Deactivation of SCR Catalysts

The major concern in applying selective catalytic reduction is deactivation or poisoning of the catalyst. One cause of deactivation may be catalyst poisons present in the flue gases. 

Clearly, a decrease in the activity of the catalyst for SCR leads to an increasing emission of NO and unreacted ammonia. 

The main cause of deactivation are elements or compounds which chemically attack the catalytically active material or its support. Also, structural changes and pore blocking are important issues of deactivation. A variety of poison compounds containing elements such as halogens, alkah metals, alkaline earth metals, arsenic, lead, phosphorus, and sulfur are mentioned in the hterature. AS2O3 is the most severe poison in coal-fired power plant operation in Germany. In power plants equipped with wet-bottom boilers alkah metal oxides mostly remain in the molten ash, whereas AS2O3 tends to escape into the flue gas and deposits on the catalyst. 

In this section the following issues will be discussed  

Two types of deactivation studies may be distinguished those in which catalysts are doped with the catalyst poison and those in which the catalyst is deactivated by compounds present in flue gases. 

---

Deactivation is not only a problem for oxidation catalysts in combustion of bio-fuel, also SCR catalysts used in biomass fined boilers are deactivated. Andersson et al. [8] investigated the deactivation of SCR catalysts in a couple of different large scale bio-fuelled boilers. The SCR catalyst works at lower temperature (not above 400 C). According to the authors, no loss of surface area occurs and the deactivation is explained by adsorption of gaseous potassium on acidic site of the catalyst. The catalyst can only be partly regenerated by washing m acidic solution. The reason for this difference between precious metal based oxidation catalyst and vanadium pentaoxide... 

It was shown by DRIFT and XAS that AS2O3 was oxidized at the surface of all catalysts forming orthoarsenate(V) species. These species are anchored to the support and to M0O3 and to WO3 as well. There is evidence from XANES measurements that W-O-As bonds are present at the surface and that these types of bonds are the cause for the deactivation of SCR catalysts. 

Honnen et al. [121] found reversible deactivation of SCR catalysts downstream a flue gas desulphurization plant of a waste incinerator. The SO2 concentration varies from 40 to 80mgm and results in the formation of ammonium sulfates at temperatures below 600 K. These sulfates could be removed easily by heating the catalyst to a temperature of about 660 K. 

Deactivation of SCR catalysts also occurs by solid-state reactions between the reagents or poisons and the catalytically active surface. The active metal oxides are thus reduced (or over oxidized) to inactive oxidation states. As an example, if, as often presumed, is the active form of vanadium, then the formation of vanadium pentoxide (as a separate phase) would result in a loss of activity. Alternately, vanadium may be reduced to vanadium +3 or less which, again, may be inactive. 

Effects of Sulfur- Deactivation of SCR catalysts by sulfur compounds can occur by several mechanisms, and the effects can be strongly dependent on temperature, gas composition, and the composition of the catalyst and its support. The most commonly encountered sulfur problems arise from the SO2 in combustion gases and/or SO3, which may come primarily from oxidation of part of the SO2 by the SCR catalyst. 

Dlugi and Gusten S observed that an acidic fly ash from a coal-fired pilot plant was a strong catalyst for the oxidation of SO2 in flue gas to SO3. This can lead to deactivation of SCR catalysts for NO reduction of NH3 through the formation and deposition of NH4HSO4 and (NH4)2 SO4 on the SCR catalyst in the lower temperature range of common operation. 

Cost of the catalyst. The transition metals used, such as rhodium, ruthenium, iridium or palladium, are extremely expensive. The same holds for complicated chiral ligands that often take six to ten synthetic steps for their production. An excellent way to beat these costs is to develop a highly active catalyst. A substrate catalyst ratio (SCR) of 1000 is often quoted as a minimum requirement. In the celebrated Metolachlor process, a SCR of over 100000 is possible. Factors determining the rate of reaction are numerous and often poorly understood. Deactivation of the catalyst also has a profound effect on the overall rate of the reaction. 

The main causes of the deactivation of diesel catalysts are poisoning by lubrication oil additives (phosphorus), and by SOx, and the hydrothermal instability. The SCR by HC is less sensitive to SOx than the NO decomposition. The Cu-based catalysts are slightly inhibited by water vapor and SOx, and suffer deactivation at elevated temperature. Noble metal catalysts such as Pt-MFI undergo low deactivation under practical conditions, are active at temperatures below 573 K but the major and undesired reduction product is N20 (56). 

The effect of SO-j on tlie rate of the selective catalytic rcductiori(SCR) of NO by NH3 over a mordenile type ztiolUe catalyst has been examined in a flotv reactor system. The deactivation of the catalyst is strongly dependent on the I eaction temperature and independent of the SO2 feed concentration. The sulfur content of the catalyst and its surface area appear to be dominant deactivation parameters. The catalytic activity is inversely related to the sulfur content of the catalyst. 

Additionally, in a microreactor the intrinsic kinetics and deactivation behavior of SCR catalysts is studied with flows up to 1.5m h . In both test facilities it is possible to vary all process parameters temperature, the ammonia to nitric oxide feed ratio, the nitric oxide and sulfur dioxide concentrations, the space velocity, and the catalyst geometry. These techniques provide information for somewhat small areas and therefore should always be performed to complement bench- or laboratory-scale activity and selectivity measurements. 

Vanadia supported on silica(with titania) promoted by Fe and CTu oxides was studied for SCR of NOx by Bjorklund et al.28 Both Fe and Cu (as oxides) enhanced the activity however, the resistance to deactivation by SO2 differed as the Fe-promoted catalyst became slightly more active with time on stream and the Cu-promoted catalyst decreased in activity to less than half the initial activity with time on stream. The activity for SCR was related to the concentration of as inferred from the solid electrical conductivity. In this case different promoters were shown to change dramatically the ability of a potential poison to deactivate the SCR catalyst Nikolov, Klissurski, and Hadjiivanov29 also studied the deactivation of vanadia/titania. A combination of ESCA, XRD, and IR were employed to characterize the surface and bulk compositions. They concluded that deactivation involved the transformation of the active anatase titania to inactive rutile. Further, there was a concomitant decrease in total surface area and a loss of phosphate promoters for the selective oxidation of xylenes to phthalic anhydride. 

In the present review, the principle causes of SCR catalyst deactivation are considered under five categories, and discussed in the order of (1) sulfur compounds, (2) alkali metal and alkaline earth metal compounds, (3) arsenic and other heavy metal compounds, (4) fouling or masking by deposits, with pore blocking or surface coating, and (5) thermal degradation. 

When the SO2 was passed through an SO2 oxidizer preceeding the reactor, there was deactivation at 330 °C even with SO3 at only 28 ppm. They concluded that the deactivation mechanism above -300 °C involves a balance between the rate of SO2 oxidation to SO3 and the adsorption equilibrium of the SO3. They developed a deactivation model on this basis, which agreed well with the observed reduction in the SCR rate constant. Finally, they showed that the deactivation of this catalyst by SO2 was reversible, and that repeated regeneration to essentially the initial activity could be achieved by heating to 380 °C. 

A simplified mechanism for the deactivation of V2O5 catalyst by alkali metal is shown in Fig. 3.19. The alkali metal ion (Na" in Fig. 3.19) reacts with the acid V-OH site and forms V-O-Na [59]. This blocks both the NH3 adsorption and the = O-V-OH site from the SCR catalytic cycle. This model is in line with the larger poisoning effect by metals with higher basicity. 

Arsenic is a poison for vanadia SCR catalysts known from stationary power plants where it is present as AS2O3 in gas-phase or in fly ashes [65]. AS2O3 is captured irreversibly to the V2O5 surface causing deactivation of the catalyst [59]. However, since As is not a common compound in diesel exhaust, this type of poisoning is not seen in mobile SCR applications. 

To understand the deactivation trend of the Cu/beta catalyst as a function of exposure time to the 670 °C/4.5 %H20 hydrothermal aging, SCR activities of the aged catalysts at 200 °C were measured at a very high space velocity (SV = 1,00,000 h ). This would render the overall NOx conversions low enough for a SCR reaction rate constant (k) to be calculated based on pseudo first-order kinetics for NO [39]. The results are plotted in Fig. 5.4. The initial 64 h of hydrothermal exposure causes a rapid deactivation of the catalyst as demonstrated by a sharp decline in the SCR reaction rate constant. Extending the aging time beyond 64 h further reduces the activity of the catalyst but at a slower rate. 

However, some challenges related to the use of urea remain. Urea solution is more difficult to dose and to mix with the exhaust gas than NH3 gas-urea may decompose incompletely, form deposits and even deactivate the SCR catalyst [11-13]. These issues are subject of ongoing research and will be treated in Sect. 16.2. 

Ammonia feed is shut off by a low temperature switch when the operating flue gas temperature drops below the minimum recommended value. This prevents deactivation of the catalyst from ammonium bisulfate deposition. This control feature is also applied during system startup and shutdown. An economizer bypass is used to maintain the flue gas temperature above the minimum recommended SCR operating temperature and an SCR bypass is shown (though not often provided). The SCR bypass is used to protect the SCR catalyst during startup and shutdown when the flue gas temperature can be below its dew point. Economizer bypasses are used at the Chambers, Indiantown, and Keystone power plants (Franklin, 1993). 

Apart from the hydrolysis step, the SCR-urea process is equivalent to that of stationary sources, and in fact the key idea behind the development of SCR-urea for diesel powered cars was the necessity to have a catalyst (1) active in the presence of 02, (2) active at very high space velocities ( 500.000 per hour based on the washcoat of a monolith) and low reaction temperatures (the temperature of the emissions in the typical diesel cycles used in testing are in the range of 120-200°C for over half of the time of the testing cycle), and (3) resistant to sulphur and phosphorus deactivation. V-Ti02-based catalysts for SCR-NH3 have these characteristics and for this reason their applications have also been developed for mobile sources. 

Even in the case of an ideal dosing strategy, ammonia slip cannot be excluded if the SCR catalyst is cold, undersized, deactivated or if load jumps are large and very fast. As a consequence, the addition of an oxidation catalyst downstream of the SCR catalyst has... 

With the honeycomb system, the catalyst pitch determines the solids handling capabilities. Pitch is the distance from centerline to centerline of one gas path in the honeycomb and typically varies from 2 to 9 mm with higher pitches being used for heavier dust applications. Typical FCC SCR catalyst would have approximately a 5 mm pitch. The following causes can lead to catalyst deactivation ... 

Fig. 4A, B C show the activity change of mordenite catalysts as a function of copper content on catalyst for the reduction of NO with the sulfur content deposited on catalyst surface. Note that catalytic activity was defined as the ratio of the reaction rate for a deactivated catalyst to that for a fresh catalyst based on the first-order reaction kinetics a = k/k. The effect of sulfur compounds deposited on the catalysts due to the presence of S02 in the feed gas stream on SCR activity significantly depends on both the reaction temperatures and the copper content of the catalyst. For HM catalyst, the catalytic activity varies with its sulfur content depending on reaction temperatures, i.e., an exponential relationship at 250 °C and a linear relationship at 400 DC as shown in Fig.4A. It has already been investigated that the surface area of deactivated HM catalyst exponentially decreases with sulfur content at lower temperature of 250 °C, while it linearly decreases at higher temperature of 400 aC as shown in Fig. 1 A. Judging from these results between catalytic activity and surface area with their catalyst sulfur content at two different reaction temperatures, the decline of the catalytic activity for deactivated HM catalyst occurs simply due to the decrease of surface area.

The dependence of SCR activity of CuHM31 on its sulfur content with respect to reaction temperatures also shows the similar behavior to HM catalyst as shown in Fig. 4B. The catalytic activity reveals an exponential decrease with the sulfur content of the catalyst at 250 °C, while no deactivation is observed at 400 °C, despite the deposition of sulfur up to 1.78 wt.% on the catalyst surface. As discussed in the previous study, it is probably due to the deposition location of the deactivating agents on the pores of catalyst structure (ref 1). 

---