CC BY
15УДК 544.431.2:544.476.2:544.478.41 Wladimir Suprun1, D. Worch2, K. Soja3 В. Супрун1, Д. Ворх2, К. Соя3
ADSORPTION OF NITROGEN OXIDES ON Ag-, Cu-AND Fe- OXIDE SUPPORTED CATALYSTS
Institute of Chemical Technology, Universität Leipzig, Linnestraße 3, 04103 Leipzig, Germany,
Faculty of Energy and Fuels, AGH University of Science and Technology, Al. Mickiewicza 30, 30-059 Krakow, Poland e-mail: [email protected]
The series of Ag-, Fe- und Cu oxide containing catalysts supported on y-AhO,SiO2, H-Y and SAPO-11 with 3 wt.-% of metal was prepared by impregnation or ion-exchange methods. The textural, acidic and redox property was determined by N2-Adsoption, H2-TPR und NH3-TPD.The NOx adsorption capacity (ACNOx) was quantified by NOx-TPD methods using on-line chemiluminiscence analysis. It was found, that ACnox depended not only from reactions temperature but also from total acidic site density as well as from Al-content of the mixed oxides. By in-situ DRIFT spectroscopy of catalytic experiments over y-Al2O3, Ag/Al2O3, FeO/Al2O3 and CuO/ Al2O3 it was found that NO in presence of O2 excess was oxidized above 673 K to NO2 and primarily adsorbed in form of nitrates. The accumulated dates indicate that the type of the transition metal strongly affected the oxidation of NO to NO2. Cu and Fe oxide containing samples show higher catalytic activity according to the oxidation of NO to NO2 as Ag containing counterparts. Ag, Cu and Fe oxide SO2 supported samples possess a low-grade ACnox in comparison to J-A2O and zeolite supported counterparts.
Keywords: AgOx, CuOx, FeOx, Catalyst, NOx-Adsorption
DOI 10.15217/issn1998984-9.2018.42.4
Introduction
The selective catalytic reduction (SCR) of NOx by hydrocarbons (HC) or oxygenates such as lower C1-C4 alcohols (ROH) over transition metal oxides supported on alumina, zir-conia, titanium or transition metal-exchanged zeolites in the presence of excess oxygen is a potential method to remove NOx from various exhaust gas streams [1-5]. As an alternative reductant, bio-ethanol for SCR has recently received attention [6]. Several studies on EtOH/SCR focusing on different types of catalysts, such as Ag, Cu, Co and Fe supported withy-Al2O3,BEA, MOR, MFI and SBA-15were reported within the last decade [6-13]. Especially alumina-supported silver catalysts have shown promising results due to their high SCR activity at low temperature (573-623 K) and their resistance towards water and sulfur dioxide inhibition.
Different mechanisms for the SCR reaction involving alkoxy-, acetate-, isocyanate- and NOx-species as intermediates for Ag containing catalysts were proposed [7, 8, 14]. However, no commonly accepted mechanism is established so far. One possible mechanism of SCR with HC and ROH, assumes
АДСОРБЦИЯ ОКСИДОВ АЗОТА НА Ag-, Cu-И Fe- ОКСИДНЫХ КАТАЛИЗАТОРАХ
Институт Технической Химии, Университет Лейпциг, Линне Штрассе 3, г. Лейпциг, 04103, Германия Факультет Энергии и Топлива, Университет Науки и Технологии, ул. А. Мицкевича 30, 30-059 г. Краков, Польша e-mail: [email protected]
Серия Ag-, Fe- и Cu- оксидных катализаторов на основе y-Al2O,SiO2, H-Y содержащих 3 мас.-% металла синтезирована методом импрегнирования и ионного обмена. Текстурные, кислотные и восстановительные свойства определены методами N2-адсорбции, H2-TPR и NH3-TPD. Адсорбционная емкость по отношению к NOx(ACnox) определена методом NOx-TPD с использованием хемилюминесцентного анализа. Установлено, что величины ACnox зависят не только от температуры, кислотных свойств катализаторов, но и от содержания Al в оксидах. Методом ИК-Фурье-спектроско-пии у-АЬОз,и Ag/A2O3, FeO/A2O3 и CUO/AI2O3 катализаторов установлено, что NO в присутствии избытка O2 подвергается окислению до NO2 с образованием поверхностных нитратов. Полученные данные свидетельствуют о том, что природа оксидов металлов переменной валентности влияет на окисление NO в NO2. Катализаторы, содержащие Cu и Fe-оксиды, проявляют большую каталитическую активность к окислению NO в NO2 чем соответствующие Ag-со-держащие катализаторы. Ag-, Cu- и Fe-оксиды, осажденные на SiO2, обладают низкой NOx адсорбционной емкостью по сравнению с катализаторам на основе Y-AI2O3 и цеолитов.
Ключевые слова: AgO x, CuOx, FeOx, Катализатор, NOx-ад-сорбция
the surface reaction between adsorbed intermediates of NOx-and hydrocarbon species [15, 16]. In a previous step, gaseous NO and NO2 were adsorbed on the catalyst surface and transformed to adsorbed nitrate (NO3-) or nitrite (NO2-) species, which in turn can react with the pre-adsorbed HC to various nitro organic species such as nitrito (R-O-N=O), nitro (R-NO2) or nitroso (R-N=O) intermediates. For EtOH/SCR-DeNOx, several reaction steps are postulated in the literature, where the formation of surface-bonded ethanol, acetaldehyde and/or acetate is discussed as a crucial step. Further more the adsorbed acetate- or enolate-species can react with adsorbed NOx to form nitrometh-ane and/or nitroethane, which can dissociate to isocyanate-spe-ciesandfinally hydrolyze to ammonia. For this suggested mechanism either the adsorption of ethanol and NOx or the formation of N-containing organic intermediate compounds play an important role. For the SCR of NOx by NH3 over Cu, Fe and H-form zeolites it is generally accepted that the oxidation of NO to NO2 (or NO2-) and the subsequent decomposition of NH4NO2 formed by the reaction of adsorbed NH4+ and NO2 (or NO2-) are important reaction steps. Further more abundant studies are focused on the SCR of NOx by NH3 over various zeolites but the mechanistic details are still not fully understand [21].
1 Wladimir Suprun, Dr. of Sci. (Chem.), Senior Research Associate, Head of Laboratory of Heterogeneous Catalysis Institute of Chemical Technology, Universität Leipzig, e-mail: [email protected]
Владимир Ярославович Супрун, д-р хим. наук, ст. науч. сотр., зав. лаб. гетерогенного катализа, Институт Технической Химии, Университет Лейпциг
2 Dr. Denis Worch, R&D Manager DBI Gas und Umwelttechik GmbH Karl-Heine Strasse 109/111 D-04229 Leipzig; Germany
Ворх Денис, инженер по развитию, ДВИ Газа и Teхники Окружающей Среды, Карл-Хайне штрассе, г. Лейпциг, Германия, е-mail: [email protected]
3 M.Sc Krzystof Soja, Scientific Coworker, Faculty of Energy and Fuels, AGH-University of Science and Technology, e-mail: [email protected] Кржистоф Соя, науч. сотр., факультет Энергетики и топлива, Краковский университет науки и технологии
Дата поступления 26 декабря 2017 года
According to Park et al. [22] an active catalyst for the HC/SCR needs four different functions: i) ability and activity to oxidize NO to NO2; ii) soft activation of HC; iii) reduction of NOx to N2;iv) oxidation of HC to CO2. In addition, Kim et al. [23] suggested further three important roles of supported transition metal oxidein EtOH/SCR reaction:i) the inhibition of the formation of acetic acid and of direct total oxidation of EtOH; ii) promotion of a soft transformation of ethanol to acetaldehyde and/orethane; information of ammonia.The adsorption of NO in absence and presence of oxygen over different catalystsuch as y-Al2O3, Ag/ Al2O3 and CuO/Al2O3were presented by numerous IR spectroscopic investigationsin an SCR relevant temperature range [2431]. It was found that various adsorbed NOx species such as mono- and bidentate nitrate und nitrite, and nitrosyl-speciesare formed as intermediates on the metal oxide surface. In this context, the reactivity of adsorbed nitrites (NOacis) to form adsorbed nitrate (NO3-)by oxidation is reliant on the reaction temperature, presence of oxygen, hydrogen and/or water [8].
Inseveral studies it was suggested, that the adsorption of NOx from gas phase on the catalyst surface, the catalytic activity for the oxidation of adsorbed NO to NO2 (and NO3-) and the amount of adsorbed NOxmust be related to textural and physico-chemical properties of the catalyst [8, 26, 30, 32]. Moreover, it was postulated that the adsorption and oxidation of NO is a crucial step in the catalytic reduction of NOxto N2[8].In this context, several studies focused onthe NOx adsorption chemistry by IR-spectroscopic investigations of surface adsorbed mono - and bidentate (bridging) nitrites and nitrates formed over Ag- und Cu-containing y-alumina and ZSM-5 catalysts [24, 26, 30, 33, 34]. In contrast, some facts concerning the NO adsorption and oxidation process, which are in detail qualitative and quantitative data of the NOX adsorption capacity, especially in the temperature range of the SCR reactions, are still not clarified.On the best of our knowledge,quantitative data of NOx adsorption capacity are only available for Ag-alumina systems investigated by Bro-sius et al. [8].
The aim of this study was to provide deeper insight into the quantitative adsorptions chemistry by investigation of the NO adsorption processin presence of oxygen and the NO oxidation to NO2 over catalysts based on Ag-, Cu- and Fe oxides on different supports. As supports, microporous materials (zeolite Y and SAPO-11) as well as micro-/mesoporous y-AhO3 and SiO2 were selected. The variety of selected samples was chosen in relation to their textural properties, presence of alumina or silica, acidic propertiesand type of transition metal oxides. Moreover, it must be mentioned that Ag, Cu- und Fe oxide are the most investigated active species for SCR in presence of lower HC or ROH
Experimental
Catalyst preparation. H-Y-zeolite (98.5 %, Abet = 625m2 g-1, Si/Al = 3.5) and SAPO-11 (99.2 <%, Abet = 113 m2 g-1, Si/Al = 0.25; P2O5 = 44.3 %) were purchased from "Sud-Chemie Zeolithes". The support materials y-Al2O3 (99.5 %; Abet = 235 m2 g-1), and SO (99.4%, Abet = 255 m2 g-1) were purchased from Alfa-Aesar Chemical. Prior to use, the samples of corresponding supports were calcined for 3 h at 773 K in dry air flow. Metal oxide containing catalysts were prepared by wet impregnation of the support materials (H-Y, SAPO-11, y-Al2O3and SiO2) with aqueous metal nitrate solution to receive a metal loading of 3.0 wt.-% [2, 35, 36]. The following metal salts were used for impregnation: Cu(NO3)2-3H2O (98.5 %), Fe(NO3)2-9H2O (98.5 %), and AgNO3-(99.0 %) (all from Fluka). The impregnation with metal precursors was performed at 298 K by contacting 10 g support with 10-15 cm3 of an aqueous solution of the corresponding metal salts under constant stirring for 120 min. The impregnated samples were dried at 383 K for 12 h and calcined in air for 4 h at 793 K. Fi-nallyall samples were pressed, sieved and a fraction with a grain sizebetween 150 and 300^m was used in the catalytic experiments. The catalyst composition and notation of the used prepared samples are listed in Table 1.The content of Al, Fe, Cu and Agwas determined by ICP-OES analysis (Optima 8000, Perkin Elmer).
Catalyst characterization.The textural properties were determined from nitrogen sorption isotherms at 77 K, using an ASAP 2010 apparatus (Micromeritics). The samples were degassed at 393 K before measurement for 24 h at 10-5 mbar. Adsorption and desorption isotherms were measured over a range of relative pressures (p/po) from 0 to 1.0. The specific surface areas were determined by applying the BET equation.
Temperature-programmed desorption of ammonia (NH3-TPD) was performed using a flow-type apparatus with a micro-reactor made from quartz glass. Prior to the analysis, the catalyst was pretreated in a He flow (30 cm3 min-1) at 523 K for 1 h. After cooling to 368 K, the catalyst was loaded with ammonia (99.9 %, Air-Liquide) byconsecutive pulses of 1 cm3 ammonia until complete saturation of the surface was reached. The evolved gases were analyzed with a quadru-pole MS spectrometer (Pfeiffer GSD 301). After the removal of physisorbed ammonia by purging with He (50 cm3 min-1) for 60 min, the sample was heated from 368 to 873 K (HR: 10 K min-1) in flowing He (50 cm3 min-1) while analyzing the desorbed NH3 by MS (m/e: 15). The amount of desorbed ammonia was quantified using a reference experiment, pulsing 1 cm3 ammonia over 50 mg of inert quartz and analyzing the amount of NH3 by MS.
Profiles of temperature-programmed reduction (TPR) were recorded on an AMI 100 (Altamira) instrument equipped with a thermal conductivity detector. Before the TPR experiments, the catalysts were pretreated in a flow of dry air (99.9 %)(50 cm3-min-1) at 573 K for 30 min and then cooled to 303 K. For the TPR measurements, a H2/Ar mixture (5 vol.-% H2; 50 cm3 min-1) was passed over the sample while heating from 303 to 1073 K (10 K min-1).
NO adsorption/desorption experiments
Dynamic NOx-TPD experiments in flow of NO and 02.These experiments were carried out at atmospheric pressure in a fixed-bed reactor made from quartz glassinserted into a cylindrical heating furnace. In a typical NOx-TPD experiment 150 mg of sample was pretreated at 623 K in He/O2 flow (6 vol.-% O2) and then cooled down to 383 K. Then the sample was saturatedin a gas containing 940 ppm NO and 6vol.-% O2 in He (flow100 cm3 min-1). During saturation step NO and NO2concentrations in the outlet stream wereon-line measured with chemiluminiscense detector (CLD 70S, Eco Physics). According to the saturation step a temperature-programmed desorption of NOx was carried out with a flow of He in presence of NO (900 ppm), O2 (6 vol.-%) with five temperature programmed steps with correspond ding plateaus at 453, 573 , 673, 743 and 793 K respectively. Every temperature plateau was then held for 20 min to reach steady statere-garding the NOx concentration. These plateaus were reached by heating with constant heating rate of 10 K min-1from 383 to 453,453 to 573, 573 to 673, 673 to 743 and 743 to 793, respectively. The NO and NOx concentrations in the gas stream were measured with CLD.
The conventional NOx-TPD. These experiments were carried out at atmospheric pressure in a fixed-bed quartz-glass reactor inserted into a cylindrical heating furnace. Prior to the se experiments, the sample was pretreated in a He/O2 flow (6 vol.-% O2; 100 cm3 min-1) at 603 K for 1 h. After cooling to 368 K, the sample was saturated with NO or NO2for 60 min. The inlet concentration of NO or NO2for the sample saturation was 900 ppm. After saturation and removal of physisorbed NOx by purging with He, the temperature was increased from 368 to 873 K with a HR: 7 K min-1.The temperature-programmed desorption of NOx was performed in He in absence or presence ofO2 (6 vol.-%). All NOx-TPD experiments were carried out with 200 mg of each sample and a total gas flow rate of 200 cm3 min-1. The NOx concentrations in the effluent gas were also measured by CLD. The total adsorption capacity of NOx(ACnox, ^ mol-g-1) in dynamic and conventional TPD-experiment was evaluated for each temperature step by integrating of the NOx desorption profiles.
Spectroscopic investigation of NO adsorption. The DRIFT spectra of adsorbed NOx species were performed on a spectrometer (Bruker, Vector 22. The samples were prepared by mixing of 100 mg of KBr and 100 mg catalyst priory grinding in micro ball-milling (Vibration Mill, Perkin Elmer). Prior to the adsorption measurement, each sample was activated by heating for 1 h at 703 K in dry air flow (Air-Liquide; 99,9 %). After cooling to 383 K the sample was saturated with NO containing mixture (900 ppm NO ; 200 cm3 min-1) for 1h and then flushed for 1 h in air flow.
Results and discussion
Textural and physico-chemical properties. Table 1 lists the samples prepared in this work, their nomenclature used in this manuscript and corresponding textural and acidic properties. In general, the samples loaded with the metals and metal oxides of Ag, Cu and Fe exhibit a lower specific surface area and average pore diameter with respect to the unloaded support. This decrease is comparable for all three metals and can be attributed to the deposition of metal oxide particles within the pore system of the corresponded support [36]. In relation to the following discussion of the results regarding the adsorption capacity experiments, the NOx adsorption study reveals a visible dependency of NOx adsorption ability in presence of silica. Therefore, according to the nature of the support, i.e. presence of alumina or silica in its structure, the investigated samples were classified as three types: i) alumina supported ii) silica supported iii) zeolite supported. The latter type included catalysts such as zeolite H-Y, SAPO-11 and Fe- and Cu-oxide containing zeolite Y.
The TPD-profiles of desorbed ammonia for some selected catalysts in the temperature region from 373 K to 773 K are depicted in Fig. 1. As shown in Table 1 the acidic site density increases after loading the y-alumina support with the metal.
According to [37] the strength of solid acidic sites can be classified as weak (390-570 K), moderate (570-700 K) and strong (over 700 K). As expected the total acidic site density of pure y-Al2O3 is much higher (104 ^ mol-g-1) than determined for pure SiO2 (6 ^ mol-g-1) [38, 39]. In this regard, y-Al2O3, which was pretreated at medium temperatures up to 773 K, will no longer have any Br0nsted acidic sites. These was proven by IR spectroscopy in presence of pyridine and therefore Br0nsted acidic sites will not be further discussed [38,39]. Doping of y-Al2O3 with Cu and Fe oxides seems to induce new weak and moderate acidic sites at least. The de-convolution of the NH3-TPD profiles for samples loaded on y-AhO3 into weak, moderate and strong Br0nsted and Lewis acidic sites was presented previously [36]. In this study we only used values of the total acidic site density (TASD) and corresponding intrinsic acidity (IASD). TASD is expressed as ^mol of NH3 desorbed per gram of catalyst. However, it was hard to define a clear boundary between a different strength of the present acidic sites of silica supported samples. The detected low values of the total acidic site density for Cu/SiO2 and Fe/SiO2 samples are in good agreement to results from Milushev et al. [40].
The NH3-TPD-profiles of the other catalysts with SP-11 and zeolite Y as support show a maximum at 443 K provided with a broad shoulder to higher temperatures (Fig. 2). This second maximum, which is more pronounced for the SAPO-11 and 3Fe/Y catalysts, indicates the presence of two different types of acidic sites on these materials. It was found that the NOx adsorption capacity depended not only from TASD as well as from the alumina content and/or presence of silica (discussed in next paragraph). The normalization of NH3-TPD data is an useful tool to compare the chemical properties and deduction of the structure-reactivity relationships for heterogeneous catalytic reactions performed over various catalysts, which in turn possess different textural properties [37, 41, 42].
Figure 1. NH3-TPD profiles given as MS-intensity (m/e: 15) as a function of temperature for samples supported over SO2(a), supported over zeolite (b) and supported over y-alumina (c).
a
b
c
Table 1.Specific surface area (SAA), average pore diameter (Dp), total acidic site density (TASD), intrinsic acidic site density (IASD), overall relative acidic site density (ORASD) and intrinsic NOx adsorption capacity (INOAC) at 363 K for Ag, Cu, and Fe containing catalysts and support materials
Abbreviation Sample SSA m2 g-1 Dp nm О CT £ ° ¡4 Q 2 3 1 ORASD a) AI2O3 wt.-% г.Ш |ОШГ| OVONI
Silica supported
SiO SiO2 330 13.5 6 0.021 1.0 < 0.01 0.003
20Cu/ SiO 20 Cu/SiO2 276 10.5 10 0.036 1.4 < 0.01 0.016
3Cu/ SiO 3 Cu/SiO2 315 13.2 22 0.069 3.1 < 0.01 0.007
3FeO/ SiO 3 Fe/SE>2 312 13.0 20 0.064 2.8 < 0.01 0.030
y-Alumina supported
AO у-А^Оз 257 12.4 104 0.410 19.5 > 99 0.214
3Ag/AO 3 Ад/у-АЬОз 232 11.0 181 0.780 37.1 97.1 0.238
3Cu/AO 3 Cu/y-АкОз 233 11.2 191 0.820 39.0 97.2 0.352
3Fe/AO 3 Fe/y-AhO3 229 11.0 200 0.882 42.0 97.2 0.314
Zeolite supported
SP-11 SAPO-11 113 0.6 1250 11.06 526.6 35.6 0.066
3Cu/ SP-11 3 Cu/SAPO-11 64 0.5 1040 16.25 773.8 33.0 0.136
HY H-Y 625 0.5 3400 5.44 259.0 17.3 0.079
3FeY 3 Fe/Y 614 0.4 3160 5.14 245.0 15.9 0.141
a) Overall relative acid site density normalized on SiO2 content: e.g. OKiSD, = IA,:IAa0:
100
'en a) □ 3CU/SP-11
О ЯО . О О ЗСи/АО
ь Д ЗСи/SiO
■и 20CU/Si0
О 60 \ я ЗАд/АО
I
400
500 600 700 Temperature / К a
800
iperatur
b
Figure 2. Adsorption capacity as a function of desorption temperature for Cu- (a) and Fe- containing materials (b) and 3 Ag/AO as reference catalyst.
Therefore, to compare the TASD of investigated samples and to normalized NOx adsorption capacity, the relative overall acidic site density (ROASD) was evaluated by setting the acidic site density of the catalyst in relation to the lowest TASD, i.e. the sample SiO2 (Table 1.) From ROASD values it is obviously that the acidic site density of the SiO2 supported catalysts is in the range of 10 to 20 times lower in comparison to the alumina supported catalysts. Furthermore, the TASD of zeolite supported samples are 200 to 700 times higher compared to the SiO2 supported samples.The loading of alumina with Cu, Fe or Ag oxides leads to a different change of ROASD, whereas the highest acidic site density was found by introducing FeOx for the alumina supported catalysts. As expected the SAPO-11 and 3Cu/SAPO-11 catalysts due to their lower specific surface exhibit the highest ROASD, i.e. about two to three times higher than HY and Fe/Y samples. These can be explained due to the differences of these samples in their specific surface area, TASD and IASD. Generally, it can be assumed that the total acidic density of the investigated samples is a function of the textural properties, the chemical nature and the ratio, density and strength of Lewis-and Br0nsted acidic sites.
The effect of temperature on adsorption of NOx. As suggested by given data from literature, one of the first reaction steps in the SCR in presence of O2 is the oxidation of NO to NO2. NO2 is then adsorbed on the catalyst and thus NOx (x= 2, 3) species are formed [43, 44]. So, it is essential to study the adsorption of NOx over samples with different support and different active metal species, in this context Ag, Cu and Fe oxides. In the temperature range from 453 K to 853 K, where the most NOx were reduced in the HC-SCR process, adsorbed NOx species must be formed, reacted with a reducing agent and/or desorbed completely. Thus, the experiments were carried out below 853 K to investigate the effect of temperature on NOx desorption. In this context, the saturation of the samples with NO in presence of oxygen was carried out at 383 K. Furthermore, the NOx desorption was performed at five consecutive temperature steps in a continuous NO and O2 containing gas.
The total NOx adsorptioncapacities for all investigated samples determined by dynamic NOx-TPD experiments are shown in Fig.2. It was found that at 383 K the alumina supported catalysts exhibit the highest NOx adsorption capacity in a range of 52 to 82 ^mol adsorbed NOx per gram samples.
The adsorption of NOx and formation of adsorbed nitrite/nitrate species on alumina is well known from literature and seems to be depended from the reaction temperature, as well as from nature of the loaded active species[8,40,43-45]. This is specifically relevant with regard to Fig. 2.Contrary, the alumina free samplessuch as 3Cu/SiO, 20Cu/SiO and 3Fe/SiO possess a relatively small NOx adsorption capaci-ty,i.e. lower than 10 ^mol adsorbed NOx per gram catalyst. As shownin Table 1 and Fig.2,the pure silica exhibitsavery low NOx uptake of 0.03 ^mol g-1. These result is in aagreement with [44], which results showed that adsorbed NOx is weak bonded on the silica surface and can be removed by evacuation even at room temperature. Therefore, according to [44 46] for Cu- and Fe-containing samples supported on silica we tentatively assumed, that NOx adsorption takes place mostly on Cu- and Fe-oxide sites. However, the possibility of NOx adsorption and/or NO oxidation by participation of the surface O-Si-O-Cu=O species cannot be totally ruled out.
In comparison to purey-Al2O3alow NOx adsorption capacity of 0.066 ^mol g-1was found for SAPO-11. In this context, it is important that these samples have extremely different textural and acidic properties (Table 1).The
3Fe/Y catalyst, having the highest overall acidic site density and BET-surface area, shows a NOx adsorption capacity of 49 ^molg-1, which are close to the alumina supported catalysts. It is important to mention, that the acidic site density of 3Fe/Y is 4 to 12 times and the BET-surface area 3 times higherthan that of the alumina supported catalysts. This indi-
cates that the adsorption of NOx is contributed to the chemical properties of the catalyst, particularly to apossible formation of nitrates.
Moreover, to compare the ability for NOx adsorption over samples with different textural properties, the total NOx adsorption capacity was normalised to 1 m2 of the BET-surface areaas intrinsic overall adsorption capacity of NOx (INO-AC)as summarised given in Table 1. The INOAC values were used to compare the relationship between NOx adsorption ability and physico-chemical properties of the investigated samples. As shown in Fig. 2, Fig. 3a and Table 1 for the alumina, silicaand zeolite supported catalystsboth the support and the contained metal oxide phase affect the adsorption properties. Thus, the adsorbed amount of NOx (pmol-g-1) increases in the order 3Cu/AO (82) > 3Fe/AO (72) > 3Ag/AO (52). An increase of metal content, as investigated for 3 and 20 wt.-% CuO containing silica also affects the amount of adsorbed NOX. This also indicates that an adsorption of NO (in form of NO2-) and/or NO2 (in form of NO3-) can occur on support (alumina, silica, zeolite) as well as on active oxide phase of Cu, Fe and Ag. Thus it can be ruled out that the NOx adsorption capacity is a function of the alumina content for all investigated catalysts.
may need the assistance of the Br0nsted acidic sites.As mentioned above, during SCR process in presence of oxygen excess, adsorbed NO species can react with active lattice oxygen. As a consequence the oxidation of NO to NO2 takes place and thus adsorbed NO2 and/or NO3- is formed [43]. Depending from the reaction temperature NO2 can also be de-sorbed from the catalyst surface to the gas phase. According to the fact, that the adsorption of NOx species can not only take place on the active phase but also the oxidation of NO to NO2 can be affected by redox active species, i.e. loaded transition metal oxides[8, 26, 49, 50]. TheDFT studies show, that the rate-determining step for the whole SCR process is identified as the oxidation of NO to NO2 [51]. Therefore, a crucial step to form adsorbed surface nitrates is the oxidation of NO to NO2 in a further adsorption/desorption process. In order to obtain information about activity of metal oxide towards oxidation of NO to NO2the data obtained by CLD analysis of desorbed NOx in the gas-phase (i.e. concentration of NOx=-NO + NO2and NO) were used for calculation of selectivity towards NO2 formation during dynamic NOx-TPD experiments. The NO2 selectivity in a temperature range from 383 K to 803 Kis depicted in Fig.4. At temperatures between 383 K and 423 K, the desorption of NO2 was detected with a low selectivity of ca. 2-4 %. These indicated, that the oxidation of NO to NO2 proceeds with low rate and/or formed NO2 is strong bonded on the catalyst surface due to formation of adsorbed nitrate [8,26,40].
Figure 3. Adsorption capacity of NOx (NO+NO2) at 383 K (a) and selectivity of NO2 formation at 673 K (b) determined by dynamic NOx-TPD in a NO and
Catalytic activity of transition metal oxide towards oxidation of NO to NO2
From results of Odenbrand et al. [47] and Halasz et al. [48] it can be suggested that the oxidation of NO to NO2
b
Figure 4.Selectivity of NO2 formation during dynamic NOx-TPD experiment as a function of temperature for Cu- (a) and Fe- (b) containing samples and 3Ag/AO as reference catalyst. Date of NO-NO2 equilibrium in presence of oxygen were retrieved from [52].
As shown in Fig. 4 the selectivity of NO2 formation reached a maximum in a range of 10 % to 15 % over all investigated catalysts in a temperature range of 723 K to 773 K. However, over the samples 20Cu/SiO, 3Cu/AO and 3Fe/ SiO a higher NO2-selectivity in the range of 20 % to 22 % at the same temperatures were determined. An increase from 3 to 20 wt.-% of copper loaded on silica leads to an increase of formed NO2. Different NO2-selectivities were found for Fe and Cu loaded catalysts with the same support. This illustrates that the support does not play an important role in the oxidation of NO to NO2.The 3CuO/AO and 3FeO/AO catalysts show the highest activity for the NO oxidation, whereas, the 3 Ag/AO catalyst show half of the NO2-yield. This can be explained by different redox-properties of these transition metal oxides[35, 53]. By our previous EtOH/SCR study it was shown in H2-TPR-experiments that hydrogen consumption for 3Ag/AO sample was 130pmol/g and the reduction take place at 430 - 500 K[36]. It must be mentioned that ^-consumption determined by TPR according to[53, 54]can be considered as an equivalent to the content of the active lattice oxygen of transition metal oxide. Contrary, the reduction of 3FeO/AO and 3CuO/AO catalysts proceed with a higher ^-consumption of 362 and 180pmol g-1 at comparatively higher temperatures of 500 K to 580 K [36]. This indicated that 3 wt.-% of Fe or Cu- oxides supported over alumina possess in comparison to 3Ag/AO about 1.4 or 2,8 time higher content of active lattice oxygen respectively.It must be mentioned that the catalytic oxidation of NO to NO2is limited by the thermodynamically equilibrium (eq. 1).
NO + 0.5 O2-NO2 ДН°298 = - 113 kJ mol-1
(1)
M-O + NO « MO-NOads
(2)
The adsorbed NO can undergo a catalytic oxidation with participation of active lattice oxygen([O*]) or by spill-over oxygen from alumina support to form NO2.The formed NO2 can react similar to NO with oxide-sites from the metal- or support-surface to form surface nitratesas given by eq. 3 and 4.
NOads+ [O*] « NO2 ads
M-O + NO2 ads
M-O-NO2 ads
(3)
(4)
As reported in [26, 40, 43] the adsorbed nitrites and nitrates can be present in three different states, i.e. uniden-tate, bidentate and/or bridging species. Moreover, these species have different binding energies leading to different desorption temperature of NOx. Therefore, it can be expected that various desorbed NOx-species are formed and must be presented at several temperature regions. However, desorption of various adsorbed surface NOx species cannot be proven by dynamic NOx-TPD tests. In this regard a conventional NOx-TPD experiment according to [8, 43] after saturating at 373 K of sample surface in a NO+O2 flow and in a NO2 flow without oxygen was carried out on 3Ag/Al2O3 catalystas shown in Fig. 5 respectively.The TPD profiles of NOx after saturation with NO + O2 or NO2 show different maximum peaks indicating the presence of various forms of adsorbed NOx
species. After saturation with a NO+O2 containing gas, two temperature ranges for desorption of NO and NO2 were detected: i) at low temperature (LT) range two desorption maxima are located at 434 K and 485 K, ii) at high temperature (HT) range two desorption maxima are located at 670 K and 725 K.
At temperatures below 323 K NO can further react with oxygen to give NO2. However, the equilibrium fully shifts to the NO2 formation in oxygen-depleted air at temperatures of 500 K or below [52].From data presented in Fig. 4 it is obviously, that NO could be fully converted to NO2 without catalyst already at low temperatures (< 323 K). Therefore, weargue that the observed conversion of NO to NO2 by employed conditions proceeds predominantly via catalytic reaction.
NO temperature programmed desorption in presence of oxygen (conventional method). As known from [8, 26, 40], the adsorption of NO in presence of excess O2 over supported transition metal oxides on different supports can take place by interaction of NO with metal oxide (eq. 2).
Figure 5. TPD profiles of NO and NO2 after saturation of the 3Ag/AO catalyst with NO and O2 (a) or NO2 (b) 373 K. (Saturation: 6 vol.-% O2; 900 ppm NO (a), 900 ppm NO2 (b).
By deconvolution of these NOx desorption profiles for both experiments the presence of two overlapped maxima at high (668/685 K and 715/724 K) and LT (434 K and 453 K) were identified. The adsorption capacity of NO and NO2 at both temperature regions and the selectivity of NO2 formation over the 3Ag/AO sample are given Table 2. It has been evident that the total NOx adsorption capacity after saturation with NO2 are about six times higher as after saturation in a NO and O2containing flow. This indicated that the ability of NO2 for adsorption on 3Ag/AO in comparison to NO is sufficiently higher. The values of NO2-selectivity indicated that over 3Ag/ AO in case of an saturation with a NO and O2 containing gas 11.6 % of NO at LT and 47.8 % at HT region was oxidized to NO2. The NO2-selectivity for sample saturation with NO2 indicated that at LT only 4,2 % was desorbed as very weak bonded NO2 species. Moreover, about 90 % of all adsorbed NO2 was desorbed at HT above 600 K and only 5 % were decomposed and eluted as NO at these temperature.
Table 2. Adsorptions capacity of NOx over 3Ag/AO catalyst and selectivity to NO2 formation during NOx-TPD experiments (SN02) after saturation with NO or NO2 in presence of oxygen.
Temperature NOx adsorption capacity (|imol g-1)
of NOx desorption after saturation with NO after saturation with NO2
NO NO2 SNO2 NO NO2 SnO2
LT (<570 K) 30.1 20.6 11.6 < 1 27 (463 K) 4.2
HT (> 570 K) 12.6 49.5 47.8 10 (723 K) 186 (623 K) 430 (723 K)
LT + HT 42.6 62.2 59.2 10 643 95.8
The NOx desorption profile for low and high temperature range show the presence of two types of adsorbed species in each case (Fig. 6). This can be attributed to the fact that the desorbed nitrites (NOads or NO2")and nitrates (NO2 or NO3") species can be present in different forms [35]. According to [24, 26, 35, 40] the LT desorption in both cases represents the evolution of monodentate adsorbed NOx species [24, 26, 35, 40]. The desorption of NO or NO2at HT can be explained by the desorptionofeither bidentate or bridged adsorbed nitrites or nitrates. Comparing the amount of desorbed NOx at HT desorption for the two different saturation methods, it is clear that for the saturation with NO2 more adsorbed nitrate species are formed. This can be explained by differences in conjugated acidic-basic properties of the molecules NO and NO2[37].
1612 1576 1554 1381 1302
N, 1 J /■ у 1 1 1 1 1 1 1 \ 3 Fe/Al205 ■
"oi с 'o> -Q _ 0" z
NO, monodentate
^ \ J 4^3 CU/AI2O3 -
3 Ag/Al2cy \ N02 bidentate
Y-AI2O3 y
1800 1700 1600 1500 1400 1300 1200 1100 1000
Wavennumber / cm"1
Figure 6. DRIFT spectra of y-AhOs, 3Ag/AO, 3Cu/AO and 3Fe/AO samples at 673 K after saturation with 900 ppm NO in presence of O2 at 363 (a). (Purge time 40 min at 673 K with N2+O2 flow(6 vol.-% O2; 200 cm3 min-1).
Comparison of adsorptions band at 1381 cm-1(b).
DRIFT spectroscopic study of NOx adsorption in presence of oxygen. According to Fig. 5at the middle temperature region (520-700 K) a low amount of NOx were desorbed. Thus the adsorbed NOx species are strong bonded to the alumina support up to these temperatures. Moreover the main reactions of both the EtOH-SCR and HC-SCR take place with high efficiency over Ag, Fe and Cu containing alumina catalysts at temperatures of 520 K to 700 K [36, 55]. Therefore the DRIFT spectra of alumina support and three representative samples loaded with Ag, Cu and Fe oxides exposed to a gas mixture of NO and O2 at 673 K were recorded in order to investigate the chemical structure of adsorbed NOx-species.A comparison of DRIFT spectra, as given in Fig. 6 show that the exposure of alumina supported samples to a NO and O2 containing gas gives various adsorption bands in the range of 1700-1200 cm-1.By analysis of these data five
adsorptions bands of quite high intensity with corresponding maxima at 1615, 1576, 1554, 1381 and 1302 cm-1 can be determined.
As known from [24, 26, 31, 40, 56-59] these data are most pronounced according to the IR-spectral identification of surface NOx species in concern to nitro, nitrito and nitrato compounds. The characteristic bands identifying these NOx compounds are overlapped and appear in the region 17001000 cm-1. Thus a spectral differentiation between these compounds is difficult. Adsorption of NO in presence of O2 excess on oxide surfaces lead to the formation of nitro and nitrito complexes as well as surface nitrates coordinated to the cat-ionic sites on the surface [24, 26, 40].
IR investigation of NO adsorption over alumina and zeolite catalysts show that with increasing temperature, nitrates are mainly formed [24, 26, 40]. Therefore, the adsorption of NO in presence of O2 at temperature above 600 K might be expected to result in formation of surface nitrates. DRIFT spectra in region of 1700 - 1000 cm-1 for y-alumina and Fe, Cu and Ag alumina supported samples at 673 K show a similar adsorption maxima (1615, 1576, 1554, 1381 and 1302 cm-1) indicating the presence of similar adsorbed nitro-nitrito species.
A carefulanalysis of the absorption band at 1381cm-1,typicalfor adsorbed surface nitrate species, over the 3CuO/AO and 3Ag/AO samples result in a higher amount of these species as found for the3FeO/AO and y alumina samples Fig. 6b. These results are in good agreement with the values of NOx adsorption capacity observed by dynamic NOx-TPD experiment and selectivity towards NO2 formation for Cu, Fe and Ag oxide loaded on alumina (Fig. 5,6). In this regard it was fo und that 3CuO/AO and Ag/AO samples possess higher NOx adsorptions capacity and higher selectivity to NO2 as 3FeO/AO and y alumina, respectively.
Relationship between observed NOx adsorption and chemical properties.
NO2 is a "radical molecule", with a high electron affinity (2.3 eV). Therefore on solid surface it can react to the metal centers such as Al-, Ti- and other transition metals through O, N or the combination of both [60, 61]. On the other site the NO molecule has an unpaired electron in its 2n* anti bonding molecular orbital and can exhibit complex adsorption reactions on transition metal oxide surfaces [61].Adsorption and interaction of chemical non-inert NO and NO2 molecules over metal oxide support surfaces such as alumina or zeolite, is useful to consider these according to so-called HSAB "hard and soft (Lewis) acids and bases" concept[54, 62]. According to HSAB concept, both gaseous NO and NO2 can be contemplate as a soft base, whereas adsorbed NO2- and NO3- species presented a more hard base. On the other hand the solid surface of alumina, silica or zeolites, depending from the environmental and pre-treatment conditions (calcination temperature, presence of water, i.e. dehydration of the surface) contained Lewis acidic sites, Br0nsted acidic sites or a mixture of both. It is obviously, that at constant temperature the ability to the adsorption/desorption of NOx of a solid surface depended of its textural and physico-chemical properties. Moreover, the acidic and basic properties of an alumina surface will be altered by loading with deferent transition metal oxides[33, 37].
Numerous IR studies show that surface NOx adsorption chemistry is quite complex and well investigated [24, 26, 40, 43, 63]. However, the quantitative NOx adsorption data is rather limited. To obtain a deeper insight into observed differences of adsorptions capacity experimental data, i.e. intrinsic NOx capacity were involve in two "simple" interdependence of alumina content and intrinsic acidity. Fig. 7a shows the total intrinsic NOx adsorption capacity at 373 K as a function of the intrinsic acidity as well as the alumina content for all samples presented in Table 1.On the one hand, the nature of the NOx interaction with the solid surface was investigated due to the comparison of the intrinsic NOx adsorption capacity given in Table 1 and the intrinsic acidity of the investigated catalysts
as shown in 7b. It clear that the SiO2 supported catalysts pos-sessquite low intrinsic acidity (0.03 - 0.06 pmol-g-1) and total intrinsic NOx-adsorption capacity (0.02 - 0.03 pmol-m-2). The low intrinsic NOx-adsorption capacities of silica supported Cu-and Ag-catalysts are in an agreement with the results of Bion et al. [7], which show that silver supported silica exhibits an extremely low SCR-activity in presence of hydrocarbons and ethanol.
о E
rx
0,4
Я 0,3
■fi 0,2
0,1
o* 0,0
1 1 ■ t>) 1 1 1 Alumina i i 3Cu/AI,0
- supported ; r3Fe/AI2p 3Ag/AI С
3 Fe/Y 3 Cu/SAPO^-T Al0-, "
Silica 0
""supported^ ...„ HY О SAPO-11
0 20 40 60 80 100 120 Content of Al203/ wt. -%
b
Figure 7. Dependency of the NOX intrinsic adsorptions capacity from the intrinsic acidity (a) and the alumina content (b) of alumina, silica and zeolite supported samples.
The alumina and zeolite supported catalysts show a comparatively higher intrinsic acidity and total intrinsic NOx-adsorption capacity. From the one hand, an increasing intrinsic acidity (compare ACNOx data of SP-11 and 3 Cu/SP-11) inhibit the NOx adsorption capacity. On the other hand, the loading of the alumina containing supports with transition metal oxides promotes the NOx-adsorption. This indicates that both support and loaded transition metal oxide contribute and promote the adsorption of NO und thus the formation of surface NOx-species. This is related to the fact that the interaction of acidic NOx and high acidic surface contained dehy-droxylated Al-OH (zeolite Y) and POx-OH (SAPO-11) groups are less pronounced. These results are in a good agreement with Brosius et al. [8]. In this context it was shown that alumina and Ag/Al2O3 samples in presence of 12 vol.-% water did not adsorb anyNOx at 423 K. On the other hand, these results support the suggestion from literature that the low temperature poisoning of SCR catalysts is due to nitrite and nitrate formation and eventually blocking of active sites on Ag- and Cu-containing catalysts [24, 32]. It must be mentioned that both values of IA and IOAC were determined in TPD experiments by previous saturation at 373 K with NO or ammonia respectively. Therefore both experimental data represented
the same thermodynamic adsorption range [54].
The first information of the both graphs in Fig. 7a,b gives the view of large discrepancy between the arbitrary parabolic line on the left and the straight line on the right as well as the experimental values of NOx adsorption capacity. Obviously these differences of the in this study gained NOx experimental data indicated the complex dependency of the adsorption chemistry and its relation to the physico-chemical properties, e.g. acidic site strength and density and/or total alumina content. According to [37]there is no direct dependency between the acidic properties and the alumina content for alumina and zeolite supported samples. Thus, from the base of these data, and to get a deeper insight in the relationship between adsorption capacity of NOx and the physico-chemical properties of the samples, it is necessary to obtain on supplementary information by DFT calculation with consideration of several energetic parameters. These are the activation energy of NOx adsorption and desorption and/or the energy of NOx-metal oxide bond [51, 64, 65]. Furthermore data of kinetic parameters for catalytic oxidation of NO to NO2 and formass transfer phenomenon of NOx adsorption and desorption are required [66].
Moreover, our results regarding the difference in NOx adsorptions capacity are in line with studies from Oh et al.[67] obtained for the SCR with propene over Pt containing alumina and silica catalysts. Briefly, it was found that NOx adsorption did not occur on the SiO2 surface so the reaction between propene and NO2 could be isolated, i.e. no nitrate effect would complicate the reactions kinetic analysis and their significance could be decoupled. The dependency of the data about NOx adsorption, Al content and acidic site strength and density of the investigated samples suggested that the ability of these to adsorb NO clearly associated to the quantity of acidic site density as well as the presence of alumina in the sample. According to the specific role of the chemical properties of the support material of catalysts for the SCR and thus the NOx adsorption, alumina has attracted attention due to its high catalytic performance. Focused on the NOx adsorption on Al2O3 and transition metal oxide surfaces, the adsorption of NO and NO2 is facilitated significantly due to the strong syn-ergetic effect between the metal oxide cluster and Al2O3 substrate [64, 68].
Conclusions
Quantitative data of the NOx adsorption capacity is only available for the state of the art catalyst for the HC/SCR of NOx, i.e. silver supported on alumina. In this work the NOx adsorption capacity (uptake) was investigated on a broad spectrum of materials and related to their physico-chemical properties. It was found that the alumina supported catalysts exhibit the highest NOx adsorption capacity. An increase of metal content, as investigated for 3 and 20 wt.-% CuO containing silica also affects the amount of adsorbed NOx. Thus an adsorption of NO and NO2 can occur on both support (alumina, silica, zeolite) as well as on active oxide phase of Cu, Fe and Ag. Alumina supported samples show about ten times higher NOx-adsorption capacities than silica supported samples. The NO adsorption ability and/or capacity increases by loading with oxides of Fe or Cu as active phase instead of Ag supported on y-Al2O3.
The oxidation of NO to NO2 took predominantly place on the active transition metal species and is directly related to their redox property. In this regard the amount of lattice oxygen, determined by H2-TPR experiments in a previous study, is directly linked to the yield of NO2. Both the ability to adsorb NOx and the activity to oxidize NO to NOx is furthermore not only depended from the total acidic site density but also from the content of alumina or silica.The formation of nitrates was investigated for pure y-alumina, whereby it can be distinguish between different surface NOx-species. These NOx species were determined by a temperature programmed desorption after saturating with NO and O2 or NO2 in absence of O2. For both experiments a formation of nitrates and nitrites could be ob-
a
served by desorption of NOx for two temperature regions. Also by DRIFTS experiments the formation and stability of different surface NOx species was identified. In this regard the reaction of NO and NO2 with lattice oxygen from the y-alumina support could be observed and the formation of nitrates is about 30 times higher for NO2 than NO using pure y-Al2O3.This paper defines anoutlet dates for further studies focusing a broader analysis of the NOx adsorption on metal carrier catalysts. From these studies itshould be possible to identify the relation of the NOx adsorption to the acidic site strength and density as well as to both the content of alumina and active metal species. Suggesting that the NO2 formation by NO oxidation and the subsequently adsorption as nitrtites and/or nitrates as a key reaction path in the OHC/HC-SCR the catalytic activity of possible catalysts can also be improved in a targeted manner.
Acknowledgements.
Financial support of this work from the Deutsche Forschungs gemeinschaft (DFG) (Pr GL 290/8-1) is gratefully acknowledged. K. S. thank the EU- Lifelong Learning "ERAS-MUS+" Programme for financial support their research practi-cum for the master diploma thesis.
References
1. Janas J., Rojek W., Dzwigaj S. The influence of C1 and C2 organic reducing agents on catalytic properties of Co(II)-single site BEA zeolite in SCR of NO // Catal. Today. 2012. V. 191. P. 32-37.
2. Worch D., Suprun W., Gläser R. Supported transition metal-oxide catalysts for HC-SCR DeNOx with propene // Catal. Today. 2011. V. 176. P. 309-313.
3. Myronyuk T.V., Orlyk S.N. Role of redox and acidic properties of CoO/ZrO2(SO4) catalysts in CH4-SCR of NO // Catal. Today. 2007. V. 119. P. 152-155.
4. Garcia-Cortes J.M., Perez-Ramirez J., Illan-Go-mez M.J, [et al.]. Effect of the support in De-NOx HC-SCR over transition metal catalysts / /React. Kinet. Catal. Lett. 2000. V. 70. P.199-206.
5. Saaid I.M., Mohamed A.R., Bhatia S. Comparative study of Cu-ZSM-5 and Fe-ZSM-5 in the SCR of NOx with i-C4H10 // React. Kinet. Catal. Lett. 2002. V. 75. P. 359-365.
6. He H., Zhang X.L, Wu Q., [et al.]. Review of Ag/ Al2O3-Reductant system in the selective catalytic reduction of NOx // Catal. Surv. Asia. 2008. V. 12. P. 38-55.
7. Bion N., Saussey J., Haneda M., [et al.]. Study by in situ FTIR spectroscopy of the SCR of NO, by ethanol on Ag/Al2O3 - Evidence of the role of isocyanate species // J. Catal. 2003. V.217. P.47-58.
8. Brosius R., Arve K., Groothaert M, [et al.]. Adsorption chemistry of NOx on Ag/Al2O3 catalyst for selective catalytic reduction of NOx using hydrocarbons // J. Catal. 2005. V. 231. P. 344-353.
9. Boutros M., Trichard J.M., Da Costa P. Silver supported mesoporous SBA-15 as potential catalysts for SCR/ NOx by ethanol // Appl. Catal. B. 2009. V. 91. P. 640-648.
10. Dzwigaj S., Janas J., Mizera, [et al.]. Incorporation of Copper in SiBEA Zeolite as Isolated Lattice Mononuclear Cu(II) Species and its Role in Selective Catalytic Reduction of NO by Ethanol // Catal. Lett. 2008. V. 126. P. 36-42.
11. Janas J., Gurgul J., Socha R.P.[,et al.] Effect of Cu content on the catalytic activity of CuSiBEA zeolite in the SCR of NO by ethanol. Nature of the copper species // Appl. Catal. B. 2009. V.91. P. 217-224.
12. Dzwigaj S., Janas J., Machej T., [et al.]. Selective catalytic reduction of NO by alcohols on Co- and Fe-Si beta catalysts // Catal. Today. 2007. V. 119. P. 133-136.
13. Janas J., Dzwigaj S. Physico-chemical properties of Fe-AlBEA and Fe-SiBEA zeolites and their catalytic activity in the SCR of NO with ethanol or methane // Catal. Today. 2011. V. 176. P. 272-276.
14. Yeom Y.H., Li M.J., Sachtler W.M.H., [et al.]. Low-temperature NOx reduction with ethanol over Ag/Y. A
comparison with Ag/gamma-Al2O3 and BaNa/Y // J. Catal. 2007. V. 246. P. 413-427.
15. Granger P., Parvulescu V.I Catalytic NOx Abatement Systems for Mobile Sources. From Three-Way to Lean Burn after-Treatment Technologies // Chem. Rev. 2011. V. 111. P. 3155-3207.
16. Djega-Mariadassou G. From three-way to DeNOx catalysis. a general model // Catal. Today. 2004. V. 90. P. 27-34.
17. Johnson W.L., Fisher G.B., Toops T.J. Mechanistic investigation of ethanol SCR of NOx over Ag/Al2O3 // Catal. Today. 2012. V. 184. P. 166-177.
18. Kim M.K., Kim P.S., Baik J.H., [et al.]. DeNOx performance of Ag/Al2O3 catalyst using simulated diesel fu-el-ethanol mixture as reductant // Appl. Catal. B. 2011. V. 105. P. 1-14.
19. Flura A., Courtois X., Can F., et al. A Study of the NOx selective catalytic reduction with ethanol and its by-products // Top. Catal. 2013. V. 56. P. 94-103.
20. Can F., Flura A., Courtois X., Royer S., [et al.]. Role of the alumina surface properties on the ammonia production during the NOx SCR with ethanol over Ag/Al2O3 catalysts // Catal. Today. 2011. V. 164. P. 474-479.
21. Wallin M., Karlsson C.-J., A. Palmqvist, et al.. Selective catalytic reduction of NOx over H-ZSM-5 under lean conditions using transient NH3 supply // Top. Catal. 2004. V. 31. P. 107-113.
22. Park J.W., Potwin C., Diega-Maridassou G. De-NOx reduction by methanol over Co/alumina // Top. Catal. 2007. V. 42-43. P. 259-262.
23. Kim M.K., Kim P.S., Cho B.K., [et al.]. Enhanced NOx reduction and byproduct removal by HC // Catal. Today. 2012. V. 184. P. 95-106.
24. Mosqueda-Jimenez B.I., Jentys A.,[et al.]. On the surface reactions during NO reduction with propene and propane on Ni-exchanged mordenite // Appl. Catal. B. 2003. V. 46. P. 189-202.
25. Anderson J.A., Marquez-Alvarez C., Lopez-Mu-noz M.J.,et al. Reduction of NOx in C3H6/ air mixtures over Cu/ Al203 catalysts // Appl. Catal. B. 1997. V. 14. P. 189-202.
26. Hadjiivanov K., Klissurski D., Ramisb G., [et al.]. Fourier transform IR study of NO, adsorption on aCuZSM-5 DeNO, catalyst // Appl. Catal. B. 1996. V. 7. P. 251-267.
27. Beutel T., Adelman B.J., Sachtler W.M.H.FTIR study of the nitrogen isotopic exchangebetween adsorbed 15N0, complexes and 14N0 overCu/ZSM-5 and Co/ZSM-5 // Appl. Catal. B. 1996. V. 9. P. 110
28. Sultana A., Nanba T., Haneda M., [et al.]. Influence of co-cations on the formation of Cu+ species in Cu/ ZSM-5 and its effect on selective catalytic reduction of NOx with NH3 // Appl. Catal. B. 2010. V. 101. P. 61-67.
29. Say Z., Vovk E.I., Bukhtiyarov V.I., [et al.]. Influence of ceria on the NOxreduction performance of NOxstor-agereduction catalysts // Appl. Catal. B. 2013. V. 142-143. P. 89-100.
30. Hodjati S., Petit C., Pitchon V., Kienneman A.Absorption/desorption of NOx process on perovskites: Nature and stability of the species formed on BaSnO // Appl. Catal. B. 2000. V. 27. P. 117.
31. Yamashita T., Vannice A. Temperature-programmed desorption of NO adsorbed on Mn2Û3 and Mn3Û4 // Appl. Catal. B. 1997. V. 13. P. 141-155.
32. Mosqueda-Jimenez B.I., Jentys A., Lercher J.A Reduction of nitric oxide by propene and propane on Ni-ex-changed mordenite // Appl. Catal. B. 2003. V. 43. P. 105-115.
33. Venkov T., Hadjiivanov K., Klissurski D. Spectroscopy study of NO adsorption and NO + O2 co-adsorption on Al2O3 // Phys. Chem. Chem. Phys. 2002. V. 4. P. 24432448.
34. Hadjiivanov K., Knozinger H. FTIR study of CO and NO adsorption and coadsorption on a Cu/SiO2 catalyst. Probing the oxidation state of copper // Phys. Chem. Chem. Phys. 2001. V. 3. P. 1132-1137.
35. Yu Y.B., Zhang X.L He., H Evidence for the formation, isomerization and decomposition of organo-nitrite and -nitro species during the NOx reduction by C3H6 on Ag/AhO3 // Appl. Catal. B. 2007. V. 75. P. 298-302.
36. Worch D., Suprun W, Glaeser R. Fe- and Cu-oxides supported on y-Al2O3 as catalysts for the SCR of NO with ethanol // Chem. Paper. 2014. P. 1228-1239.
37. Tanabe K., Misono M., Ono Y, New Solid Acid and bases // Stud. Surf. Sci. Catal. 1989. P. 27-213.
38. Daniel W, Schubert U., Gloeckner R., [et al.]. Enhanced surface acidity in mixed alumina-silicas: alow-tem-peratureFTIR study // Appl. Catal. A. 2000. V. 196. P. 247-260.
39. Morterra C., Magnacca G. Surface characterization of modified aluminas. Part 5.—Surface acidity and basicity of CeO2-Al2O3 systems // J. Chem. Soc. Faraday Trans. 1996. V. 92. P. 5111-5116.
40. Morterra C.,Magnacca G. Surface characterization of modified aluminas. Part 4. Surface hydration and Lewis acidity of CeO2-Al2O3 systems // J. Chem. Soc. Faraday Trans. 1996. V. 92. P. 1991-1999.
41. MilushevA., HadjiivanovK.FTIR study of CO and NOx adsorption and co-adsorption on Cu/silicalite-1 // Phys. Chem. Chem. Phys. 2001. V. 3. P. 5337-5341.
42. Christodoulakis A., Machli N., Lemonidou A.A., et al.Molecular structure and reactivity of vanadia-based catalysts for propane oxidative dehydrogenation studied by in situ Raman spectroscopy and catalytic activity measurements // J. Catal. 2004. V. 222. P. 293-306.
43. Wachs I.E. Recent conceptual advances in the catalysis science of mixed metal oxide catalytic materials // Catal. Today. 2005. V. 100. P. 79-94.
44. Zhang J., Liu Y.,. Fan W, He Y., Selective synthesis of mixed alcohols from syngas over catalyst Fe2O3/AhO3 in slurry reactor // Fuel Proces. Techn. 2000, V. 91 P. 379-382.
45. Hadjiivanov K.I. Identification of neutral and charged NxOysurface species by IR spectroscopy // Catal. Rev. Sci. Eng. 2000. V. 42. P. 71-144.
46. Mosqueda-Jimenez B.I., Lahougue A., Bazin P., [et al.]. Operando systems for the evaluation of the catalytic performance of NOx storage and reduction materials // Catal. Today. 2007. V. 119. P. 73-77.
47. Balkenende A.R., Vandergrift C.J.G., Meulen-kamp E.A. [et al]. Characterization of the surface of a Cu/SiO, catalyst exposedto NO and CO using IR spectroscopy // Appl. Surf. Sci. 1993. V. 68. P. 161-171.
48. Odenbrand C.U.I., Andersson L.A.H, Brandin J.G.M., [et al.]. Dealuminatedmordenites as catalyst in the oxidation and decomposition of nitric oxide and in the decomposition of nitrogen dioxide: characterization and activities // Catal. Today. 1989. V. 4. P. 155-172.
49. Halasz I., Brenner A., Simon K.YActive sites of H-ZSM5 catalysts for the oxidation of nitric oxide by oxygen // Catal. Lett. 1995. V. 34. P. 151-161.
50. De Mello L.F., Baldanza M.A.S., Noronha F.B., Schmal M. NO reduction with ethanol on MoO3/Al2O3 and CeO2-ZrO2-supported Pd catalysts // Catal. Today. 2003. V. 85. P. 3-12.
51. Ramis G., Larribis M.A. An FT-IR study of the adsorption and oxidation of N-containing compounds over Fe2O3/
AI2O3 SCR catalysts // J. Mol. Catal. A. 2004. V. 215. P. 161-167.
52. Li J., Li S. New insight into selective catalytic reduction of nitrogen oxides by ammonia over H-form zeolites: a theoretical study // Phys. Chem. Chem. Phys. 2007. V. 9. P. 3304-3311.
53. Busca G,.Larrubia M.A,.Arrighi L, Ramis, G. Catalytic abatement of NOx: Chemical and mechanistic aspects // Catal. Today. 2005. V. 107-108. P. 139-148.
54. Hurst N.W., Gentry S.J, Jones. A. Temperature Programmed Reduction // Catal. Rev. Sci. Eng. 1982. V. 24. P. 233-309.
55. Deutschmann O., Knözinger H., Kochloefl K., Turek T. Ullmann's Encyclopedia of Industrial Chemistry, Wi-ley-VCH Verlag GmbH & Co. KGaA. 2000. P. 89-175.
56. Johnson T.V. Review of diesel emissions and control // Inter. J. Eng. Res. 2009. V. 10. P. 275-285.
57. Yeom Y.H., Li M., Sachtler W.M.H., [et al.]. A study of the mechanism for NOx reduction with ethanol on y -alumina supported silver // J. Catal. 2006. V. 238. P. 100-110.
58. Centi G., Perathoner S. Nature of active species in copper-based catalysts and their chemistry of transformation of nitrogen oxides // Appl. Catal. A. 1995. V. 132. P. 179295.
59. Richter M., Bentrup U., Eckelt R., Schneider, [et. al.]. The effect of hydrogen on the selective catalytic reduction of NO in excess oxygen over Ag/AlÄ // Appl. Catal. B. 2007. V. 51. P. 261-274.
60. Padley M., Rochester C.H., Gutchings G., [et al.]. FTIR study of the effects of sulfur poisons on NO adsorption on supported copper oxide catalysts // J. Chem. Soc. Faraday Trans. 1995. V. 91. P. 141-144.
61. Garin F Mechanism of NOx decomposition // Appl. Catal. A. 2001. V. 222. P. 183-219.
62. Sivachandiran L., Thevenet F., Gravejat P et al.Dissociative Adsorption of NO on TiO2 (110) Surface //Appl. Catal. B. 2013. V. 142. P. 196-204.
63. H. Bögel, S. Tobisch, T. Nowak DFT Investigations of the Structure and Bonding Between Transition Metals and Olefins" // Int. J. Quantum Chem. 1998. V. 69. P. 387396.
64. Stern K.H. High temperature properties and decomposition of inorganic salts // J. Phys. Chem. Ref. Data. 1972. V. 1. P. 747-772.
65. Liu Z., Li J., Woo S.I., Hu H. Density functional theory studies of NO and NO2 adsorption on Al2O3 supported SnO2 cluster // Catal. Lett. 2013. V. 143. P. 912-918.
66. Grybos R., Hafner J., Benco L., [et al.]. Adsorption of NO on Pd-exchanged mordenite: Ab initio DFT modeling // J. Phys. Chem. 2008. V. 112. P. 12349-12362.
67. Mhadeshwar A.B, Winkler B.H., Eiteneer B.,[et al.]. Microkinetic modeling for hydrocarbon (HC)-based selective catalytic reduction(SCR) of NOx on a silver-based catalyst // Appl. Catal. B. 2009. V. 89. P. 229-238.
68. Oh H., Pieta I.S., Luo J., Epling WS. Reaction kinetics of C3H6 oxidation for various reaction pathways over diesel oxidation catalysts // Top. Catal. 2013. V. 56. P. 1916-1921.
69. Liu Z.M., Li J.H., Hao J.M. Selective catalytic reduction of NOx with propene over SnO2/Al2O3 catalyst // Chem. Eng. J. 2010. V. 165. P. 420-425.
