stoichiometry of NH3 adsorption on copper species. The ΔNH3/Cu ratio of the catalysts is: Cu-BEA (0.88)>Cu-MOR (0.78)>Cu-ZSM5 (0.44). The increase in NH3 desorption observed on Cu-BEA compared to that of Cu-MOR and Cu-ZSM5 catalyst could be due to large pore size. It is expected that the large pore zeolites have improved accessibility of the Cu-species which are more available for ammonia adsorption.
Activity vs. temperature of Cu-MOR, Cu-ZSM5 and Cu-BEA catalysts are shown in
The potassium poisoned catalysts showed decreased SCR activity with increasing K/Cu molar ratios. At low K/Cu molar ratio up to 0.20 essentially no influence on the SCR activity was observed and only upon further increase of potassium concentration a slight decrease in activity was observed. The superior performance of these catalysts can be observed even at high potassium concentrations, whereas conventional catalysts like, e.g. V2O5/WO3—TiO2, V2O5/Sulphated-ZrO2 and V2O5/WO3—ZrO2 are poisoned severely at a K/V molar ratio of 0.3 where the active hydroxy vanadates are transferred to potassium vanadates [J. P. Chen et al. J. Catal. 125 (1990) 411, J. Due-Hansen et al. J. Catal. 251 (2007) 459, Y. Zheng et al. Ind. Eng. Chem. Res. 43 (2004) 941, A. L. Kustov et al. Appl. Catal. B 58 (2008) 97].
The relative activity of Cu/zeolites and VWT catalysts with different potassium loadings is shown in
Potassium poisoned V2O5/TiO2 catalysts can change redox and acidic properties. Kamata et al. [H. Kamata et al. J. Mol. Catal. A 139 (1999) 189.] made an infrared spectroscopic study, which showed that the addition of K2O to the catalyst modified the structure of the surface vanadium species. K2O added to the catalyst might also partially react with V2O5 to form KVO3. It has also been shown that K2O directly coordinates with the surface vanadium oxide phase [G. Deo, I. E. Wachs, J. Catal. 146 (1994) 335, J. P. Dunn et al. J. Catal. 181 (1999) 233.]. Progressive addition of K2O to V2O5—TiO2 catalyst gradually titrates the surface vanadium oxide sites. Several authors [J. P. Chen et al. J. Catal. 125 (1990) 411, H. Kamata et al. J. Mol. Catal. A 139 (1999) 189, D. A. Bulushev et al. Langmuir 17 (2001) 5276] have reported the effect of alkaline metals on the activity of V2O5/TiO2 catalysts. Most of them conclude that poisonous additives are affecting the Brønsted acid sites which are responsible for the ammonia adsorption, thus decreasing both their number and activity in NO reduction. For the commercial vanadium catalyst the potassium poisoning is very sever due to the change of the active vanadium phase and loss of Brønsted acid sites.
When doping the Cu/zeolites with potassium (500 μmol/g) a slight decrease in catalytic activity occured. Detailed characterization of these deactivated Cu/zeolite catalysts with EPR, H2-TPD and NH3-TPD techniques could give further information on the actual cause of deactivation.
Table 3 shows the surface Fe content of the catalysts measured by EDX analysis. Fe-MOR, Fe-ZSM5 and Fe-BEA catalysts were found to be have Fe composition of 3.0, 3.2 and 5.6 wt. %, respectively. In general, the metal exchange capacity of a zeolite is a function of both the Si/Al ratio, surface area and pore size [S. S. R. Putluru et al. Appl. Catal. B 97 (2010) 333]. In iron-exchanged zeolites Fe cations are located in coordinative unsaturated sites that are accessible to reactant gases with the cations being stabilized in specific environments and oxidation states that may be essential for the catalytic activity. Accordingly, it is important to characterize the local electronic environment of the Fe exchanged zeolites. EPR spectroscopy allows to examine such environments and has been used here to probe the iron (Fe2+, d6; Fe3+, d5) exchanged into the zeolites.
1K-doped catalysts are prepared with 500 μmol/g of potassium
2Fe-content measured by EDX analysis
X-band EPR spectra of Fe3+-containing zeolites usually consist of three different signals: a sharp signal with g=4.3, a broad signal with g=2.0-2.3 and a very sharp signal with g=2.0 [M. Schwidder et al. J. Catal. 231 (2005) 314-330, D. Goldfarb et al. J. Am. Chem. Soc. 116 (1994) 6344]. The commonly accepted assignments of the three signals are: framework iron (tetrahedral lattice Fe3+ ions), iron oxide cluster with Fe3+ neighbours, and iron oxide clusters in cation exchange sites (isolated Fe3+ ions or FeOx oligomers), respectively.
The catalytic activity of undoped and potassium-doped Fe-zeolite catalysts from 200-600° C. is reported in
The potassium-doped Fe-zeolite catalysts with a maximum potassium concentration of 500 μmol/g of catalyst showed decreased catalytic activity compared to that of the fresh catalysts in the order: K—Fe-MOR (780 cm3/g·s)>K—Fe-ZSM5 (680 cm3/g·s)>K—Fe-BEA (300 cm3/g·s) at 525° C.
SCR of NO with ammonia is a feasible reaction on a catalyst having optimal redox and acidic properties. The difference in acidic and redox properties of the freshly prepared and the potassium poisoned catalysts could give more insight about the observed catalyst deactivation. NH3-TPD is a frequently used method for determining the surface acidity of solid catalysts as well as the acid strength distribution [S. S. R. Putluru et al. Appl. Catal. B 97 (2010) 333, G. I. Kapustin et al. Appl. Catal. 42 (1988) 240]. Surface acidity of the zeolites, the Fe-zeolite catalysts and the potassium doped Fe-zeolites catalysts are reported in Table 3. Acidity of the pure zeolites were in the order of MOR>ZSM5>BEA. This order remained also upon Fe exchange into the zeolites, where the acidity increased even further by around 20%. Due to the pronounced acidity originating from the support these are promising acid scavenging materials for alkali resistant SCR applications.
The acidity of the potassium poisoned catalysts were also found to be in the order of K—Fe-MOR>K—Fe-ZSM5>K—Fe-BEA, as expected. However, after doping with potassium (500 μmol/g of catalyst) Fe-MOR, Fe-ZSM5 and Fe-BEA catalysts lost surface acidic sites of 932, 626 and 605 μmol/g of catalyst, respectively, corresponding to about half their acidic sites. Thus, overall the Fe-zeolite catalysts maintained appreciable surface acidity also at high potassium concentration. In comparison, the commercial vanadium catalyst loos all surface acidic sites even at low potassium concentration of 100 μmol/g of catalyst [A. L. Kustov et al. Appl. Catal. B 58 (2005) 97, J. Due-Hansen et al. J. Catal. 251(2007) 459, J. Due-Hansen et al. Appl. Catal. B 66 (2006) 161]. This superior potassium resistivity is directly related to the high surface area and acidity of the zeolite supports as compared to those of conventional metal oxides like, e.g. ZrO2 and TiO2, since these characteristics allow the SCR active metals to be well protected from the potassium poisions by exposing the dense acidic sites as hosts.
The redox properties of the Fe-zeolite and K—Fe-zeolite catalysts were also characterized by H2-TPR method. The obtained results are shown in
| Number | Date | Country | Kind |
|---|---|---|---|
| 10174381.3 | Aug 2010 | EP | regional |
| Number | Date | Country | |
|---|---|---|---|
| 61377672 | Aug 2010 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 13819227 | Mar 2013 | US |
| Child | 15007098 | US |
