Insight into Deactivation of Commercial SCR Catalyst by Arsenic: An

Nov 7, 2014 - Arsenic (As) is a serious poison for commercial SCR catalysts from stationary sources, where it is present as As2O3 in the gas phase of ...
1 downloads 0 Views 1MB Size
Download PDF
Article pubs.acs.org/est

Insight into Deactivation of Commercial SCR Catalyst by Arsenic: An Experiment and DFT Study Yue Peng,†,‡ Junhua Li,*,† Wenzhe Si,† Jinming Luo,‡ Qizhou Dai,‡ Xubiao Luo,‡ Xin Liu,† and Jiming Hao† †

Environ. Sci. Technol. 2014.48:13895-13900. Downloaded from pubs.acs.org by IOWA STATE UNIV on 01/10/19. For personal use only.

State Key Joint Laboratory of Environment Simulation and Pollution Control, School of Environment, Tsinghua University, Beijing, 100084, China ‡ School of Civil and Environmental Engineering, Georgia Institute of Technology, 800 West Peachtree Street, Suite 400 F-H, Atlanta, Georgia 30332-0595, United States S Supporting Information *

ABSTRACT: Fresh and arsenic-poisoned V2O5−WO3/TiO2 catalysts are investigated by experiments and DFT calculations for SCR activity and the deactivation mechanism. Poisoned catalyst (1.40% of arsenic) presents lower NO conversion and more N2O formation than fresh. Stream (5%) could further decrease the activity of poisoned catalyst above 350 °C. The deactivation is not attributed to the loss of surface area or phase transformation of TiO2 at a certain arsenic content, but due to the coverage of the V2O5 cluster and the decrease in the surface acidity: the number of Lewis acid sites and the stability of Brønsted acid sites. Large amounts of surface hydroxyl induced by H2O molecules provide more unreactive As−OH groups and give rise to a further decrease in the SCR activity. N2O is mainly from NH3 unselective oxidation at high temperatures since the reducibility of catalysts and the number of surface-active oxygens are improved by As2O5. Finally, the reaction pathway seems unchanged after poisoning: NH3 adsorbed on both Lewis and Brønsted acid sites is reactive.

■

INTRODUCTION Selective catalytic reduction (SCR) with ammonia is a wellproven technique for NO removal in coal-fired power plants and steel sintering boilers.1,2 In the SCR process, NO is reduced to N2 over a catalyst, which is usually V2O5 supported on TiO2 and works at ∼350−400 °C.3 The catalytic convertor is often placed right after the vessel to take advantage of the heat from hot exhaust gas; however, the major drawbacks of this arrangement are the facts that the catalytic convertor operates under high temperatures for a long period (thermal aging) and is exposed to high concentrations of fly ash and poisoning elements, including SO2, alkali/alkaline-earth metals, and heavy metals (arsenic and lead).4−7 To improve catalyst thermal aging resistance and lifetime, WO3 or MoO3 additives are introduced to V2O5/TiO2. These metal oxides enhance the catalyst activity and thermal durability by enhancing the number of Brønsted acid sites and stabilizing TiO2 from phase change (anatase to rutile), respectively. In addition, the catalyst is generally robust against SO2 poisoning above 300 °C.8−10 Previous studies proposed the deactivation mechanism of alkali metals: in addition to the decrease in the surface acidity, the loss of reducibility and the accumulation of inactive nitrite/nitrate species might also restrain catalyst activity.11,12 However, these metals are highly mobile and soluble in a water form, and water washing or electrophoresis methods are quite effective for regenerating the poisoned catalysts.13 Further, the © 2014 American Chemical Society

honeycomb monolithic catalyst shows good durability to dust and resistance to alkali metals. Arsenic (As) is a serious poison for commercial SCR catalysts from stationary sources, where it is present as As2O3 in the gas phase of power plants in concentrations between 1 μg/ m3 and 10 mg/m3.14,15 As2O3 diffuses into the catalyst surface and is adsorbed on both active (mainly V2O5) and nonactive sites (TiO2 support), resulting in the replacement of hydroxyl groups by newly formed As hydroxyls. When the catalyst is poisoned by sufficient amounts of As, As2O3 is oxidized to As2O5 by surface superoxide species, and neither CO nor NH3 is adsorbed on the Lewis acid sites of the poisoned catalyst.16−18 Gao et al. used Pb as a probe atom to reveal the deactivation on V2O5/TiO2 by density functional theory (DFT) calculations. The poisoning mechanism was attributed to the coverage of vanadium active sites by Pb and the loss of surface acidity.19,20 However, few works have been published to systematically elucidate the influences on surface acidity, reducibility, and reaction pathway for poisoned catalysts. The objective of this work is to clarify the poisoning mechanism of As on a commercial V2O5−WO3/TiO2 catalyst. Received: Revised: Accepted: Published: 13895

July 18, 2014 November 5, 2014 November 7, 2014 November 7, 2014 dx.doi.org/10.1021/es503486w | Environ. Sci. Technol. 2014, 48, 13895−13900

Environmental Science & Technology

Article

Table 1. Physical Parameters of Fresh and Poisoned Catalysts fresh poisoned a

SA (m2/g)

DPore (nm)

As contenta

As contentb

surface Asc

surface SO42− c

V contentb

W + Ti + Si contenta

52.7 52.0

113.9 115.4

0 1.51%

0 1.40%

0 1.82%

1.43% 1.42%

0.66% 0.53%

97.1% 95.4%

Acquired from XRF results. bAcquired from ICP results. cAcquired from XPS spectra.

■

plane.24,25 The TiO2 (001) model was constructed by cutting TiO2 anatase and a (3 × 3) supercell of the slab model. A vacuum gap of 15 Å was employed to separate subsequent slabs (Ti36O72). The slab thickness was optimized according to previous TiO2 (001) slab results, in which a 3-layer slab model with the bottom layer fixed to the bulk parameters is sufficient.26,27 For surface relaxation, no symmetry was used, and a dipole correction was included (SI Figure S1). All the calculations are based on DFT and performed using Material Studio 5.5 modeling DMol3.28,29 The doublenumerical-quality basis set with polarization functions and GGA-PBE were used for all calculations.30,31 Core electrons were treated with DFT semicore-pseudo potentials.32 Spin polarization was also applied, and the real space cutoff radius was maintained as 4.2 Å. The adsorption energy (Ead) of the NH3 molecule was calculated as follows:

EXPERIMENT AND CALCULATION Catalysts Preparation and Activity Measurement. The investigated monolith catalyst (fresh) was V2O5−WO3/TiO2 with SO42−, SiO2, and CaO (shown in Supporting Information, SI, Table S1). The deactivated catalyst (poisoned) was the same catalyst, which was obtained from a power plant in southwest China after running under actual working conditions. Both fresh and poisoned catalysts are crushed and sieved within 40−60 meshes for activity measurement. Activity was performed in a fixed-bed quartz reactor (5 mm i.d.) using 100 mg of catalyst. The feed gas contained 500 ppm of NO, 500 ppm of NH3, 3% O2, 5% H2O (when used) for SCR reactions, and 500 ppm of NH3, 3% O2 for NH3 oxidations. The balance was N2. The concentrations of NO, NO2, N2O, NH3, and H2O were continually monitored by an FTIR spectrometer (MultiGas TM 2030 FTIR). Catalysts Characterizations. The BET surface area was carried out with a Micromeritics ASAP 2020 apparatus. The element contents were characterized by both ICP with an IRIS Intrepid II XSP apparatus (Thermo Fisher Scientific Inc.) and XRF with an XRF DEX-LE from Shimadzu. X-ray photoelectron spectroscopy (XPS) was performed with an ESCALab220i-XL electron spectrometer from VG Scientific using 300 W Al Kα radiations. The binding energy was referenced to the C 1s line at 284.8 eV. Temperature-programmed reduction (TPR) of H2 was performed on a chemisorption analyzer (Micromeritics, ChemiSorb 2720 TPx) under 10% H2/Ar gas flow (50 mL/ min) at a rate of 10 °C/min up to 900 °C. Temperatureprogrammed desorption (TPD) of NH3 was performed in a fixed-bed quartz reactor. Before the test, each sample (100 mg) was pretreated in N2 (200 mL/min) gas at 350 °C. The sample was purged under NH3 at room temperature. After isothermal desorption under N2 gas flow (200 mL/min) at 100 °C, the temperature was elevated to 600 °C at a rate of 10 °C/min. The concentrations of NH3 were monitored by MultiGas TM 2030 FTIR. In-situ IR spectra were recorded on a Fourier transform infrared spectrometer (FTIR, Nicolet Nexus 870) equipped with the Harrick IR cell and an MCT detector cooled by liquid N2. The catalyst was first heated to 350 °C under N2 at a total flow rate of 100 mL/min for 1 h to remove adsorbed impurities. The background spectrum was collected in a flowing N2 atmosphere and subtracted from the sample spectra. The IR spectra were recorded by accumulating 32 scans at a resolution of 4 cm−1. Model Selection and DFT Calculation. Bulk V2O5 is also active and selective to produce N2, and the V6O20H10 cluster could be selected as the modeled surface. The cluster model exposing the V2O5 (010) plane was used to investigate the reaction mechanism of the SCR reaction and NH 3 oxidation.21−23 The model was constructed by cutting the V6O20 moiety from a V2O5 (010) surface, and the dangling bonds were saturated by using H atoms as terminators. To further study the interaction between As and V on support, we also calculated V2O5 and As2O5 doped on the TiO2 (001)

Ead = Esurface + Eammonia − Eammonia/surface

(1)

Esurface is the energy of the surface, Eammonia is the energy of an isolated NH3 molecule, and Eammonia/surface is the total energy of the same molecule adsorbed on the surface. Note that a positive value for Ead suggests a stable adsorption. The net charge of atoms during NH3 adsorption is calculated as follows: net charge = ESP chargeafter − ESP charge before

(2)

ESP chargeafter is the charge of atoms after NH3 is adsorbed, and ESP chargebefore is the charge of atoms before adsorption or gaseous NH3 molecules. Note that a positive value for net charge suggests the loss of electrons.

■

RESULTS AND DISCUSSION Catalyst Structure and SCR Performance. Table 1 summarizes the properties of fresh and poisoned catalysts. Compared with the poisoned catalyst, the BET surface area of fresh decreased slightly, and the average pore diameter was improved, which is attributed to the inevitable thermal aging effect. Nevertheless, the differences are not apparent. It seems that the change in the surface area or pore diameter does not play a key role during the poisoning process. The As loading is a little higher on the surface than that in the bulk, and only As5+ cations are found in XPS spectra (SI Figure S2), implying the enrichment of As2O5 on the catalyst surface.16 Other additive remnants during the catalyst procedure, especially alkali metals, were also observed by XRF (SI Table S1). The contents of K2O, Na2O, and CaO of the two samples are nearly the same. Therefore, the influence of alkali metals on the catalyst activity can be neglected in this study, and the catalyst can be defined as merely poisoned by As. The crystal structures of the catalysts are obtained by XRD and Raman spectra (SI Figure S3, S4). Only anatase phase TiO2 is present in both the XRD and Raman spectra, and both the W−O−W (794 cm−1) and WO (980 cm−1) vibrational modes are observed on the Raman spectra.33,34 The results indicate that As2O5 at a certain amount (1.40% in the bulk and 1.82% on the surface) does not change the crystal phase of TiO2 anatase or surface WO3 species. 13896

dx.doi.org/10.1021/es503486w | Environ. Sci. Technol. 2014, 48, 13895−13900

Environmental Science & Technology

Article

conversion of catalysts under a higher GHSV (240 000 mL/g· h) in 5% stream. Catalytic activity decreases at higher GHSV compared with Figure 1a, and H2O further restrains NO conversion above 300 °C. Although the activity of fresh catalyst under stream is not severe, the activity of a poisoned sample significantly decreases: only 35% of NO conversion is observed at 350 °C, compared with 55% of NO conversion for fresh catalyst under the same conditions. The results reveal there might be a further inhibition effect between As and stream at relatively high temperatures. Therefore, chemical characterizations and DFT calculations were then employed to elucidate the loss of SCR activity, the increase in N2O formation, and the further deactivation under H2O by arsenic poisoning. N2O formations were also further studied for different poisoned catalysts and for different test conditions (SI Figure S5). The results show that the increase in N2O during the SCR process is from arsenic poisoning at high temperatures. Reducibility and Surface Acidity. The XPS spectra of O 1s for fresh and poisoned catalysts are shown in Figure 2a. The

Figure 1a shows the differences in NO conversion and N2O formation between fresh and poisoned catalysts under a gas

Figure 1. Comparisons on the SCR performance of fresh and poisoned catalysts. Reaction conditions: catalyst mass = 100 mg, (a) [NO] = 500 ppm, [NH3] = 500 ppm, [O2] = 3%, total flow rate = 200 mL/min, GHSV = 120 000 mL/g·h, (b) [NH3] = 500 ppm, [O2] = 3%, total flow rate = 200 mL/min, GHSV = 120 000 mL/g·h at 450 °C, (c) [NO] = 500 ppm, [NH3] = 500 ppm, [O2] = 3%, [H2O] = 5%, total flow rate = 400 mL/min, GHSV = 240 000 mL/g·h.

Figure 2. (a) XPS spectra of O 1s and (b) H2-TPR profiles of fresh and poisoned catalysts.

O 1s bands is fitted into two independent peaks, referred to as the stable oxygen in lattice at 530.2 eV (Oβ) and the surfaceactive oxygen at 531.9 eV (Oα),39,40 where Oα is more active in oxidations than Oβ as a result of its higher mobility. The fractions of Oα are 20.1% and 23.7% for fresh and poisoned, respectively, suggesting that surface-active oxygen species could be improved by As. Wei et al. studied the role of surface oxygen on TiO2 anatase (101) by DFT calculation and proposed that O2 could easily bond to the TiO2 surface, creating superoxide or hydroperoxyl radical species with As5+ cations.18 Figure 2b presents the H2-TPR profiles of fresh and poisoned catalysts in the range of 200−900 °C. Two reduction peaks are obtained for the fresh catalyst. The peak at 494 °C can be attributed to the reduction of V5+ to V0 and W6+ to W4+; a peak at 792 °C can be attributed to the reduction of W4+ to W0; and a sharp peak for poisoned catalyst at 416 °C, which could be assigned to the reduction of As5+ to As0.37,41 The onset reduction temperature shifts to lower temperature for the poisoned catalyst, and peaks of tungsten reduction are nearly unaffected (792 °C) or are overlapped (494 °C) by the large reduction peak. Surface As5+ cations increase the reducibility and the number of active surface oxygens. This might be one of the factors for the improvement of N2O formation. The NH3-TPD profiles of fresh and poisoned catalysts were performed below 600 °C (SI Figure S6). Relatively weak (266

hourly space velocity (GHSV) of 120 000 mL/g·h. Fresh catalyst shows higher activity than poisoned catalyst: nearly 85% of NO conversion is acquired for fresh at 450 °C, poisoned catalyst yields only 60% of NO conversion. The N2O concentrations for poisoned sample are higher and increase fast above 400 °C. As an important trace gas, N2O could cause global warming and stratospheric ozone depletion.35 To study the origin of N2O, NH3 oxidation was studied at 450 °C, with 500 ppm of the inlet NH3 (Figure 1b). NO2 was not detected over the two catalysts, and N2 concentrations from NH3 oxidation were calculated by subtracting the concentrations of N2O and NO from the inlet NH3. Poisoned catalyst yields the formation of more N2, N2O, and NO than fresh catalyst. The N2O originates mainly from unselective NH3 oxidation under O2-rich flow at high temperatures36 and could be improved by As poisoning, that is, As2O5 not only reduces the SCR activity but also generates more N2O. Stream is always present in the flue gas, and the catalyst activity temporarily decreases from the competitive adsorption of H2O with NH3 on surface acid sites. Once H2O is removed, the activity will be enhanced.37,38 Figure 1c presents the NO 13897

dx.doi.org/10.1021/es503486w | Environ. Sci. Technol. 2014, 48, 13895−13900

Environmental Science & Technology

Article

cm−1) at 300 °C, but with respect to the fresh catalyst, the band of Lewis acidity (1603 cm−1) is still observed at 350 °C. The thermal stability of the Brønsted acid sites under 300 °C is slightly enhanced by As2O5, but they vanish at 350 °C. Combined with NH3-TPD results, the stability of the surface acid sites could be inferred. There are significantly fewer Lewis acid sites after arsenic poisoning, and parts of the surface As atoms provide weak Brønsted acid sites. The strength of the Lewis acid sites at both the low and high temperature decreases, that is, the adsorptions of NH3 on VO groups become weak, whereas the stability of the Brønsted acid sites slightly increases on As−OH groups at low temperature but rapidly declines above 300 °C. Furthermore, a weak peak at 1539 cm−1 occurs only on the poisoned catalyst at 200−350 °C. This peak could be attributed to the δ(NH2−) mode,42 which could serve as a probe for the oxidation ability of catalysts, formed by reaction between adsorbed NH3 and surface-active oxygen above 200 °C. The formation of this species shows good agreement with the improved oxidation ability of As2O5 on the catalyst surface. DFT Calculations. The optimized structures of acid sites, poisoned sites, and the corresponding adsorptions of NH3 molecule on V2O5 cluster are displayed in Figure 4, and the Ead and net charges of the surface atoms are summarized in Table 2. Both Lewis and Brønsted acid sites are stably constructed as

°C) and strong (360 °C) acidities can be identified on the TPD curves on the basis of the temperature regions. The total acidity apparently decreases after poisoning, including part of the weak acidity and most of the strong acidity. To study the types of surface acid sites, in situ FTIR spectra of NH3 adsorption/ desorption behaviors were carried out over the two catalysts from 100 to 350 °C (Figure 3). Both Lewis and Brønsted acid

Figure 3. In situ FTIR spectra of NH3-TPD of (a) fresh and (b) poisoned catalysts.

Table 2. Ead and Net Charges (ESP charge) during the NH3 Adsorption Based on Figure 4

sites stably bond to the surface of the catalyst at 100 °C. The Lewis acid sites displayed broad bands in the 1150−1300 cm−1 as the δs(NH3) mode and a relatively weak peak at 1603 cm−1 (fresh) or 1607 cm−1 (poisoned) as the δas(NH3) mode, whereas Brønsted acid sites present a broad band at 1390− 1480 cm−1 as the δas(NH4+) mode and a peak at 1670 cm−1 as the δs(NH4+) mode.2,42−44 To diminish the influence of IR absorbance, the absorbance intensity was set to 3.00 at 100 °C by calibrating the angle of incident light when collecting the background spectra. The Brønsted acid sites seemed to be improved after As doping at 100 °C, whereas the Lewis acid sites decreased by comparing the intensity of the peaks at 1603 and 1607 cm−1. Arsenic atoms block the surface Lewis acid sites (VO) and provide new Brønsted acid sites (As−OH) of vanadia-based catalyst. Previous studies also ascertained the formation of As−OH groups for the WO3(MoO3)/TiO2 catalyst.17 However, whether As−OH groups are active has not been demonstrated. With an increase in the temperature, the Lewis acid sites of the poisoned catalyst decrease rapidly and nearly disappear (1607

freshL Ead (eV) net charge of As net charge of N net charge of Oa a

freshB

poisonedL

poisonedB

0.67

2.13

−0.26

0.14

1.17 0.20 0.01 −0.08

3.37 0.29 0.58 −0.02

The O atom bond between As and V.

previous reports by other groups;19,21,22 the bond length of V O and V−O−V are 1.6 and 1.8 Å, respectively; and the O−H distance is nearly 1.0 Å. Once the As atom was doped on the Lewis acid sites (Figure 4c), it covers the regions formed by four VO groups, and the As−O bond length becomes longer (3.1 Å). With respect to Brønsted acid sites, As slightly moves against the H atom. Figure 4e,f shows NH3 adsorptions on fresh models. The Ead is 0.67 and 2.13 eV for Lewis and Brønsted acid sites, respectively. By checking the net charges of V and N atoms, the adsorption of NH3 on Lewis acid sites is

Figure 4. Optimized V2O5 cluster structures of (a) Lewis acid sites, (b) Brønsted acid sites, (c) poisoned Lewis acid sites, and (d) poisoned Brønsted acid sites and (e−h) the corresponding NH3 adsorption configurations. Vanadium atoms are green, oxygen atoms are red, arsenic atoms are purple, nitrogen atoms are blue, and hydrogen atoms are white. 13898

dx.doi.org/10.1021/es503486w | Environ. Sci. Technol. 2014, 48, 13895−13900

Environmental Science & Technology

Article

OH groups are produced, and the deep inhibition effect occurs. N2O is generated from the unselective oxidation of NH3 resulting from the improvement of reducibility and surfaceactive oxygen by As. Reaction mechanism over both catalysts was carried out by transient in situ FTIR spectra at 250 and 350 °C (SI Figure S7−S9). The key pathways (Eley−Rideal) and intermediate species were not changed over the poisoned catalyst compared with fresh V2O5−WO3/TiO2: NH3 adsorbed on both Lewis and Brønsted acid sites were active with gaseous NO above 250 °C. Arsenic improved the NH3 oxidation and produced more N2O at high temperatures.

weak, and some electrons (0.14) transfer from N to H atoms as a covalent band on the Brønsted acid sites. When the NH3 molecule is adsorbed onto poisoned Lewis acid sites (Figure 4e), a larger Ead is obtained, indicating a more stable adsorbed species, but the net charge of N is only 0.01, and the net charge of As is 0.20, indicating the adsorption of NH3 on an As atom is unfavorable and some electrons of the As shift to the O atom (−0.08). For the NH3 adsorption on Brønsted acid sites (Figure 4f), the Ead for NH3 adsorption is higher (3.37 eV) than that on clean Brønsted acid sites, and the net charge of N was 0.58, which is larger than the adsorption structure (0.14) in Figure 4f. The NH3 adsorption on Brønsted acid sites could be strengthened by the As atom. The results show good agreement with the data of in situ FTIR spectra. Figure 5 gives the optimized structure of the V2O5 and As2O5 cluster on the TiO2 (001) plane. The vanadia dimer (V/Ti) is

■

ASSOCIATED CONTENT

S Supporting Information *

Related table and figures. This material is available free of charge via the Internet at http://pubs.acs.org/.

■

AUTHOR INFORMATION

Corresponding Author

*Tel.: +86 10 62771093; Fax.: +86 10 62771093. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors gratefully acknowledge financial support from the National Natural Science Foundation of China (21325731, 21407088, and 21221004), National High-Tech Research and the Development (863) Program of China (2013AA065401), and the International Postdoctoral Exchange Fellowship Program of China (20130032). The authors appreciate support from the Brook Byers Institute for Sustainable Systems, Hightower Chair, and Georgia Research Alliance at Georgia Institute of Technology and CPI YUANDA EnvironmentalProtection Engineering Co., Ltd.

Figure 5. Optimized structures of V2O5, As2O5 clusters doped on the TiO2 (001) plane, and the PDOS of As and V orbitals on the slab models.

modeled as having a 4-fold coordination (three V−O bonds and a VO bond) with two adjacent vanadium sites in a configuration similar to that found on the (010) plane of bulk V2O5 and the V6O20 cluster. The As2O5 cluster stably bonds to the (001) plane (As/Ti) with the same coordination as with V2O5. Further, As2O5 and V2O5 bond together atop the (001) plane (AsVTi), indicating the two clusters have similar bond lengths and structures. To study the influence of As on the V2O5 cluster, we have calculated the project density of states (PDOS) of As and V orbitals for the three models. The valence band of As in AsV/Ti is nearly unchanged compared with that in As/Ti, whereas the bottom of the conduction band slightly shifts to a lower energy region (at 3.2 eV). These results suggest that V2O5 improves the reactivity of surface As2O5 and agrees with the TPR results. Therefore, As doped on a vanadiabased catalyst provides new surface acid sites, and As2O5 itself is promoted by V2O5 for its reactivity on the TiO2 surface. Deactivation Mechanism of Arsenic. The physical properties of the poisoned catalyst, including the surface area, pore diameter, and crystal phase are similar to those of a fresh catalyst at certain As contents. However, significant changes of acidity and reducibility by As2O5 hinder the NO conversions and improve the N2O formations. The loss of SCR activity is due to the decrease in the number of Lewis acid sites and unstable Brønsted acid sites (As−OH) at high temperatures. The cosuppression of As and H2O above 300 °C is attributed to the competitive adsorption of H2O on the surface As2O5. Once H2O bonds to acid sites close to the As atom, more As−

■

REFERENCES

(1) Ramis, G.; Bregani, F. Fourier transform-infrared study of the adsorption and coadsorption of nitric oxide, nitrogen dioxide and ammonia on vanadia−titania and mechanism of selective catalytic reduction. Appl. Catal. 1990, 64, 259−278. (2) Topsoe, N. Mechanism of the selective catalytic reduction of nitric oxide by ammonia elucidated by in situ on-line Fourier transform infrared spectroscopy. Science 1994, 265 (5176), 1217−9. (3) Busca, G.; Lietti, L.; Ramis, G.; Berti, F. Chemical and mechanistic aspects of the selective catalytic reduction of NOx by ammonia over oxide catalysts: A review. Appl. Catal., B 1998, 18 (1− 2), 1−36. (4) Chen, J.; Buzanowski, M.; Yang, R.; Cichanowicz, J. Deactivation of the vanadia catalyst in the selective catalytic reduction process. J. Air Waste Manage. Assoc. 1990, 40 (10), 1403−1409. (5) Zheng, Y.; Jensen, A.; Johnsson, J. Deactivation of V2O5−WO3− TiO2 SCR catalyst at a biomass-fired combined heat and power plant. Appl. Catal., B 2005, 60 (3−4), 253−264. (6) Kröcher, O.; Elsener, M. Chemical deactivation of V2O5/WO3− TiO2 SCR catalysts by additives and impurities from fuels, lubrication oils, and urea solution. Appl. Catal., B 2008, 77 (3−4), 215−227. (7) Nicosia, D.; Czekaj, I.; Kröcher, O. Chemical deactivation of V2O5/WO3−TiO2 SCR catalysts by additives and impurities from fuels, lubrication oils and urea solution. Appl. Catal., B 2008, 77 (3−4), 228−236. (8) Ciambelli, P.; Fortuna, M. E.; Sannino, D.; Baldacci, A. The influence of sulphate on the catalytic properties of V2O5−TiO2 and 13899

dx.doi.org/10.1021/es503486w | Environ. Sci. Technol. 2014, 48, 13895−13900

Environmental Science & Technology

Article

WO3−TiO2 in the reduction of nitric oxide with ammonia. Catal. Today 1996, 29 (1−4), 161−164. (9) Amiridis, M.; Duevel, R.; Wachs, I. The effect of metal oxide additives on the activity of V2O5/TiO2 catalysts for the selective catalytic reduction of nitric oxide by ammonia. Appl. Catal., B 1999, 20 (2), 111−122. (10) Sun, C.; Dong, L.; Yu, W.; Liu, L.; Li, H.; Gao, F.; Dong, L.; Chen, Y. Promotion effect of tungsten oxide on SCR of NO with NH3 for the V2O5−WO3/Ti0.5Sn0.5O2 catalyst: Experiments combined with DFT calculations. J. Mol. Catal. A: Chem. 2011, 346 (1−2), 29−38. (11) Peng, Y.; Li, J.; Chen, L.; Chen, J.; Han, J.; Zhang, H.; Han, W. Alkali metal poisoning of a CeO2−WO3 catalyst used in the selective catalytic reduction of NOx with NH3: An experimental and theoretical study. Environ. Sci. Technol. 2012, 46 (5), 2864−2869. (12) Peng, Y.; Li, J.; Huang, X.; Li, X.; Su, W.; Sun, X.; Wang, D.; Hao, J. Deactivation mechanism of potassium on the V2O5/CeO2 catalysts for SCR reaction: acidity, reducibility and adsorbed NOx. Environ. Sci. Technol. 2014, 48 (8), 4515−4520. (13) Peng, Y.; Li, J.; Shi, W.; Xu, J.; Hao, J. Design strategies for development of SCR catalyst: improvement of alkali poisoning resistance and novel regeneration method. Environ. Sci. Technol. 2012, 46 (22), 12623−12629. (14) Valdés, T.; Marbán, G.; Fuertes, A. Low-temperature SCR of NOx with NH3 over carbon-ceramic supported catalysts. Appl. Catal., B 2003, 46 (2), 261−271. (15) Senior, C.; Lignell, D.; Sarofim, A.; Mehta, A. Modeling arsenic partitioning in coal-fired power plants. Combust. Flame 2006, 147 (3), 209−221. (16) Lange, F.; Schmelz, H.; Knözinger, H. Infrared-spectroscopic investigations of selective catalytic reduction catalysts poisoned with arsenic oxide. Appl. Catal., B 1996, 8 (2), 245−265. (17) Hums, E. Is advanced SCR technology at a standstill? A provocation for the academic community and catalyst manufacturers. Catal. Today 1998, 42 (1−2), 25−35. (18) Wei, Z.; Zhang, S.; Pan, Z.; Liu, Y. Theoretical studies of arsenite adsorption and its oxidation mechanism on a perfect TiO2 anatase (101) surface. Appl. Surf. Sci. 2011, 258 (3), 1192−1198. (19) Gao, X.; Du, X.; Fu, Y.; Mao, J.; Luo, Z.; Ni, M.; Cen, K. Theoretical and experimental study on the deactivation of V2O5-based catalyst by lead for selective catalytic reduction of nitric oxides. Catal. Today 2011, 175 (1), 625−630. (20) Jiang, Y.; Gao, X.; Zhang, Y.; Wu, W.; Song, H.; Luo, Z.; Cen, K. Effects of PbCl2 on selective catalytic reduction of NO with NH3 over vanadia-based catalysts. J. Hazard. Mater. 2014, 274 (0), 270−278. (21) Yuan, R.; Fu, G.; Xu, X.; Wan, H. Bronsted-NH4+ mechanism versus nitrite mechanism: New insight into the selective catalytic reduction of NO by NH3. Phys. Chem. Chem. Phys. 2011, 13 (2), 453− 460. (22) Yuan, R.; Fu, G.; Xu, X.; Wan, H. Mechanisms for selective catalytic oxidation of ammonia over vanadium oxides. J. Phys. Chem. C 2011, 115 (43), 21218−21229. (23) Witko, M.; Grybos, R.; Tokarz-Sobieraj, R. Heterogeneity of V2O5(010) surfaces−the role of alkali metal dopants. Top. Catal. 2006, 38 (1−3), 105−115. (24) Calatayud, M.; Mguig, B.; Minot, C. A DFT study on the hydrated V2O5−TiO2-anatase catalyst: Stability of monomeric species. Theor. Chem. Acc. 2005, 114 (1−3), 29−37. (25) Calatayud, M.; Minot, C. Reactivity of the V2O5−TiO2-anatase catalyst: Role of the oxygen sites. Top. Catal. 2006, 41 (1−4), 17−26. (26) Suarez Negreira, A.; Wilcox, J. DFT Study of Hg oxidation across vanadia−titania SCR catalyst under flue gas conditions. J. Phys. Chem. C 2012, 117 (4), 1761−1772. (27) Suarez Negreira, A.; Wilcox, J. Role of WO3 in the Hg oxidation across the V2O5−WO3−TiO2 SCR catalyst: A DFT study. J. Phys. Chem. C 2013, 117 (46), 24397−24406. (28) Becke, A. Density functional exchange-energy approximation with correct asymptotic behavior. Phys. Rev. A 1988, 38 (6), 3098− 3100.

(29) Delley, B. An all-electron numerical method for solving the local density functional for polyatomic molecules. J. Chem. Phys. 1990, 92 (1), 508−517. (30) Perdew, J.; Burke, K.; Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 1996, 77 (18), 3865− 3868. (31) Delley, B. From molecules to solids with the DMol3 approach. J. Chem. Phys. 2000, 113, 7756−7764. (32) Delley, B. Hardness conserving semilocal pseudopotentials. Phys. Rev. B 2002, 66 (15), 155125. (33) Bañares, M.; Wachs, I. Molecular structures of supported metal oxide catalysts under different environments. J. Raman Spectrosc. 2002, 33 (5), 359−380. (34) Wachs, I.; Roberts, C. Monitoring surface metal oxide catalytic active sites with Raman spectroscopy. Chem. Soc. Rev. 2010, 39 (12), 5002−5017. (35) Kobayashi, M.; Hagi, M. V2O5−WO3/TiO2−SiO2−SO42− catalysts: Influence of active components and supports on activities in the selective catalytic reduction of NO by NH3 and in the oxidation of SO2. Appl. Catal., B 2006, 63 (1−2), 104−113. (36) Yang, S.; Xiong, S.; Liao, Y.; Xiao, X.; Qi, F.; Peng, Y.; Fu, Y.; Shan, W.; Li, J. Mechanism of N2O formation during the lowtemperature selective catalytic reduction of NO with NH3 over Mn− Fe spinel. Environ. Sci. Technol. 2014, 48 (17), 10354−10362. (37) Giakoumelou, I.; Fountzoula, C.; Kordulis, C.; Boghosian, S. Molecular structure and catalytic activity of V2O5/TiO2 catalysts for the SCR of NO by NH3: In situ Raman spectra in the presence of O2, NH3, NO, H2, H2O, and SO2. J. Catal. 2006, 239 (1), 1−12. (38) Huang, Z.; Liu, Z.; Zhang, X.; Liu, Q. Inhibition effect of H2O on V2O5/AC catalyst for catalytic reduction of NO with NH3 at low temperature. Appl. Catal., B 2006, 63 (3−4), 260−265. (39) Peng, Y.; Wang, C.; Li, J. Structure−activity relationship of VOx/CeO2 nanorod for NO removal with ammonia. Appl. Catal., B 2014, 144 (0), 538−546. (40) Liu, F.; Asakura, K.; He, H.; Shan, W.; Shi, X.; Zhang, C. Influence of sulfation on iron titanate catalyst for the selective catalytic reduction of NOx with NH3. Appl. Catal., B 2011, 103 (3−4), 369− 377. (41) Wang, C.; Yang, S.; Chang, H.; Peng, Y.; Li, J. Dispersion of tungsten oxide on SCR performance of V2O5−WO3/TiO2: Acidity, surface species and catalytic activity. Chem. Eng. J. 2013, 225 (0), 520− 527. (42) Hadjiivanov, K. Identification of neutral and charged NxOy surface species by IR spectroscopy. Catal. Rev. 2000, 42 (1), 71−144. (43) Peña, D.; Uphade, B.; Reddy, E.; Smirniotis, P. Identification of surface species on titania-supported manganese, chromium, and copper oxide low-temperature SCR catalysts. J. Phys. Chem. B 2004, 108 (28), 9927−9936. (44) Topsoe, N. Vanadia/titania catalysts for selective catalytic reduction (SCR) of nitric oxide by ammonia i. combined temperatureprogrammed in situ FTIR and on-line mass spectroscopy studies. J. Catal. 1995, 151 (1), 226−240.

13900

dx.doi.org/10.1021/es503486w | Environ. Sci. Technol. 2014, 48, 13895−13900