Ce1âxTixO2 Catalysts for Low-Temperature NH3-SCR

Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)

Article

Efficient VOx/Ce1-xTixO2 catalysts for low-temperature NH3-SCR: Reaction mechanism and active sites assessed by in situ/operando spectroscopy Thanh Huyen Vuong, Jörg Radnik, Jabor Rabeah, Ursula Bentrup, Matthias Schneider, Hanan Atia, Udo Armbruster, Wolfgang Grünert, and Angelika Brückner ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.6b03223 • Publication Date (Web): 18 Jan 2017 Downloaded from http://pubs.acs.org on January 18, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Catalysis is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Efficient VOx/Ce1-xTixO2 catalysts for lowtemperature NH3-SCR: Reaction mechanism and active sites assessed by in situ/operando spectroscopy Thanh Huyen Vuong,†,‡ Jörg Radnik,† Jabor Rabeah,† Ursula Bentrup,† Matthias Schneider,† Hanan Atia,† Udo Armbruster,† Wolfgang Grünert,§ and Angelika Brückner*,† †

Leibniz Institute for Catalysis at the University of Rostock, Albert-Einstein-Str. 29a, D-18059

Rostock, Germany ‡

School of Chemical Engineering, Hanoi University of Science and Technology, 1 Dai Co Viet,

10000 Hanoi, Vietnam §

Faculty of Chemistry and Biochemistry, Ruhr University Bochum, D-44780 Bochum, Germany

Dedicated to Prof. Dr. Bernhard Lücke on the occasion of his 80th birthday

ABSTRACT: Supported V2O5/ Ce1-xTixO2 (3, 5 and 7 wt. % V; x = 0, 0.1, 0.3, 0.5, 1) and bare supports have been tested in selective catalytic reduction (SCR) of NO by NH3 at different gas hourly space velocities (GHSV) and were comprehensively characterized using XRD, pseudo-in

ACS Paragon Plus Environment

1

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 41

situ XPS, UV-Vis-DRS as well as EPR and DRIFTS in in situ and operando mode. The best V/Ce1-xTixO2 (x = 0.3, 0.5) catalysts showed almost 100% NO conversion and N2 selectivity already at 190 °C with a GHSV of 70,000 h-1 which belongs to the best performances observed so far in low-temperature NH3-SCR of NO. The corresponding bare supports still converted around 80 % to N2 under the same conditions. On bare supports, SCR proceeds via a LangmuirHinshelwood mechanism comprising reaction of adsorbed surface nitrates with adsorbed NH3. On V/Ce1-xTixO2, nitrate formation is not possible, and an Eley-Rideal mechanism is working in which gaseous NO reacts with adsorbed NH3 and NH4+. Lewis and Brønsted acid sites, though adsorbing NH3, do not scale with the catalytic activity which is rather governed by the redox ability of the materials. This is boosted in the supports by replacing Ce with the more redoxactive Ti and in catalysts by tight connection of vanadyl species via O-bridges to the support surface forming –Ce–O–V(=O)–O–Ti– units in which the equilibrium valence state of V under reaction conditions is close to +5.

KEYWORDS: VOx/Ce1-xTixO2 catalysts, Low temperature NH3-SCR of NO, Mechanism, Operando EPR, In situ DRIFTS, Pseudo-in situ XPS.

INTRODUCTION Selective catalytic reduction (SCR) of nitrogen oxides by ammonia over V2O5-WO3/TiO2 catalysts is an established technology for cleaning flue gases from power plants which, however, operates only in a rather high and narrow temperature range of 300-500 °C. The temperature of NOx-containing exhaust gases from other sources such as diesel or lean-burn gasoline engines is


2

Page 3 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

much lower. Moreover, the restricted volume of catalytic converters in these engines requires catalysts that are active under high gas hourly space velocity (GHSV) conditions but at low pressure drop. Therefore, the design of SCR catalysts being sufficiently active and selective at low temperature (LT) and high GHSV is highly desired. A variety of catalysts has already been tested for LT-NH3-SCR.1-3 A survey of selected representative examples is given in Table 1. Table 1. Maximum NO conversion and N2 selectivity reached at respective temperatures and GHSV for selected representative LT-NH3-SCR catalysts. No.

Catalyst

Xmax(NO)/%

T/°C

S(N2)/%

GHSV/h-1

Ref.

1

β-MnO2

86

150

56

90.000

4

2

MnOx

98

80

n. a.

47.000

5

3

MnOx-CeO2

80

140

57

24.000

6

4

Mn4Ce6Ox

98

100

87

64.000

7

5

Mn0.3Ce0.7O2

100

110

n. a.

42.000

8

6

Ce0.5Zr0.5O2

35

220

n. a.

30.000

9

MnOx/Ce0.5Zr0.5O2

95

120

95

CeO2

18

220

45

70.000

10

VOx/CeO2

100

220

100

Ce0.7Zr0.3O2

53

300

98

70.000

11

VOx/Ce0.7Zr0.3O2

100

220

100

9

Ce0.6Ti0.4O2

98

300

100

50.000

12

10

Ce0.16Ti0.84O2

100

220

95-100

50.000

13

11

Ce0.3TiOx

100

175

98

25.000

14

7

8


3

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

12

Page 4 of 41

Ce0.7Ti0.3O2

80

200

100

70.000

This

VOx/Ce0.7Ti0.3O2

100

200

100

13

2.5% V2O5, 7% WO3/TiO2

95

200

n. a.

14.200

15

14

2% V2O5, 5% WO3/TiO2

80

200

100

40.000

16

15

3.5% V2O5, 10% WO3/TiO2

100

350

n. a.

50.000

17

work

With MnOx-based catalysts, almost total NO conversion has been obtained already at temperatures well below 150 °C (entries 1-5). However, such catalysts frequently suffer from deactivation by other flue gas components such as SO2 or H2O and/or from undesired N2O formation.1, 7, 18 The latter effect is usually more pronounced for pure MnOx and can be significantly suppressed in catalysts containing additional oxide components besides MnOx such as ceria, though also in these cases significant differences in catalytic performance appeared, depending on elemental composition, synthesis method and structure (Table 1, entries 3-6). Besides ceria-based catalysts, copper-exchanged zeolites have been reported as promising catalysts for diesel vehicles, however, their limited hydrothermal stability and N2 selectivity is a disadvantage for applications.19-20 Among the wide variety of catalysts tested in recent years, those based on (modified) ceria belong to the most promising ones (Table 1).21-23 CeO2 revealed to be a beneficial catalyst component not only for SCR but also for other redox reactions.24-26 This is due to the fact that it can store and release oxygen very efficiently by changing the Ce valence state between +4 and +3. This property is very important for a Mars-van-Krevelen mechanism (postulated, too, for NH3-SCR) in which the substrate is oxidized by lattice oxygen and the resulting O vacancies are replenished by uptake of gas-phase O2.27-30 For ceria-based materials, oxygen mobility is


4

Page 5 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

promoted even more, when isovalent metal cations of smaller diameter such as Zr4+ or Ti4+ are incorporated in Ce lattice positions.24 Thus, SCR catalysts based on ceria-containing mixed oxide supports almost always revealed to be more active than those supported on pure ceria (compare entries 7-12 in Table 1). An additional enhancement of catalytic activity could frequently be obtained by depositing small amounts of highly dispersed vanadia on such supports (entry 7, 8 and 12). Such materials showed even higher performances than commercial V2O5/WO3/TiO2 catalysts (compare with entries 13-15 in Table 1). Recently, we have shown this for series of V/CexZr1-xO2 catalysts in which almost 100 % NO conversion and N2 selectivity have been reached already around 200 °C with no deactivation during 190 h at a GHSV of 70,000 h-1 which outperformed the state of the art significantly.11 Based on these results, we anticipated that mixed oxide supports containing ceria in junction with a more redox-active component such as titania could lead to even more active catalysts for low-temperature SCR of NO. Indeed, a few recent papers suggest that such systems may be promising. Thus, 90 % of NOx was obtained at 200 °C with a 3 % V2O5/CeTiO2 catalyst, yet only at a GHSV of 50,000 h-1.31 With 1 % V2O5 – 5 % CeO2/TiO2 and 7 % V2O5 – 20 % CeO2/TiO2 catalysts a temperature of 250 - 300 °C was needed to obtain almost full NOx conversion at GHSV values of 128,000 h-1 and 100,000 h-1, respectively.32-33 As far as the mechanism of NH3-SCR over vanadia-containing catalysts is concerned, there seems to be agreement that acidic sites are needed for facilitating ammonia adsorption. However, while some authors believe that Lewis sites are preferred for this purpose,34-35 others think that it is the Brønsted –OH surface groups,36-37 or both Lewis and Brønsted sites on which NH3 is adsorbed.23, 38 In contrast, the adsorption of NO depends on the supports and the oxidation state of vanadium. Thus, NO was found to adsorb significantly on V2O5/activated semi-coke,39 V2O5 –


5

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 41

CeO2/TiO2,32 and reduced V2O5/TiO2 catalysts but weakly on oxidized V2O5/TiO2 catalysts.40-41 Moreover, different opinions exist about the role of redox-active sites. Both, a Ce4+/Ce3+ or a V5+/V4+ redox cycle and even the participation of V3+ have been discussed.42 In our recently studied V/CexZr1-xO2 catalysts we could show by in situ-EPR and pseudo-in situ-XPS that vanadium shuttles between V5+ and V4+ while Ce4+ and Zr4+ do not change their valence states,11 yet this may be different in a respective V/CexTi1-xO2 system. Furthermore, it is controversially discussed whether a Langmuir-Hinshelwood43-44 or an Eley-Rideal mechanism is relevant.45 To clarify these issues for the V/CexTi1-xO2 system, detailed spectroscopic in situ studies are needed that address both the interaction of feed components with surface adsorption sites as well as changes of the valence states of metal ions in the catalyst. In the present work, we have explored a series of catalysts containing V2O5 dispersed on mixed CeO2/TiO2 supports for low-temperature NH3-SCR of NO, since we expected that partial replacement of Ce4+ by smaller Ti4+ ions with a higher redox potential promotes defect formation and oxygen transport within the catalyst lattice and, thus, catalytic activity even more than observed for the corresponding V/CexZr1-xO2 system.11 Various techniques such as X-ray diffraction and photoelectron spectroscopy (XRD, XPS), electron paramagnetic resonance (EPR), UV-vis and IR spectroscopy in diffuse reflectance mode (UV-vis-DRS, DRIFTS) were used partly under reaction conditions, to relate catalyst properties such as redox behavior and surface acidity with catalytic performance and to derive conclusions on the reaction mechanism.


6

Page 7 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

EXPERIMENTAL SECTION Catalyst preparation Different Ce1-xTixO2 supports (x = 0, 0.1, 0.3, 0.5, and 1) were prepared by a coprecipitation method. Appropriate amounts of 0.125 M ammonium titanyl oxalate and 0.125 M cerium nitrate solutions were added dropwise to an aqueous solution of ammonia under vigorous stirring at room temperature to adjust the desired Ce/Ti ratio. During this time, the pH was kept at about 10. After stirring at room temperature for 1 h, the obtained solution was aged at 60 °C for 6 h, filtered and the precipitate was washed with deionized water. The obtained solid was dried first at room temperature, then at 100 °C for 12 h and subsequently calcined in air at 550 °C for 5 h. V/Ce1-xTixO2 catalysts with 3, 5, and 7 wt.% V2O5 loading were prepared by wet impregnation. A certain quantity of ammonium metavanadate was dissolved in 0.2 M oxalic acid solution. The required amount of the calcined Ce1-xTixO2 powder support was suspended in this aqueous solution. After 2 h stirring at room temperature, the excess water was evaporated using a water bath, and the solid residue was oven dried at 120 °C for 12 h and subsequently calcined in air at 400 °C for 5 h.11, 46 For comparison, a commercial 2% V2O5/8% WO3-TiO2 catalyst (Argillon GmbH) was used. Catalyst Characterization XRD powder patterns were recorded in the 2 Theta range from 5 – 80° by a theta/theta diffractometer (X’Pert Pro, Panalytical, Almelo, Netherlands) equipped with a X’Celerator RTMS Detector using Cu Kα radiation. The phase composition was determined using the program suite WinXPOW (STOE&CIE) with an inclusion of the Powder Diffraction File PDF2


7

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 41

of the International Centre of Diffraction Data (ICDD). Specific surface areas were determined by nitrogen adsorption at -196 °C using the single-point BET procedure (Gemini III 2375, Micromeritics). Raman spectra of the catalysts were recorded at room temperature using a Renishaw inVia Raman microscope. For the measurements, a 633 nm laser with a power of 0.17 mW was used. X-ray photoelectron spectra were obtained by a Thermo ESCALAB 220 iXL spectrometer (ThermoFisher) at room temperature using monochromatic Al Kα radiation. Binding energies were corrected with reference to C 1s value of 284.8 eV. Signal intensities were normalized using the sensitivity factors of Scofield 47 and the transmission function of the spectrometer. For pseudo-in situ measurements, catalysts have been pretreated at 200 °C in a reaction chamber attached to the spectrometer first under air flow for 30 min and subsequently under a flow of NH3-SCR feed (0.4% NH3, 0.4% NO, 10% O2/He (25 ml/min)) for 30 min. After each step, the samples were cooled to room temperature and transferred to the analysis chamber without intermediate contact to the ambient atmosphere. EPR spectra were measured by an X-band cw-spectrometer ELEXSYS 500-10/12 (Bruker) using a microwave power of 6.3 mW, and a modulation frequency and amplitude of 100 kHz and 5 G, respectively. In situ and operando EPR experiments were performed with 110 mg of catalyst particles of 250-350 µm size and a total gas flow of 50 ml/min in a home-made quartz plug-flow reactor connected to a gas-dosing device with mass flow controllers (Bronkhorst) at the inlet and a quadrupole mass spectrometer (Omnistar, Pfeiffer Vacuum GmbH) at the outlet for on-line product analysis. To study the impact of oxidizing, reducing and SCR reaction conditions separately, the catalysts were first pretreated in O2 flow for 1 h at 300 °C, cooled to


8

Page 9 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

200 °C, flushed with argon and then exposed for 30 min to a flow of 0.2 % NH3/He. Since the signal-to-noise ratio at 200 °C was high, the catalysts were cooled in this gas stream to 20 °C for recording the EPR spectrum. Subsequently, they were reheated to 200 °C in argon, treated for another 30 min in flowing 0.2 % NO, 5 % O2/He and cooled in this flow to 20 °C for spectra recording, before reheating again to 200 °C in argon and switching to the total feed stream for 30 min. In this case, NO and NH3 concentrations in the effluent were measured by on-line mass spectrometry. In situ-DRIFTS was performed on a Nicolet 6700 FTIR spectrometer using a high temperature reaction cell (Harrick) equipped with a temperature programmer (Eurotherm) and connected to a gas-dosing device with mass flow controllers (Bronkhorst) at the inlet. 145 mg of catalyst particles (250-350 µm) were pretreated for 1 h at 300 °C in air and subsequently exposed at 200 °C to a flow of different gases (30 ml/min). Background spectra were recorded in a flow of He and subtracted from the sample spectra for each measurement at the experiment temperature. Surface acidity was analyzed by FTIR spectroscopy of adsorbed pyridine using a Tensor 27 spectrometer (Bruker) equipped with a heatable and evacuable IR cell with CaF2 windows connected to a gas dosing evacuation system. For each experiment, 50 mg of powder were pressed into self-supporting wafers with a diameter of 20 mm, which were activated by heating to 400 °C in synthetic air followed by cooling to room temperature. Pyridine was adsorbed at room temperature until saturation. Spectra were recorded at room temperature after evacuation to remove the gas phase and subsequently at different temperatures from 100 °C to 400 °C. Background spectra of the activated samples were recorded in a flow of He and subtracted from the spectra of samples with adsorbed pyridine.


9

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 41

H2-TPR measurements were performed using a Micromeritics Autochem II 2920 instrument. For each experiment, 30 mg of support or 200 mg of V-containing catalyst particles (250-350 µm) were loaded into a U-shaped quartz reactor and heated from room temperature to 400 °C with 20 K/min in 5% O2/He (30 ml/min), kept under these conditions for 30 min and cooled to room temperature under flushing with Ar. TPR runs were carried out from room temperature to 800 °C in a 5% H2/Ar flow (50 ml/min) with a heating rate of 10 K/min. Hydrogen consumption was monitored by a TCD detector. NH3-SCR activity test Catalytic tests were conducted in a fixed-bed quartz plug-flow reactor (length 200 mm, internal diameter 6 mm) with a feed of 1000 ppm NO, 1000 ppm NH3, 5 vol.% O2/He and a GHSV of 70,000 h-1 using 100 mg of catalyst particles of 250-350 µm size and a total flow of 100 ml/min. Products leaving the reactor were analyzed by on-line gas chromatography (HP 6890) using a molecular sieve 5A column for analysis of N2, N2O and O2. Simultaneously, a multigas sensor (Limas 11HW, ABB, Germany), including a catalytic converter was used to determine NO, NO2, and NH3 concentrations. Due to inlet temperature and chemical equilibrium, NO was partly converted into NO2 prior to entering the reactor and this has been considered when calculating conversion and selectivity. Reactant and product contents in the effluent were monitored during 1 h of continuous reaction in steady state at each temperature chosen. N2O formation was below 10 ppm and is therefore not considered in due course. For analyzing the effect of space velocity on the catalytic behavior, catalysts were tested with the same feed composition in the range of 150-400 °C at higher GHSV of 300,000 and 750,000 h-1.


10

Page 11 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

RESULTS AND DISCUSSION Catalytic behavior Figure 1 shows the effect of temperature on NOx and NH3 conversion as well as on N2 selectivity for all catalysts with a loading of 5 % V2O5 in comparison to the corresponding bare supports. It can be seen that, apart from pure CeO2 and TiO2, even the CexTi1-xO2 supports show significant activity which increases upon replacement of about one third of the Ce sites by Ti and does not change much upon further raising the Ti content (compare Ce0.7Ti0.3O2 and Ce0.5Ti0.5O2), reaching complete NOx and NH3 conversion between 225 °C to 300 °C. The performance of the corresponding V-containing catalysts is significantly higher than that of the bare supports. With the best V/Ce0.7Ti0.3O2 and V/Ce0.5Ti0.5O2 catalysts full conversion is already reached slightly below 200 °C. Catalysts with 3 and 7 % V2O5 deposited on the most active support Ce0.5Ti0.5O2 as well as a commercial 2% V2O5/8% WO3-TiO2 catalyst have also been tested for comparison, yet the catalytic performance was in all cases lower than that of the best 5 % V2O5/Ce0.5Ti0.5O2 catalyst (Figure S13). This is particularly true for the commercial catalyst, depite the fact that its surface atomic ratio of V/(W+Ti) = 0.35 derived by XPS is higher than the corresponding ratio V/(Ce+Ti) = 0.27 of the best V/Ce0.5Ti0.5O2 catalyst described below. Therefore, these samples are not included in further studies on structure-reactivity relationships described below. It should be mentioned that an optimum loading of 5 % V2O5 has been identified, too, in our previous work on V2O5/CexZr1-xO2. This is also in line with the observations of Li et al.21 Interestingly, the gain in catalytic activity upon vanadia deposition decreases in the order Ce0.5Ti0.5O2 ≈ Ce0.7Ti0.3O2 > Ce0.9Ti0.1O2 >>CeO2 > TiO2. On the first glance, one could think


11

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 41

that this behavior is related to a drop of the specific surface area since the SBET values follow the same trend (Table 2). However, surface areas are most probably not a major factor that governs catalytic activity. This is evident by comparing NO conversions of two 5% V/TiO2 catalysts prepared by the same synthesis procedure but with TiO2 supports of the different surface area (SBET = 18.6 m2g-1 and SBET = 350 m2g-1, Figure S12). The two catalysts show almost the same activity despite the huge difference in surface area. A similar observation was also made by Wachs et al.48 who investigated NH3-SCR over a series of supported vanadia catalysts. They showed that, although the BET surface area of the supports increased in the order TiO2 (50m2/g) < Al2O3 (180 m2/g) < SiO2 (300 m2/g), the catalytic performance of the corresponding catalysts decreased in the order V/TiO2>V/Al2O3>V/SiO2. In agreement with these authors, we anticipate that in our case it is the interface between the deposited vanadia and the TiO2 support rather than the BET surface area of the latter which is crucial for catalytic activity since this might govern the redox properties of the catalysts. As expected from the almost equal slope of the NO and NH3 conversion curves, N2 selectivity is close to 100 % for all catalysts, except for the bare CeO2 and TiO2. In these two cases, N2 selectivities are slightly lower, due to the formation of some N2O, which does not occur with the V- containing catalysts.


12

Page 13 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Figure 1. NOx (A), NH3 (B) conversion, and N2 selectivity (C) over pure Ce1-xTixO2 supports (open symbols, dashed lines) and respective V/Ce1-xTixO2 catalysts (filled symbols, solid lines) as a function of temperature. Feed composition: 0.1 % NO, 0.1 % NH3, 5 % O2/He, GHSV = 70,000 h-1. For practical applications in diesel or lean-burn engines, the volume of the catalysts should be as low as possible to account for the limited volume of the catalytic converters.21, 49 This means that the catalysts must enable high conversions also at short contact times. Therefore, the effect of GHSV on the catalytic behavior of the V/Ce1-xTixO2 catalysts has been studied (Figure 2). It can be seen that the NOx conversion over all catalysts decreased significantly when the GHSV was raised from 70,000 to 300,000 h-1 and further to 750,000 h-1 in the whole temperature range


13

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 41

(compare Figures 1A and 2A). However the best catalyst (V/Ce0.5Ti0.5O2) is still remarkably active below 300 °C with X(NOx) ≈ 70 % at GHSV = 300,000 h-1 and X(NOx) ≈ 50 % at 750,000 h-1. Moreover, with the best mixed oxide catalysts V/Ce0.7Ti0.3O2 and V/Ce0.5Ti0.5O2 the undesired formation of N2O remained negligible over the whole temperature range while this became significant for V/TiO2 and V/CeO2 particularly at higher temperatures (Figure 2B). This indicates that incorporation of titanium into ceria does not only help to retain high activity but also high N2 selectivity, which is a promising basis for further catalyst optimization.

Figure 2. NOx conversion (A) and N2O concentration (B) over V/Ce1-xTixO2 catalysts (x = 0, 0.3, 0.5 and 1) at a GHSV of 300,000 h-1 (filled symbols, solid lines) and 750,000 h-1 (open symbols, dashed lines) as a function of temperature. Feed composition: 0.1% NO, 0.1% NH3, 5% O2/He. To gain more information about the beneficial role of V and Ti in V/Ce1-xTixO2 catalysts for the performance in NH3-SCR, a comprehensive characterization study has been performed. The following description of results is, however, mainly restricted to the best VOx/Ce0.5Ti0.5O2 catalyst in comparison to V/CeO2 and V/TiO2.


14

Page 15 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Bulk and surface properties of supports and catalysts The XRD powder patterns indicate that pristine TiO2 is anatase while CeO2 obeys the cubic fluorite structure. The latter also persists upon incorporation of Ti, yet all reflections become broader with rising Ti content (Figure S1). This indicates a decrease in crystallite size (Table 2) and/or higher lattice disorder due to replacement of Ce4+ (97 pm) by smaller Ti4+ ions (74 pm) and confirms the formation of solid Ce1-xTixO2 solutions.50-51 The XRD patterns of the corresponding V-containing mixed oxide catalysts do not show any additional reflections, pointing to highly dispersed and/or amorphous vanadium oxide species on the surface of the supports. When Ti is incorporated into CeO2, BET surface area and pore volumes increase significantly and reach a maximum value of 113.5 m2g-1 for Ce0.5Ti0.5O2, which does not change much upon deposition of vanadia, showing that pore blockage is negligible. The pore size distribution of all supports and catalysts does not differ much and is in the range of 3 to 7 nm in agreement with previous observations.52 Information on surface composition and valence states is provided by XPS data listed in Table 3. The corresponding spectra are shown in the supporting information (Figure S2). The Ti 2p EB values of both catalysts V/TiO2 and V/Ce0.5Ti0.5O2 after oxidative pretreatment are attributed to Ti4+ states (459.1 – 458.8 eV).52-53 Moreover, samples V/CeO2 and V/Ce0.5Ti0.5O2 contain exclusively Ce4+, indicated by main peaks at 919 eV and 901 eV. Any peaks between 880 and 881 eV characteristic for Ce3+ were not found.31, 54-55 The binding energies (EB) of the O 1s peaks are very similar with values known for CeO2, TiO2 and their solid solutions.52, 54, 56-57 Sample V/TiO2 shows a V 2p3/2 peak at about EB = 517.2 eV being characteristic for V5+,58-60 together with a shoulder at EB = 518.4 eV. The latter may stem from highly dispersed VOx species while


15

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 41

the main peak around EB ≈ 517 eV might arise from larger V2O5 nanocrystals.58 The latter have been evidenced in this sample by Raman spectra discussed below. Table 2: Crystallite size, specific surface area and pore volume, band gap energy of supports and catalysts Sample

Mean crystallite size (nm) [a]

Surface area (m2 g-1)

Pore volume (cm3 g-1)

Average pore size (nm)

Band gap energy (eV) [b]

CeO2

10.8

61.2

0.060

3.16

2.72

Ce0.9Ti0.1O2

10.0

94.9

0.188

6.54

2.86

Ce0.7Ti0.3O2

8.7

91.1

0.134

4.74

2.66

Ce0.5Ti0.5O2

8.3

113.5

0.169

4.44

2.70

TiO2

24.0

20.1

0.025

3.06

2.90

V/ CeO2

10.8

45.4

0.053

3.63

2.25

V/Ce0.9Ti0.1O2

10.5

73.2

0.162

7.05

2.65

V/Ce0.7Ti0.3O2

8.6

77.2

0.128

5.10

2.60

V/Ce0.5Ti0.5O2

6.2

119.0

0.157

4.10

2.57

V/TiO2

23.0

18.6

0.025

3.64

2.11

a

derived by the Scherrer equation from XRD data, bresults from UV-vis

In both samples V/CeO2 and V/Ce0.5Ti0.5O2 only one peak at 517.0-516.9 eV is seen, though in the V/Ce0.5Ti0.5O2 catalyst VOx dispersion is highest and no V2O5 nanocrystals were found (vide infra). Possibly, the peak at 518.4 eV in V/TiO2 originates from few V single sites persisting besides the dominating tridimensional V2O5 clusters and nanoparticles while the shift to lower EB


16

Page 17 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

in V/Ce0.5Ti0.5O2 is caused by abundant small VxOy clusters. In summary, all surface metal ions are in their highest valence state after oxidative pretreatment. The surface V/(Ce + Ti) ratios are equal to V supported on TiO2 and CeO2 and almost twice as high as on the mixed oxide support (Table 3). A possible reason may be the much higher BET surface area of Ce0.5Ti0.5O2 which, despite a higher dispersion of the VOx sites, may lead to a lower V surface concentration at the same total V2O5 content of 5 wt. %. Table 3: XPS binding energies (eV) and surface atomic ratios obtained after treatment in air and SCR feed without contact to the ambient atmosphere. Sample

V/CeO2

Binding energy (eV)

Atomic ratio

O 1s

Ti 2p

V 2p3/2

Ce 3d

Ce/Ti

V/(Ce+Ti)

air

528.9

-

517.0

885.5

-

0.52

SCR

529.0

-

516.2

-

0.44

885.5 882.5 V/Ce0.5Ti0.5O2

air

529.4

458.8

516.9

884.8

1.18

0.27

SCR

529.4

458.9

516.9

884.8

1.51

0.20

air

530.2

459.1

-

-

0.53

-

-

0.41

517.2 V/TiO2

518.4 SCR

530.1

458.4

517.2

EPR spectra of fresh catalysts are almost identical with those of the bare supports, showing only weak signals of paramagnetic oxygen defects such as O•- and/or O2•- species.61 An exception is V/TiO2, in which a broad signal of magnetically interacting VO2+ species is seen around g = 1.963 (Figure S3). Since VOx clusters might dominate in V/TiO2, it is probable that


17

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 41

this signal results from partially reduced V sites in subsurface regions of these agglomerates since, according to XPS (Table 3), the surface of this catalyst exposes V5+ only. UV-Vis diffuse reflectance spectroscopy (DRS) is another useful method to provide information on the structure of VOx species in the catalysts. Since the wavelength range below 400 nm is superimposed by ligand-metal charge-transfer (LMCT) transitions of Ce4+ and Ti4+ (Figure 3A), it is not possible to derive reliable conclusions about the coordination of V5+ single sites, the LMCT transitions of which fall into this range, too. However, from the position of the absorption edge above 400 nm, at least qualitative information can be obtained about the extent of VOx agglomeration since it has been shown that the edge energy derived from UV-Vis-DRS (Table 2, Figure S6) is related to the number of V-O-V bridges.62 The lowest values of the band gap energy Eg = 2.11 and 2.25 eV are observed for V/TiO2 and V/CeO2 suggesting a rather low VOx dispersion. This is in agreement with Raman data discussed below and with EPR results described above. In contrast, higher and similar Eg values of 2.57-2.65 eV have been derived for V/Ce1-xTixO2 catalysts, indicating the prevalence of highly dispersed VOx sites on the surfaces of the mixed oxide supports. These results are also supported by Raman spectra in Figure 4, in which catalysts V/CeO2 and V/TiO2 show weak bands of (XRD-silent) nanocrystalline V2O5 at 145, 195, 282, 702, 864 and 994 cm-1,53, 63 besides the typical bands of ceria at 463 cm-1,53, 56 and of anatase at 144, 395, 515 and 614 cm-1, respectively.53, 64 The V/Ce1-xTixO2 catalysts show only the band of cubic ceria at 463 cm-1, yet with lower intensity, due to decreasing crystallite size and/or increasing structural disorder with rising Ti content (Table 2). No Raman bands of V2O5 nanocrystals can be seen in these catalysts which indicate, in agreement with the other spectroscopic data discussed above,


18

Page 19 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

that the incorporation of titanium into ceria improves the dispersion of deposited VOx surface sites.

Figure 3. UV-Vis DRS spectra of A) bare supports and B) 5% V2O5/Ce1-xTixO2 (V/Ce1-xTixO2) catalysts.

Figure 4. Raman spectra of the catalysts VCe1-xTixO2 (x = 0-1). Pyridine adsorption studied by FTIR spectroscopy has been used to compare the acidity of supports and catalysts. Spectra measured at 150 °C are shown in Figure 5. All samples show bands at 1608-1597 and 1445-1442 cm-1, which are assigned to pyridine coordinated to Lewis


19

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 41

acid sites (L-Py).65-66 The strength of Lewis sites depends on the position of the ν(8a) vibration in the range between 1590 and 1630 cm-1.67 The more these bands are shifted to higher wavenumbers, the stronger are the corresponding Lewis sites. Inspection of Figure 5A and B shows, that the Lewis sites of bare TiO2 and V/TiO2 give rise to bands at almost the same position (1605 and 1607 cm-1), indicating that Ti4+ and V5+ are sites of similar Lewis strength. In comparison, Ce4+ might be a weaker Lewis site as suggested by the shift to lower wave numbers in the bare CeO2 support (1600 cm-1) and the V/CeO2 catalyst (1601 cm-1).68 In the mixed oxide support Ce0.5Ti0.5O2 and the corresponding V-catalyst, this band falls at an intermediate value, due to the superposition of Lewis sites related to Ce4+, Ti4+ and V5+. The relative amount of Lewis sites reflected by the integrated area of the L-Py band at 1445 cm-1 (Figure 5, ILewis in Table S1) increases in the order CeO2 < Ce0.5Ti0.5O2 V/CeO2 > V/TiO2. The same trend is observed for the catalytic activity of the V-containing catalysts as well as for the pure Ce-containing supports (Figure 1). This suggests that Lewis acidity may be beneficial for the SCR reaction, though it is definitely not the main factor since pure TiO2 with the highest Lewis acidity shows the lowest catalytic activity. Interestingly, normalization of the band area of the Lewis sites at 1445 cm-1 leads to a different order for the V-catalysts, which does not follow the trend in catalytic activity: V/CeO2 > V/Ce0.5Ti0.5O2 ≈ V/TiO2. This again supports the conclusion made above, that surface areas are most probably not a major factor that governs catalytic activity. Bands at 1537 and 1635 cm-1 are related to PyH+ ions, i. e. to pyridine adsorbed on Brønsted sites.67, 69 They are very weak and only observed in V-containing catalysts. Probably they


20

Page 21 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

originate from V-OH groups. These bands decrease in the order V/TiO2 > V/Ce0.5Ti0.5O2 > V/CeO2 (Table S1) and do not correlate with observed differences in the catalytic behavior, suggesting that Brønsted sites might not play an important role in NH3-SCR of NOx over these catalysts.

Figure 5. FTIR spectra of adsorbed pyridine on A) bare supports Ce1-xTixO2 (x = 0; 0.5; 1) and B) supported vanadium catalysts V/Ce1-xTixO2 recorded after evacuation at 150 °C. As explained in the introduction, reducibility may be an important catalyst property anticipating that NH3-SCR involves a Mars-van-Krevelen redox cycle.28-29 Therefore, selected catalysts and the corresponding bare supports were also analyzed by H2-TPR. However, as evident from Figure S7, these results might be of rather limited relevance since measurable reduction under TPR conditions starts only above 350°C while the catalysts show high activity already well below 200°C. Therefore, the results obtained from in situ spectroscopy, particularly those of operando EPR and in situ XPS described below, are certainly more relevant for assessing the redox properties of the catalysts during NH3-SCR, since they have been obtained under conditions very close or even identical to that of the catalytic reaction. Nevertheless, the


21

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 41

TPR findings seem to be somehow in line with the observed trend in catalytic activity. The least active bare TiO2 support shows negligible reducibility, followed by bare CeO2 for which a weak broad signal with a maximum at 490 °C is seen which can be assigned to the consumption of the surface capping oxygen of CeO2 while above 600°C reduction of bulk CeO2 starts.52, 70 Incorporation of Ti into ceria enhances reducibility of Ce0.5Ti0.5O2 reflected by higher H2 uptake and a shift of the peak maximum to lower temperature. Interestingly, a weak peak appears already at about 225°C in this sample, which can be assigned to the reduction of Ce tightly bound to Ti species.31 This could explain why the bare Ce0.5Ti0.5O2 support is markedly more active than bare CeO2 and TiO2 (compare Figure 1). The peak maximum of the least active V/TiO2 catalyst falls at significantly lower temperature compared to the more active samples V/CeO2 and V/Ce0.5Ti0.5O2 and this catalyst also shows the highest percentage of reduced V species under NH3-SCR conditions, as evident from operando EPR results described below. EPR and XPS results also show that V/CeO2 is markedly more reduced that V/Ce0.5Ti0.5O2 under NH3-SCR conditions. This effect, which is discussed in terms of its relevance for catalytic activity below, is hardly reflected by TPR results, illustrating again the limited usefulness of this method for the present catalytic system.

Spectroscopic in situ and operando studies It has been shown previously for V2O5/TiO2 catalysts that the SCR of NO by NH3 comprises reaction steps catalyzed by both acid and redox sites, though the interpretation of these results differs. While a direct correlation between NO conversion and the concentration of Brønsted sites (responsible for NH3 adsorption as NH4+) was postulated by Topsoe,45 such a correlation


22

Page 23 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

was proposed between NO conversion and number of V5+=O redox sites but not the amount of adsorbed NH4+ elsewhere.40 In any case, these results show that it is of utmost importance to study the behavior of V/Ce1-xTixO2 catalysts under reaction conditions to derive reliable conclusions about mechanistic details. Therefore, in situ DRIFTS, pseudo-in situ XPS, and operando EPR spectroscopy have been applied, the results of which are presented below. DRIFTS investigations In situ DRIFT spectra of the bare supports and the V-containing catalysts in flowing 0.1 % NH3/He contain bands at 3376-3150 cm-1 arising from ν(N-H) stretching vibrations of NH3 adsorbed on Lewis sites (Figure S8).41, 45 The corresponding bending vibrations are observed at 1225-1156 and 1603 cm-1 (Figure 6).41, 45, 71 Among the catalysts, the intensities of these bands are highest for V/Ce0.5Ti0.5O2 suggesting the highest amount of Lewis acid sites on this catalyst. Besides, additional bands at 1520-1510 cm-1 of NH2- species,72 only observed on pure CeO2 and Ce0.5Ti0.5O2, point to dissociative NH3 adsorption on the support surface. The characteristic bands for NH4+ formed by the interaction of NH3 with Brønsted sites around 1420 cm-1 are only observed for the V-containing catalysts and might stem from V-OH moieties.41, 45, 71 This is also supported by the negative band at 3646-3640 cm-1 (Figure S8) which indicates the consumption of the latter. These results are in good agreement with pyridine adsorption data presented above. Another negative band at 2043 cm-1 is due to the 2ν overtone vibration of V5+=O groups which are partly reduced to V4+=O and/or covered by NH3.45 This reduction is also evident from in situXPS and -EPR data discussed below.


23

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 41

Figure 6. Difference DRIFT spectra of bare supports and supported vanadium catalysts recorded at 200 °C after 45 min in 0.1 % NH3/He flow. When the bare supports with preadsorbed NH3 are subsequently treated at 200 °C in a flow of 0.1 % NO, 5 % O2/He, a similar behavior was observed for CeO2 and Ce0.5Ti0.5O2, in which ν(N-H) and δ(N-H) bands of NH3 adsorbed on Lewis sites above 3150 cm-1 (Figure S9A) and between 1150 and 1603 cm-1 (Figure 7A) disappear gradually while new bands arise. Those below 1600 cm-1 are due to monodentate (1533 and 1258-1253 cm-1), bridged (1600-1591 and 1205 cm-1) and bidentate nitrate species (1570-1565 and 1218-1200 cm-1).72-74 The weak features between 2200 and 1900 cm-1 may stem from combination bands of adsorbed NOx species.75-76


24

Page 25 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Additionally, bands at 1995-1935 cm-1 of M-NO mononitrosyl species (M= Ce, Ti), adsorbed NO2 (1689 cm-1) and N2O4 (1728-1726 cm-1) are seen.72 A broad band at 2220 cm-1 can be assigned to adsorbed NO+ cations.72,77 In contrast, for pure TiO2, nearly no change was observed in the spectra after switching from NH3 to NO+O2, indicating that adsorbed NH3 is very stable and poorly reactive on the surface of this support. Note that this support shows also negligible catalytic activity at 200 °C (Figure 1). Interestingly, on supported vanadium catalysts, the bands of preadsorbed NH3 disappear in the flow of 0.1% NO, 5% O2/He, too. This happens much slower on the least active catalyst V/TiO2, pointing again to the high stability and poor reactivity of adsorbed NH3. Remarkably, no stable adsorbed nitrate, nitrosyl and NO+ ions are formed (Figure 7B). Moreover, on both catalysts V/CeO2 and V/Ce0.5Ti0.5O2 the negative intensities of the 2ν(V=O) and the ν(V-O-H) bands at 2047-2043 cm-1 (Figure 7B) and 3646-3640 cm-1 (Figure S9B), respectively, turn into positive features due to the reoxidation of V4+ to V5+=O and V5+-OH surface sites, which is also supported by the EPR data described below. In contrast, for the least active catalyst V/TiO2 the 2ν(V=O) band remains negative, indicating that reoxidation of V4+ on this sample is much slower and more difficult, possibly due to coverage of these sites by adsorbed NH3. These results clearly show that both V-OH Brønsted, as well as V=O redox sites, participate in the NH3-SCR mechanism. This was found previously for V2O5/TiO2 and is here confirmed for V/Ce1-xTixO2 catalysts, too.45 However, our results suggest that the impact of V=O redox sites might be more important since the differences in Brønsted acidity between the V-containing catalysts are marginal (Figure 5 and 6).


25

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 41

In a separate experiment, bare supports and catalysts were first exposed to NO/O2 and subsequently to NH3/He flow (Figure 8). After exposure to NO/O2, the same bands appeared as observed after first exposing to NH3/He and afterwards to NO/O2 flow (compare grey and black spectra in Figure 7A), confirming the formation of differently bound nitrate, M-NO (M= Ce, Ti) and NO+ species on CeO2 and Ce0.5Ti0.5O2. On TiO2, the amount of adsorbed nitrates is much lower compared to CeO2 and Ce0.5Ti0.5O2 and no Ti-NO and NO+ species are observed. This suggests that nitrate formation from NO/O2 might be promoted by the higher oxygen mobility and oxidation activity of Ce-containing supports. Moreover, it suggests that it may be preferentially the oxygen in the vicinity of Ce that participates in nitrate formation, since the amount of these adsorbates is somewhat higher on CeO2, despite the lower BET surface area compared to Ce0.5Ti0.5O2 (Table 2). Interestingly, no bands at all are created upon exposure of the V-containing catalysts to NO/O2 flow at 200 °C (Figure S10). Therefore, the corresponding reference spectra are not shown in Figure 7B. This evidences clearly that NO does not interact with the vanadium catalyst surface but only with adsorbed NH3 and NH4+ species whereby, however, no surface nitrates are formed in the latter case (compare Figure 7B).


26

Page 27 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Figure 7. Difference DRIFT spectra recorded at 200 °C of bare supports (A) and V-containing catalysts (B) after 45 min exposure at 200 °C to 0.1 % NH3/He flow and subsequent switch to 0.1 % NO, 5 % O2/He flow. Top traces in plot A depict spectra recorded after 45 min exclusive exposure to 0.1 % NO, 5 % O2/He flow without pretreatment in NH3/He. For plots of the range 2500-4000 cm-1 see Figure S9.


27

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 41

Upon subsequent exposure of the NO/O2-pretreated supports to a flow of 0.1 % NH3/He (Figure 8) the bands of nitrates, M-NO (M = Ce, Ti) and NO+ species decrease on Ce0.5Ti0.5O2 and bands already known for adsorbed NH3 arise, indicating the reaction of adsorbed NOx with (adsorbed) NH3 most probably to N2 and H2O (compare red spectrum in Figure 7A and black line in Figure 8). In contrast, on pristine CeO2, the bands of all pre-adsorbed NOx species do not change under NH3/He flow at 200°C (Figure 8), which clearly indicates that these species are stable against NH3 attack. This might be one reason for the much lower catalytic activity of bare CeO2 in comparison to Ce0.5Ti0.5O2 (Figure 1). On TiO2, only weak bands of adsorbed NOx species were formed during pretreatment in NO/O2 which under NH3/He flow are replaced immediately by adsorbed NH3 (band at 1603 cm-1). On V-containing catalysts (not shown in Figure 8), only rising bands of adsorbed NH3 are seen (Figure S10). Comparison of all these in situ DRIFTS results points to interesting mechanistic differences in the NH3-SCR over bare CeO2, TiO2 and Ce1-xTixO2 supports and the corresponding V-containing catalysts. While NH3 can be adsorbed and activated on all surfaces, NOx species are only adsorbed on the bare supports but not on the catalysts. A possible reason may be that Ce-O moieties responsible for oxidation and fixation of NO/O2 as surface nitrates are covered by VOx. This suggests that NH3-SCR proceeds after a Langmuir-Hinshelwood mechanism on the bare supports, in which both NO and NH3 are first adsorbed to give activated surface species that further react to produce N2 and H2O. To further confirm this experimentally, an additional DRIFTS experiment has been performed, in which NH3 and NO/O2 was subsequently adsorbed at room temperature. Upon heating the bands of adsorbed NH3 and nitrate disappeared and simultaneously N2 was detected by on-line mass spectrometry (Figure S11). In the case of pure CeO2 and TiO2 the adsorbed NOx and NH3 species are obviously so stable that their catalytic


28

Page 29 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

conversion is hindered, which explains the low activity of these solids. In contrast, an EleyRideal mechanism might be operative in the case of the V-containing catalysts, since only NH3 but no NOx is adsorbed on the surface. This means that gas-phase NO/O2 must react with adsorbed NH3 and NH4+. This is in line with proposals made for NH3-SCR over pure V2O5,36 V2O5/TiO2,45 and V2O5/CeO2,31 but contradicts other studies in which a Langmuir-Hinshelwood mechanism is suggested for the same reaction on V2O5, V2O5/alumina,43 and V2O5/TiO2.44 While in situ DRIFTS provides insights on the nature of adsorbed intermediates formed upon interaction of feed components with surface sites, results of operando EPR and pseudo-in situXPS described in the next section allow conclusions on reaction-dependent valence changes of V, Ce and Ti.

Figure 8. Difference DRIFT spectra recorded at 200 °C of bare supports under flowing of 0.1% NH3/He after pretreatment at 200°C for 45min in 0.1 % NO, 5 % O2/He flow.


29

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 41

EPR investigations Operando EPR spectra after subsequent treatment in different gas flows are exemplarily shown for the most active catalyst V/Ce0.5Ti0.5O2 in Figure 9 (for the spectra of the other catalysts and results of the simultaneous mass spectrometric analysis of the effluent gas stream see Figures S4 and S5). After oxidative pretreatment, only weak signals tentatively assigned to paramagnetic oxygen defects in the support are seen (compare Figure S3). Treatment in NH3/He flow gives rise to the characteristic signal of VO2+ with hyperfine structure (hfs) from the coupling of the single electron spin of V4+ (d1, S = ½) with the nuclear spin of V (I = 7/2, 100 % natural abundance), indicating that V5+ is partially reduced in the presence of NH3. This signal completely disappeared when the catalyst is subsequently exposed to a flow of 0.1 % NO, 5 % O2/He (Figure 9C), pointing to reoxidation. This reduction/reoxidation of vanadium agrees pretty well with the behavior of the ν(V-O-H) and 2ν(V=O) bands at 3646-3640 and 2043 cm-1 in the DRIFT spectra (Figures S9B and 7B). Moreover, it indicates that oxygen participating in LTNH3-SCR is activated via a Mars-van Krevelen mechanism.28, 34 Afterwards, upon switching to the total SCR feed flow, the hfs signal of VO2+ appears again, yet with lower intensity, which relates to the equilibrium V valence state under reaction conditions. Interestingly, the total EPR intensity of the most active V/Ce0.5Ti0.5O2 catalyst under SCR conditions is significantly lower compared to the less active V/CeO2 and V/TiO2 catalysts (Figure 9, right). This suggests that the equilibrium concentration of V5+ in the active state is highest for the most active catalyst.


30

Page 31 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Figure 9. Left: Operando EPR spectra of sample V/Ce0.5Ti0.5O2 recorded at 20 °C after (A) 1 h pretreatment in O2 flow at 300°C, (B) 30 min exposure to 0.1 % NH3/Ar (C) 30 min exposure to 0.1 % NO, 5 % O2/Ar and (D) 30 min exposure to total SCR feed flow (spectrum c subtracted, dashed line shows spectrum fitted with spin Hamiltonian parameters in Table 3); Right: Comparison of spectra (D) for V/Ce0.5Ti0.5O2, V/CeO2 and V/TiO2. As we have shown recently for the V/Ce1-xZrxO2 system,11 spin Hamiltonian parameters of the EPR spectra under SCR feed provide valuable information about the location and structure of single V4+=O sites, being considered as representative also for the corresponding EPR-silent V5+ species from which they are formed by reduction.78-79 Thus, A|| increases when the V=O bond shortens, ∆g||/∆g⊥ (with ∆g|| = g|| - ge, ∆g⊥ = g⊥ - ge, ge = 2.0023) increases with rising axial distortion and the coefficient β*22 (Eq. 1, with P = 184.5 G being the strength of the electronnuclear dipole-dipole interaction for the free V4+ ion)80 is a measure for the degree of covalence of the V-O bonds. It is unity for a pure VO2+ cation and decreases with rising covalent character.79, 81 β*22 = (7/6)∆g|| - (5/12)∆g⊥ -(7/6)[(A|| - A⊥)/P]

(1)


31

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 41

For all catalysts, two hfs signals A and B for different single VO2+ species and a broad isotropic singlet C for magnetically interacting VO2+ species had to be assumed to obtain satisfactory fits of the experimental spectra (Table 4, Figures 9 and S4). First of all, it can be seen that the contribution of the broad singlet C decreases in the order V/TiO2 > V/CeO2 > V/Ce0.5Ti0.5O2, which agrees well with the dispersion of VOx sites being highest for the catalyst with the mixed oxide support. The similarity of the parameters of the single VO2+ sites A suggests that these species might have a similar environment in all three catalysts. Possibly, they are located more in subsurface regions of the V2O5 agglomerates without direct contact to the support surface, while sites B may be rather situated on the support interface. Remarkably, the in-plane delocalization coefficient β*22 of VO2+ sites B on the mixed oxide support Ce0.5Ti0.5O2 is significantly lower than the respective values of B sites on pure CeO2 and TiO2, suggesting that the former are more covalent and, thus more effectively bound to the support. A similar though less significant effect was also found for the best catalyst V/Ce0.7Zr0.3O2 from the V/Ce1-xZrxO2 series.11 In analogy to these results we propose that the V sites B in catalyst V/Ce0.5Ti0.5O2 may be part of –O–Ce–O–V(=O)–O–Ti–O– surface moieties. This confinement could facilitate electron transfer between the Vn+=O site and the support and, thus, promote redox activity of these sites. The fact that best catalyst from the V/Ce1-xTixO2 series is more active than the analogous catalyst from the V/Ce1-xZrxO2 series might be due to the higher redox potential of Ti compared to Zr.82


32

Page 33 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Table 4. Spin-Hamiltonian parameters derived by simulation of the difference EPR spectrum (NO/NH3/O2)-(NO/O2) of V/Ce1-xTixO2 catalysts

V/CeO2

V/Ce0.5Ti0.5O2

V species

g

g⊥

A / G

A⊥ / G ∆g/∆g⊥

β*22

Irel / %

a

1.922

1.976

185.4

62.6

3.05

0.859

19

b

1.933

1.967

175.7

55.8

1.96

0.824

24

c

1.963

-

-

-

-

57

a

1.924

1.977

180.0

63.5

3.14

0.818

35

b

1.933

1.964

182.5

77.3

1.81

0.730

38

c

1.963

-

-

-

-

27

a

1.927

1.979

179.9

63.4

3.16

0.814

10

b

1.938

1.971

178.3

52.5

2.05

0.857

11

c

1.963

-

-

-

-

79

V/TiO2

XPS investigations Since EPR can only detect V4+ at temperatures ≥ 20°C, pseudo-in situ-XPS experiments have been performed after treatment of the catalysts in a reaction chamber attached to the spectrometer to analyze the behavior of Cen+ and Tin+ in addition to Vn+. As discussed above, all surface metal ions are in their highest valence state after oxidative pretreatment (section 3.2, Table 3). After treatment in SCR feed, the V 2p3/2 signal of the most active catalyst V/Ce0.5Ti0.5O2 remains almost at the same position (516.9 eV) indicating that the majority of the surface V species stays pentavalent. In contrast, this peak shifts to lower binding energy after SCR in


33

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 41

sample V/CeO2 and in catalyst V/TiO2 the shoulder at 518.4 eV vanished. This points, in agreement with operando EPR data, to a partial reduction of V5+ to V4+ (Table 3). This effect is hardly detectable for the most active catalyst V/Ce0.5Ti0.5O2 which is also in line with the much smaller VO2+ EPR signal intensity of this catalyst after SCR treatment in comparison to that of V/TiO2 and V/CeO2 (Figure 9, right). The Ti 2p binding energies are between 458.4 and 459.1 eV in both Ti-containing catalysts and do not change after SCR treatment, indicating that Ti remains essentially tetravalent.52-53 This holds true also for cerium in the mixed oxide support of V/Ce0.5Ti0.5O2, yet a partial reduction of Ce4+ to Ce3+ under SCR conditions is seen for V/CeO2 indicated by a shoulder at 882.5 eV. This is particularly interesting since such reduction has not been observed under SCR feed in our recent study of V/Ce1-xZrxO2 catalysts11 in which, however, the CeO2 support was prepared by a citrate method. It indicates that the synthesis procedure might have a crucial impact on the redox properties and, thus the catalytic activity of V/CeO2 materials. This effect has been studied separately in more detail.83 The V/(Ce + Ti) surface ratio decreases slightly for all catalysts after SCR, suggesting diffusion of a minor amount of V into subsurface layers and/or an agglomeration of dispersed VOx sites. However these effects are almost negligible. More significant is the increase of the Ce/Ti ratio in the V/Ce0.5Ti0.5O2 catalyst which points to an enrichment of Ce on the surface.

CONCLUSIONS Incorporation of Ti into CeO2 leads to solid Ce1-xTixO2 solutions with smaller crystallite size and/or higher disorder that might improve oxygen mobility, reducibility and create catalytic activity in NH3-SCR already in the absence of VOx surface species. While CeO2 and TiO2 are


34

Page 35 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

poorly active even at 300 °C, Ce1-xTixO2 with x ≥ 0.3 show NO and NH3 conversions around 80 % at 200 °C. Deposition of vanadia on the surface of these supports boosts SCR activity tremendously. With the best 5 wt.% V2O5/Ce1-xTixO2 catalysts (x = 0.3-0.5) NO conversion and N2 selectivity of ≈100 % were reached already below 200 °C at a space velocity of 70000 h-1 and performance was still appreciable at a tenfold higher space velocity of 750000 h-1. Remarkably, no undesired formation of N2O was observed. Comprehensive studies of structure-reactivity relationships using in situ and operando spectroscopy revealed significant mechanistic differences. While on bare supports NH3-SCR proceeds after a Langmuir-Hinshelwood mechanism implying the reaction with adsorbed NH3 and nitrate species (formed on the surface by reaction of NO/O2), an Eley-Rideal mechanism operates on V-containing catalysts, in which adsorbed NH3 and NH4+ react with NO/O2 from the gas phase. The switch in reaction mechanism has its roots in structural differences of catalysts and supports. While NH3 adsorbs on both Lewis and Brønsted sites present in all samples (though to a different extent), nitrate is preferentially formed from NO/O2 by the participation of Ce-O moieties. Within the bare supports, the observed maximum catalytic activity of Ce1-xTixO2 (x = 0.3-0.5) can be explained by the poor formation of surface nitrates on Ce-free TiO2 and, despite being abundant, their low reactivity on CeO2. In V/Ce1-xTixO2 catalysts, Ce-O sites are obviously effectively covered by VOx species which hinders the formation of surface nitrates and causes the switch in the reaction mechanism. Without doubt, both Lewis and Brønsted surface sites support NH3 adsorption, yet the difference in their concentration does not correlate with the observed activity differences,


35

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 36 of 41

indicating that it is the redox properties of the materials that govern catalytic activity, i. e. the intrinsic oxygen mobility of the supports boosted by replacing Ce by redox-active Ti cations as well as the nature of VOx species in the V/Ce1-xTixO2 catalysts. V/TiO2 and, to a lesser extent, also V/CeO2 contain lower dispersed VOx species up to amorphous surface aggregates and even V2O5 nanoparticles, not all V sites of which might be accessible by reactant molecules. In contrast, a unique property of the most active V/Ce0.5Ti0.5O2 catalyst is the presence of highly dispersed vanadyl species connected via oxygen bridges to both Ce and Ti. These are highly redox-active. Obviously, the intimate contact with Ce/Ti renders the V species in their highest and most active equilibrium valence state +5. On the condition of a Mars-van Krevelen mechanism, which is commonly accepted for NH3-SCR28-29, 34, 38 and also supported by the results of in situ spectroscopy in this work, oxygen vacancies created upon reduction of V5+ by NH3 in the immediate vicinity of the resulting V4+ must be replenished during reoxidation of the latter. Possibly this does not happen directly by gas-phase O2 in the very same position but by a lattice oxygen in the immediate vicinity, whereby other vacancies at higher distances are created, that are then filled by gas-phase oxygen (Scheme 1). This would mean that supply and uptake of oxygen according to Mars and van Krevelen is not restricted to the outermost surface but also comprises deeper layers of the catalyst lattice. In such a case, a high efficiency of oxygen and electron transport through the catalysts lattice, being a speciality of ceria-based oxides, might boost the reaction rate.


36

Page 37 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Scheme 1. Tentative mechanism of LT-NH3-SCR on V/Ce1-xTixO2 catalysts In previous studies of vanadium phosphate catalysts during selective oxidation of hydrocarbons which also follow a Mars-van Krevelen mechanism, we have experimentally verified by in situ EPR that the catalyst bulk does indeed participate in oxygen transport.84 Therefore, we assume that it is the efficiency of this process which is responsible for the catalytic performance of our V/Ce1-xTixO2 (x = 0.3-0.5) catalysts which belongs to the best observed so far for low-temperature NH3-SCR. Apart from this, a high steady-state concentration of V5+ maintained under SCR conditions in the best catalysts might improve Lewis acidity and, thus, NH3 adsorption.


37

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 38 of 41

ASSOCIATED CONTENT Supporting Information. The supporting Information is available free of charge on the ACS Publications website at http://pubs.acs.org. XRD and XPS details, additional data (including EPR, UV-Vis, DRIFTS and H2-TPR analysis) of pure supports and supported vanadium catalysts. AUTHOR INFORMATION Corresponding Author * E-mail: [email protected] ACKNOWLEDGMENT The authors thank Dr. Jana Engeldinger, Mrs. Christine Rautenberg, and Mr. Reinhard Eckelt for experimental supports. Thanh Huyen Vuong acknowledges a grant of the Vietnamese Ministry of Education and Training. ABBREVIATIONS SCR selective catalytic reduction, EPR electron paramagnetic resonance, XPS X-ray photoelectron spectroscopy, XRD X-ray powder diffraction, UV-Vis-DRS UV-Visible diffuse reflectance spectroscopy, and DRIFTS diffuse reflectance infrared Fourier transform spectroscopy. REFERENCES 1. 2. 3. 4. 5.

Li, J.; Chang, H.; Ma, L.; Hao, J.; Yang, R. T., Catal. Today 2011, 175, 147-156. Liu, C.; Shi, J.-W.; Gao, C.; Niu, C., Appl. Catal., A 2016, 522, 54-69. Tang, C.; Zhang, H.; Dong, L., Catal. Sci. Technol. 2016, 6, 1248-1264. Tang, X.; Li, J.; Sun, L.; Hao, J., Appl. Catal., B 2010, 99, 156-162. Tang, X.; Hao, J.; Xu, W.; Li, J., Catal. Commun. 2007, 8, 329-334.


38

Page 39 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

6. Xiao, X.; Sheng, Z.; Yang, L.; Dong, F., Catal. Sci. Technol. 2016, 6, 1507-1514. 7. Liu, Z.; Yi, Y.; Zhang, S.; Zhu, T.; Zhu, J.; Wang, J., Catal. Today 2013, 216, 76-81. 8. Qi, G.; Yang, R. T., Chem. Commun. 2003, 848-849. 9. Shen, B.; Wang, Y.; Wang, F.; Liu, T., Chem. Eng. J. 2014, 236, 171-180. 10. Guo, R.-T.; Zhen, W.-L.; Zhou, Y.; Pan, W. G.; Xu, H.-J.; Jin, Q.; Ding, C.-G.; Guo, S.Y., Asian J. Chem. 2014, 26, 407-410. 11. Vuong, T. H.; Radnik, J.; Kondratenko, E.; Schneider, M.; Armbruster, U.; Brückner, A., Appl. Catal. B: Environ. 2016, 197, 159-167. 12. Gao, X.; Jiang, Y.; Zhong, Y.; Luo, Z.; Cen, K., J. Hazard. Mater. 2010, 174, 734-739. 13. Shan, W.; Liu, F.; He, H.; Shi, X.; Zhang, C., Catal. Today 2012, 184, 160-165. 14. Li, P.; Xin, Y.; Li, Q.; Wang, Z.; Zhang, Z.; Zheng, L., Environ. Sci. Technol. 2012, 46, 9600-9605. 15. Deng, Y.; Chen, X.; Shao, R.; Hu, L.; Tang, J.; Wang, C., Key Eng. Mater. 2016, 697, 275-278. 16. Gan, L.; Guo, F.; Yu, J.; Xu, G., Catalysts 2016, 6, 25. 17. Marberger, A.; Elsener, M.; Ferri, D.; Kröcher, O., Catalysts 2015, 5, 1704-1720. 18. Zuo, J.; Chen, Z.; Wang, F.; Yu, Y.; Wang, L.; Li, X., Ind. Eng. Chem. Res. 2014, 53, 2647-2655. 19. Granger, P.; Parvulescu, V. I., Chem. Rev. 2011, 111, 3155-3207. 20. Chen, H.-Y., Cu/Zeolite SCR Catalysts for Automotive Diesel NOx Emission Control. In Urea-SCR Technology for deNOx After Treatment of Diesel Exhausts, Nova, I.; Tronconi, E., Eds. Springer New York: 2014; pp 123-147. 21. Li, C.; Li, Q.; Lu, P.; Cui, H.; Zeng, G., Front. Environ. Sci. Eng. 2012, 6, 156-161. 22. Lian, Z.; Liu, F.; He, H., Catal. Sci. Technol. 2015, 5, 389-396. 23. Peng, Y.; Wang, C.; Li, J., Appl. Catal., B 2014, 144, 538-546. 24. Trovarelli, A., Catal. Rev. 1996, 38, 439-520. 25. Beckers, J.; Rothenberg, G., Green Chem. 2010, 12, 939-948. 26. Vivier, L.; Duprez, D., ChemSusChem 2010, 3, 654-678. 27. Beck, B.; Harth, M.; Hamilton, N. G.; Carrero, C.; Uhlrich, J. J.; Trunschke, A.; Shaikhutdinov, S.; Schubert, H.; Freund, H.-J.; Schlögl, R.; Sauer, J.; Schomäcker, R., J. Catal. 2012, 296, 120-131. 28. Tronconi, E.; Nova, I.; Ciardelli, C.; Chatterjee, D.; Weibel, M., J. Catal. 2007, 245, 110. 29. Busca, G.; Lietti, L.; Ramis, G.; Berti, F., Appl. Catal., B 1998, 18, 1-36. 30. Wachs, I. E., Dalton Trans. 2013, 42, 11762-11769. 31. Lian, Z.; Liu, F.; He, H., Ind. Eng. Chem. Res. 2014, 53, 19506-19511. 32. Liu, Z.; Zhang, S.; Li, J.; Zhu, J.; Ma, L., Appl. Catal., B 2014, 158–159, 11-19. 33. Cha, W.; Ehrman, S. H.; Jurng, J., J. Environ. Chem. Eng. 2016, 4, 556-563. 34. Ramis, G.; Yi, L.; Busca, G., Catal. Today 1996, 28, 373-380. 35. Vittadini, A.; Casarin, M.; Selloni, A., J. Phys. Chem. B 2005, 109, 1652-1655. 36. Inomata, M.; Miyamoto, A.; Murakami, Y., J. Catal. 1980, 62, 140-148. 37. Soyer, S.; Uzun, A.; Senkan, S.; Onal, I., Catal.Today 2006, 118, 268-278. 38. Topsoe, N. Y.; Dumesic, J. A.; Topsoe, H., J. Catal. 1995, 151, 241-252. 39. Wang, J.; Yan, Z.; Liu, L.; Chen, Y.; Zhang, Z.; Wang, X., Appl. Surf. Sci. 2014, 313, 660-669.


39

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 40 of 41

40. Centeno, M. A.; Carrizosa, I.; A. Odriozola, J., Phys. Chem. Chem. Phys. 1999, 1, 349354. 41. Centeno, M. A.; Carrizosa, I.; Odriozola, J. A., Appl. Catal., B 2001, 29, 307-314. 42. Boningari, T.; Koirala, R.; Smirniotis, P. G., Appl. Catal., B 2013, 140–141, 289-298. 43. Takagi, M.; Kawai, T.; Soma, M.; Onishi, T.; Tamaru, K., J. Catal. 1977, 50, 441-446. 44. Ramis, G.; Busca, G.; Bregani, F.; Forzatti*, P., Appl. Catal. 1990, 64, 259-278. 45. Topsoe, N. Y., Science 1994, 265, 1217-1219. 46. Reddy, B. M.; Lakshmanan, P.; Loridant, S.; Yamada, Y.; Kobayashi, T.; Lopez-Cartes, C.; Rojas, T. C.; Fernandez, A., J. Phys. Chem. B 2006, 110, 9140-9147. 47. Scofield, J. H., J. Electron Spectrosc. 1976, 8, 129-137. 48. Wachs, I. E.; Deo, G.; Weckhuysen, B. M.; Andreini, A.; Vuurman, M. A.; Boer, M. d.; Amiridis, M. D., J. Catal. 1996, 161, 211-221. 49. Yu, J.; Si, Z.; Zhu, M.; Wu, X.; Chen, L.; Weng, D.; Zou, J., RSC Adv. 2015, 5, 8359483599. 50. Hori, C. E.; Permana, H.; Ng, K. Y. S.; Brenner, A.; More, K.; Rahmoeller, K. M.; Belton, D., Appl. Catal., B 1998, 16, 105-117. 51. M. Alifanti, B. B., N. Blangenois, J. Naud, P. Grange, B. Delmon, Chem. Mater. 2003, 15, 395-403. 52. Watanabe, S.; Ma, X.; Song, C., J. Phys. Chem. C 2009, 113, 14249-14257. 53. Reddy, B. M.; Khan, A.; Yamada, Y.; Kobayashi, T.; Loridant, S.; Volta, J.-C., J. Phys. Chem. B 2003, 107, 5162-5167. 54. Bêche, E.; Charvin, P.; Perarnau, D.; Abanades, S.; Flamant, G., Surf. Interface Anal. 2008, 40, 264-267. 55. Reddy, B. M.; Reddy, G. K.; Ganesh, I.; Ferreira, J. M. F., J. Mater. Sci. 2009, 44, 27432751. 56. Reddy, B. M.; Lakshmanan, P.; Khan, A., J. Phys. Chem. B 2004, 108, 16855-16863. 57. Bensalem, A.; Bozon-Verduraz, F.; Delamar, M.; Bugli, G., Appl. Catal., A 1995, 121, 81-93. 58. Biesinger, M. C.; Lau, L. W. M.; Gerson, A. R.; Smart, R. S. C., Appl. Surf. Sci. 2010, 257, 887-898. 59. Moulder, J. F.; Stickle, W. F.; Sobol, P. E.; Bomben, K. D., Handbook of x-ray photoelectron spectroscopy : a reference book of standard spectra for identification and interpretation of XPS data. Eden Prairie, Minn. : Physical Electronics: 1995. 60. Rybarczyk, P.; Berndt, H.; Radnik, J.; Pohl, M. M.; Buyevskaya, O.; Baerns, M.; Brückner, A., J. Catal. 2001, 202, 45-58. 61. Che, M.; Tench, A. J., Characterization and Reactivity of Molecular Oxygen Species on Oxide Surfaces. In Adv. Catal., D.D. Eley, H. P.; Paul, B. W., Eds. Academic Press: 1983; Vol. Volume 32, pp 1-148. 62. Gao, X.; Wachs, I. E., J. Phys. Chem. B 2000, 104, 1261-1268. 63. Rodella, C. B.; Franco, R. W. A.; Magon, C. J.; Donoso, J. P.; Nunes, L. A. O.; Saeki, M. J.; Aegerter, M. A.; Sargentelli, V.; Florentino, A. O., J. Sol-Gel Sci. Technol. 2002, 25, 83-88. 64. Inoue, F.; Ando, R. A.; Corio, P., J. Raman Spectrosc. 2011, 42, 1379-1383. 65. Corma, A., Chem. Rev. 1995, 95, 559-614. 66. Hughes, T. R.; White, H. M., J. Phys. Chem. 1967, 71, 2192-2201. 67. Busca, G., Physi. Chem. Chem. Phys. 1999, 1, 723-736.


40

Page 41 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

68. Wu, Z.; Mann, A. K. P.; Li, M.; Overbury, S. H., J. Phys. Chem. C 2015, 119, 73407350. 69. Lercher, J. A.; Gründling, C.; Eder-Mirth, G., Catal. Today 1996, 27, 353-376. 70. Zhu, H.; Qin, Z.; Shan, W.; Shen, W.; Wang, J., J. Catal. 2004, 225, 267-277. 71. Gu, X.; Ge, J.; Zhang, H.; Auroux, A.; Shen, J., Thermochim. Acta 2006, 451, 84-93. 72. Davydov, A. A., Molecular spectroscopy of oxide catalyst surfaces. John Wiley & Son Ltd,: The Atrium, Southern Gate, Chichester, England, 2003. 73. Ruggeri, M. P.; Nova, I.; Tronconi, E.; Pihl, J. A.; Toops, T. J.; Partridge, W. P., Appl. Catal., B 2015, 166-167, 181-192. 74. Philipp, S.; Drochner, A.; Kunert, J.; Vogel, H.; Theis, J.; Lox, E. S., Top. Catal. 2004, 30/31, 235-238. 75. Martinez-Arias, A.; Soria, J.; Conesa, J. C.; Seoane, X. L.; Arcoya, A.; Cataluna, R., J. Chem. Soc., Faraday Trans. 1995, 91, 1679-1687. 76. Kantcheva, M., J. Catal. 2001, 204, 479-494. 77. Azambre, B.; Zenboury, L.; Koch, A.; Weber, J. V., J. Phys. Chem. C 2009, 113, 1328713299. 78. Nowińska, K.; Więckowski, A. B., Z. Phys. Chem. NF 1989, 231-244. 79. Risse, T.; Hollmann, D.; Bruckner, A., Chapter 1 In situ electron paramagnetic resonance (EPR) - a unique tool for analysing structure and reaction behaviour of paramagnetic sites in model and real catalysts. In Catalysis: Vol. 27, The Royal Society of Chemistry: 2015; Vol. 27, pp 1-32. 80. McGarvey, B. R., J. Phys. Chem. 1967, 71, 51-66. 81. Boucher, L. J.; Tynan, E. C.; Yen, T. F., Electron Spin Resonance of Metal Complexes. Plenum Press: New York, 1969. 82. Nanostructured Photocatalysts: Advanced Functional Materials. Springer International Publishing: Switzerland, 2016. 83. Vuong, T. H.; Radnik, J.; Schneider, M.; Atia, H.; Armbruster, U.; Brückner, A., Catal. Commun. 2016, 84, 171-174. 84. Brückner, A., Catal. Rev. 2003, 45, 97-150.

FOR TABLE OF CONTENT ONLY


41

Ce1âxTixO2 Catalysts for Low-Temperature NH3-SCR

Recommend Documents

Ce1âxTixO2 Catalysts for Low-Temperature NH3-SCR