Article pubs.acs.org/est

Adsorption and Oxidation of Elemental Mercury over Ce-MnOx/TiPILCs Chuan He, Boxiong Shen,* Jianhong Chen, and Ji Cai College of Environmental Science and Engineering, Nankai University, Tianjin 300071, P. R. China S Supporting Information *

ABSTRACT: A series of innovative Ce−Mn/Ti-pillared-clay (Ce−Mn/Ti-PILC) catalysts combining the advantages of PILCs and Ce−Mn were investigated for elemental mercury (Hg0) capture at 100−350 °C in the absence of HCl in the flue gas. The fresh and used catalysts were characterized by scanning electron microscopy (SEM), nitrogen adsorption− desorption, X-ray diffraction (XRD), and X-ray photoelectron spectroscopy (XPS). The catalyst characterization indicated that the 6%Ce-6%MnOx/Ti-PILC catalyst possessed a large specific surface area and high dispersion of Ce and Mn on the surface. The experimental results indicated that the 6%Ce-6% MnOx/Ti-PILC catalyst exhibited high Hg0 capture (>90%) at 100−350 °C. During the first stage of the reaction, the main Hg0 capture mechanism for the catalyst was adsorption. As the reaction proceeded, the Hg0 oxidation ability was substantially enhanced. Both the hydroxyl oxygen and the lattice oxygen on the surface of the catalysts participated in Hg0 oxidation. At a low temperature (150 °C), the hydroxyl oxygen and lattice oxygen from Ce4+ → Ce3+ and Mn3+ → Mn2+ on the surface contributed to Hg0 oxidation. However, at a high temperature (250 °C), the hydroxyl oxygen and lattice oxygen from Mn4+ → Mn3+ contributed to Hg0 oxidation. Hg0 oxidation was preferred at a high temperature. The 6%Ce-6%MnOx/Ti-PILC catalyst was demonstrated to a good Hg0 adsorbent and catalytic oxidant in the absence of HCl in the flue gas.

■

oxidation would be achieved over these catalysts in flue gas containing no or low concentrations of HCl, especially in the flue gas after dry desulfurization. Therefore, a more effective catalyst for Hg0 oxidation without HCl in the flue gas is required. Among the catalysts for Hg0 oxidation without HCl, manganese oxides are the preferred ones.11 It was believed that Hg0 can be oxidized by MnOx to form HgO.14,15 Yan and co-workers11,16 reported that the Hg0 oxidation efficiency over MnOx/alumina and modified MnOx/alumina catalysts was more than 90%, and Mn4+ species were the most active components among the various manganese valence states. In recent years, CeO2 has been extensively studied as an oxygen provider for heterogeneous catalytic reactions due to its large oxygen storage capacity and the ability to shift between CeO2 and Ce2O3 under different conditions.17−20 Cerium oxides provide oxygen for manganese oxides at low temperatures, which improved the oxidation activity of manganese.17,18 However, a detailed mechanism for the Hg0 oxidation over

INTRODUCTION Mercury is a volatile and persistent pollutant that has various adverse effects on human health and has attracted considerable worldwide attention.1 The Mercury and Air Toxics Standard (MATS) announced in 2011 by the U.S. Environmental Protection Agency (EPA) emphasized mercury emission control from power plant.2 Coal-fired boilers are considered to be one of the largest anthropogenic sources of mercury emissions.3 The mercury in coal combustion flue gas exists in three forms: elemental mercury (Hg0), oxidized mercury (Hgoxi), and particle-bound mercury (Hgp).4 Oxidized mercury and particle-bound mercury can be easily removed by a wet flue gas desulfurization (WFGD) system and electrostatic precipitator or baghouses.5 However, elemental mercury is difficult to capture by typical air pollution control devices (APCDs).6 Activated carbon and modified activated carbon are the most common adsorbents for elemental mercury removal. However, large consumption of carbon-based adsorbents is greatly restricted by the operation cost because of the adsorption saturation by carbon materials.5−8 Some metal oxides including selective catalytic reduction (SCR) catalysts have been extensively investigated as oxidants for elemental mercury.9−12 It is known that the presence of HCl could substantially enhance the Hg0 conversion efficiency where the general mechanism involves a Deacon process.5,13 However, less Hg0 © 2014 American Chemical Society

Received: Revised: Accepted: Published: 7891

February 14, 2014 May 13, 2014 June 23, 2014 June 23, 2014 dx.doi.org/10.1021/es5007719 | Environ. Sci. Technol. 2014, 48, 7891−7898

Environmental Science & Technology

Article

pipelines warm (120 °C), the condensation of mercury on the surface of silicone pipelines was less than 2%. To identify mercury speciation in the outlet flue gas, a gas washing bottle containing a 10% SnCl2 aqueous solution or 10% KCl aqueous solution was set in front of the mercury analyzer as a mercury speciation conversion system when necessary. The total concentration of mercury (HgT) was measured with the flue gas passing through a SnCl2 solution (10%) to reduce the oxidized mercury in the gas into Hg0. The concentration of oxidized mercury was determined by the difference in the total concentration of mercury HgT and the concentration of Hg0 (by passing through a 10% KCl aqueous solution). In all of the experiments, Hg0 is the only mercury in the inlet flue gas, and the elemental mercury and oxidized mercury in the outlet flue gas coexists in the outlet flue gas. The total Hg0 removal efficiency (ET), the Hg0 oxidation efficiency (Eoxi) and the Hg0 adsorption efficiency (Eads) can be defined as follows:

Ce−Mn mixed-oxide catalysts in the absence of HCl is still needed to study. TiO2 and γ-Al2O3 supported MnOx-CeO2 mixed-oxide showed high Hg0 removal efficiency at low temperature in the presence of HCl. However, in the absence of HCl, Hg0 oxidation efficiency decreased greatly for these materials.15,21,22 The pillared clays (PILCs) are two-dimensional zeolite-like materials that can be employed as adsorbents or catalyst supports due to their large specific surface area.23−28 Compared to TiO2, γ-Al2O3 and carbon based materials, the modified PILCs can be considered as a promising material for Hg0 removal in a real power plant due to its low operation cost. To the best of our knowledge, no reports on Ce−Mn mixed-oxide loaded on PILCs for the oxidization of Hg0 in flue gas have been published. Therefore, an innovative material that combines the advantages of PILCs and Ce−Mn is investigated for Hg0 adsorption and oxidation in the absence of HCl in the flue gas. In this work, TiO2 pillared bentonite was used as a support to synthesize Ce−Mn mixed-oxide modified titania-PILC (CeMnOx/Ti-PILC). The Hg0 removal tests were performed with a simulated flue gas at 100−350 °C in the absence of HCl. The adsorption and catalytic oxidation mechanisms for Hg0 removal over the catalysts were investigated via activity tests and characterization techniques.

ET =

Eoxi =

■

EXPERIMENTAL SECTION Catalyst Preparation. The catalysts employed in this study were prepared with a cerium nitrate (Ce(NO3)3·6H2O) and/or manganese nitrate (Mn(NO3)2) aqueous solution as precursors and Ti-PILC as the support using an impregnation method. The details of the synthetic procedure are available in the Supporting Information (SI) (Part 1). The three catalysts used in this work are denoted as 12%CeOx/Ti-PILC, 12%MnOx/TiPILC and 6%Ce-6%MnOx/Ti-PILC, and the total metal mass percentage is selected as 12% (M/(M+Ti-PILC), M = Ce and/ or Mn) in each catalyst, according to the previous studies results of Ce-MnOx/Ti-PILC in SCR reactions.26,29 Catalyst Characterization. The textural characteristics of the catalysts were measured using the nitrogen (N2) adsorption−desorption method. The specific surface area and pore volume of the catalysts were calculated from the adsorption isotherm via the BET method, and the pore diameters were determined using the BJH method. The morphology analysis was performed with scanning electron microscopy (SEM). X-ray diffraction (XRD) and X-ray photoelectron spectroscopy (XPS) measurements were used to analyze the catalysts. The details of the material characterization methods are described in the SI (Part 2). Catalytic Activity Test. The elemental mercury (Hg0) removal activity test over the catalysts was performed with a bench scale catalytic activity test system (Part 3 in the SI). To investigate the Hg0 removal activity, a 0.5 mL (0.25 g) sample was loaded in the quartz reactor, and simulated flue gas was introduced into the reactor with a gas hourly space velocity (GHSV) of approximately 100 000 h−1. The inlet and outlet Hg0 concentrations were measured using an online mercury analyzer (H11-QM201H, Qingshan Co. LTD, China). The exhaust gas from the mercury analyzer as well as the reactor was introduced into the carbon trap and then expelled into the atmosphere. To avoid mercury condensation, heating belts were used to keep the silicone

Eads =

Hg oin − Hg oout Hg oin

× 100%

T Hg out − Hg oout

Hg oin T Hg oin − Hg out

Hg oin

E T = Eoxi + Eads

(1)

× 100% (2)

× 100% (3) (4)

Where the subscript “in” represents the reactor inlet and “out” represents the reactor outlet. As the elemental mercury and oxidized mercury in the outlet flue gas might be adsorbed on the surfaces of the adsorbents, the tests of E(2 h) at 2 h and E(equilibrium) at equilibrium were carried out in our experiments. The E(equilibrium) was calculated by the mercury concentrations when the total mercury inlet and outlet of the reactor was balanced and the fluctuation of Hg0 outlet was less than 5%.

■

RESULTS AND DISCUSSION Comparisons of the Hg0 Removal over Different Materials. To compare the different Hg0 removal efficiencies of the catalysts, the experiments were performed to evaluate the overall Hg0 removal efficiency of the catalysts. In these experiments, the simulated flue gas was introduced to the reactor under the same conditions (i.e., O2 at 5%, Hg0 at 45 μg/ m3, GHSV at 100 000 h−1 and the reaction temperatures at 100−350 °C). The concentrations of Hg0out and Hg0out (at 2 h) were detected. According to eq 1, the total Hg0 removal efficiency ET(2 h) was obtained at different temperatures for the catalysts. In fact, the calculated ET(2 h) included both the Hg0 adsorption efficiency (Eads(2 h)) and the Hg0 oxidation efficiency (Eoxi(2 h)). A comparison of ET(2 h) for the original clay, Ti-PICL and the Ce-MnOx based catalysts at 2 h are shown in Figure 1. The original clay exhibited the lowest Hg0 removal efficiencies (ET(2 h) 250 °C), which indicated that the adsorption of Hg0 was gradually replaced by the oxidation of Hg0 as the reaction proceeded at higher temperatures. 12% CeOx/Ti-PILC achieved an Eoxi(equilibrium) that was nearly the same as the ET(2 h) obtained in Figure 1, which indicated that the adsorption of Hg0 was gradually replaced by the oxidation of Hg0 as the reaction continued. However, the 12% CeOx/Ti-PILC catalyst exhibited higher Eoxi(equilibrium) compared to the other two catalysts at 300−350 °C, which may be due to the better catalytic oxidation property of the CeOx catalyst at high temperatures. By comparison of the results in Figures 1−3, in most cases, the 6%Ce-6%MnOx/TiPILC catalyst possessed better Hg0 adsorption and oxidation ability and a larger active temperature range than the other two catalysts. TiO2 supported MnOx-CeO2 showed high Hg0 oxidation efficiency at low temperature in the presence of HCl. However, in the absence of HCl, the Hg0 oxidation efficiency decreased to only approximately 75% at 200 °C.15,21 Al2O3 supported MnOx-CeO2 showed less activity than TiO2 supported MnOxCeO2.22 From Figure 3, it is known that 6%Ce-6%MnOx/TiPILC demonstrated higher Hg0 oxidation efficiency (>80%) than MnOx-CeO2/TiO2 in the absence of HCl. Characterization of the Materials. The BET specific areas and other pore structure parameters of the materials are listed in SI Table S1. The Ti-PILC catalyst possessed a much higher surface area (163.30 m2·g−1) than the original clay (70.47 m2·g−1) and the largest pore volume (0.1680 cm3·g−1) among all of the samples. The increasing BET area of Ti-PILC was due to successful pillaring.31 As shown in the SEM images (Figure S3 in SI Part 4), Ti-PILC exhibited a textural complexity compared to the original clay, which indicated that the pillaring process reorganized the layers of clay, and interlamellar layers were formed. 31 When loaded with

Eoxi(2 h) increased gradually as the temperature increased. In most cases, the Eoxi(2 h) and Eads(2 h) in the presence of O2 were higher than that in the absence of O2 for the catalysts at high temperatures (>250 °C) (except for 12%MnOx/Ti-PILC and 6%Ce-6%MnOx/Ti-PILC at 250 and 300 °C). The results indicated that O2 participated in the adsorption and oxidation of Hg0. The catalytic oxidation of Hg0 is favored at high temperatures. A comparison of ET(2 h) at 5% O2 and 10% O2 resulted in no apparent difference. This result suggested that 5% O2 was sufficient for improvement of ET(2 h) over the catalysts. As the reaction proceeds, the Hg0 adsorption efficiency will decrease and disappear when equilibrium is achieved, whereas the mercury oxidation efficiency will increase and reach balance when equilibrium is achieved. It should be noticed that the adsorption of oxidized mercury (Hgoxi) would occur during this process. The equilibrium can be obtained when the total mercury inlet and outlet of the reactor was balanced and the fluctuation of the outlet Hg0 was less than 5%. To verify the correctness of the equilibrium, a 72 h experiment over 6%Ce6%MnOx/Ti-PILC at two selected temperatures (150 and 250 °C) was carried out. As shown in SI Figure S2, when the ET value got stable with the equality of the total mercury inlet and outlet of the reactor, the Eads disappeared and the Eoxi kept constant with the reaction time. This verified that when the equilibrium was obtained, the difference between the Hg0 concentrations at inlet and outlet was due to the catalytic oxidation. The accumulative adsorption amounts (Q) of total mercury for 6%Ce-6%MnOx/Ti-PILC were 442.7 μg/g (150 °C) and 223.8 μg/g (250 °C) (the computational method is shown in the SI Part 6). So the stable ET can be identified as the Hg0 oxidation efficiency (SI Figure S2 (B)). The following experiments were aimed at identifying the Hg0 oxidation ability of the catalysts according to the equilibrium determined by the Hg0 balance discussed above. The experimental conditions were the same as those in Figure 1. After the reactions were continuously conducted for 40−100 h, the mercury adsorption process was determined to reach equilibrium (the adsorption equilibrium time was listed in SI Table S3). Then, the HgTout, Hg0out, and Hg0in concentrations were detected to calculate the potential Hg0 oxidation efficiency (Eoxi(equilibrium)) according to eq 3. Figure 3 shows the

Figure 3. Hg0 oxidation efficiency of the catalysts at adsorption equilibrium (5% O2). 7894

dx.doi.org/10.1021/es5007719 | Environ. Sci. Technol. 2014, 48, 7891−7898

Environmental Science & Technology

Article

Figure 4. XPS spectra of 6%Ce-6%MnOx/Ti-PILC over the spectral regions of (A) O 1s, (B) Ce 3d, (C) Mn 2p, and (D) Hg 4f.

surface of the clay.33 The XRD pattern of 12%CeOx/Ti-PILC contained four distinct peaks at 28.7°, 33.1°, 47.5°, and 56.6°, and all four peaks corresponded to cerium oxide (CeO2). In addition, the SEM images (SI Figure S3) of 12%CeOx/Ti-PILC also displayed crystalline particles that were identified as metal oxides on the surface of the Ti-PILC support. SI Figure S4 shows the formation of MnO2 in 12%MnOx/Ti-PILC, and the pyrolusite MnO2 crystallization was confirmed by the presence of four peaks at 28.6°, 37.3°, 42.7°, and 56.8°. However, no characteristic peaks of other manganese oxides were found in the XRD pattern of 12% MnOx/Ti-PILC. It is important to note that characteristic peaks corresponding to cerium oxide and pyrolusite MnO2 are not distinct in the XRD pattern of 6% Ce-6%MnOx/Ti-PILC, which indicated that the Ce−Mn mixed oxide was well dispersed on the Ti-PILC support compared to the other catalysts.35 As demonstrated in the activity test, 6%Ce-6%MnOx/TiPILC exhibited the best Hg0 oxidation capacity, and the Hg0

manganese and/or cerium oxide, the BET surface areas of the 12%CeO x /Ti-PILC, 12%MnO x /Ti-PILC, and 6%Ce-6% MnOx/Ti-PILC catalysts moderately decreased to 105.08 m2· g−1, 125.15 m2·g−1, and 93.46 m2·g−1, respectively. This result may be due to partial blockage of the pores by a metallic oxide.25,26 The XRD patterns of the clay, Ti-PILC, and Ce-MnOx/TiPILC catalysts are shown in Figure S4 in SI Part 4. The XRD pattern of clay exhibited a basal 001 reflection at 5.2° and a two-dimensional diffraction (hk) at approximately 19.7° and 35.2°, and these peaks are characteristic of the type of mineral clays.32 The peaks at 21.9° and 29.4° were assigned to cristobalite and calcite impurity.33 However, the basal 001 line barely appeared in the pattern of Ti-PILC, which was most likely due to the mineral structure being disordered by the calcination process.34 As observed, the characteristic peaks of the clay mineral were weakened after the pillaring process, which was caused by a pillaring agent that was deposited on the 7895

dx.doi.org/10.1021/es5007719 | Environ. Sci. Technol. 2014, 48, 7891−7898

Environmental Science & Technology

Article

used at 150 °C, the Mn2+ peak was observed at 641.4 eV with a ratio of 20.9%. In addition, the ratio of Mn3+ decreased from 61.4% to 25.0%, and the Mn4+ ratio increased from 38.6% to 54.1%. The decrease in Mn3+ was equal to the increase in Mn4+ and Mn2+. The result indicated that both the reduction and oxidation of Mn3+ occurred in the presence of O2 at 150 °C. In addition, the reduction of Mn3+ to Mn2+ contributed to Hg0 oxidation. After Hg0 oxidation at 250 °C, the ratio of Mn4+ decreased to 35.5%, and the ratio of Mn3+ increased accordingly without the appearance of a peak corresponding to Mn2+ on the surface of the 6%Ce-6%MnOx/Ti-PILC catalyst, which suggested that the reduction of Mn4+ to Mn3+ on the surface of the catalyst contributed to the Hg0 oxidation at this temperature. The XPS spectra for Hg 4f are shown in Figure 4(D). The peak at 102.9 eV for the fresh 6%Ce-6%MnOx/Ti-PILC catalyst corresponded to Si 2p. For the spent catalyst at 150 °C, Hg 4f peaks appeared at approximately 104 eV, which were assigned to mercury oxide (HgO). However, no adsorbed Hg0 was observed on the surface of the catalysts, which may be due to the Hg0 escaping when the samples were treated prior to XPS analysis. Hg0 can be easily sublimed at 45 °C. The presence of HgO on the surface of the catalyst indicated that the oxidation reaction occurred on the surface of the catalyst. The spent catalyst at 250 °C exhibited a much smaller Hg 4f peak than that at 150 °C, which indicated partial volatilization of HgO at high temperatures. Mechanism for the Hg0 Oxidation over 6%Ce-6% MnOx/Ti-PILC. From the above studies, during the first stage of the reaction, the main Hg0 capture mechanism was adsorption. As the reaction proceeded, the Hg0 oxidation ability would replace part of or the entire adsorption ability. When the adsorption reached equilibrium, the Hg0 oxidation ability exhibited the highest value. The Hg0 oxidation preferred to occur at a high temperature. Based on the catalyst characterization, both the hydroxyl oxygens and the lattice oxygen participated in the Hg0 oxidation. The lattice oxygen ([O]) came from the valence state changes of Ce and Mn, which can be described as follows (eqs 5−7):

removal mechanisms might be different at low temperature and high temperature. To gain more insight into the Hg0 oxidation mechanisms, the fresh and spent 6%Ce-6%MnOx/Ti-PILC catalyst with a Hg0 saturation adsorption at 150 and 250 °C were selected for XPS analysis. The XPS spectra over the spectral regions of O 1s, Ce 3d, Mn 2p, and Hg 4f are shown in Figure 4. And surface atomic concentration of fresh and spent 6%Ce-6%MnOx/Ti-PILC is listed in SI Table S2. In Figure 4 (A), for the fresh 6%Ce-6%MnOx/Ti-PILC catalyst, the O 1s spectrum was divided into two main peaks at 529.7 eV (assigned to lattice oxygen) and 532.1 eV (assigned to the hydroxyl oxygen).36,37 After the Hg0 saturation adsorption at 150 °C (in the presence of 5% O2), a new peak at approximately 534.6 eV appeared in the O 1s spectrum on the surface of the catalyst, and this peak was assigned to H2O. In addition, the ratio for the peak at 532.1 eV (attributed to the hydroxyl oxygen) decreased from 81.9% to 61.2% compared to that of the fresh catalyst, and the ratio of lattice oxides decreased from 18.1% to 12.4%. These observations suggested that both hydroxyl oxygen (OH) and lattice oxygen participated in Hg0 oxidation at 150 °C, and H2O was simultaneously produced on the surface of the catalysts. After the Hg0 saturation adsorption at 250 °C (in the presence of 5% O2), the O 1s spectrum corresponding to H2O at 533.2 eV also appeared. The ratio of hydroxyl oxygen (OH) decreased from 81.9% to 54.9% compared to that of the fresh catalyst, whereas the ratio of lattice oxygen increased from 18.1% to 24.4%. These results indicated that only hydroxyl oxygen (OH) participated in Hg0 oxidation at a high temperature (i.e., 250 °C). The ratio of the O 1s spectrum for H2O on the surface of the catalyst at 250 °C was 20.7%, which was lower than that at 150 °C (26.4%). As shown in Figure 3, at the adsorption equilibrium, the 6%Ce-6%MnOx/Ti-PILC catalyst demonstrated a higher Eoxi(equilibrium) at 250 °C than at 150 °C, which indicated that more H2O was produced on the surface of the catalyst at 250 °C than at 150 °C. The contradictory results for H2O might be due to the partial volatilization of H2O from the surface of the catalysts as the temperature increased. The increase in the ratio for lattice oxygen at 250 °C can be explained by two factors. First, O2 in the flue gas compensated for the lattice oxygen consumed in the reaction. Second, H2O volatilization resulted in a decrease in hydroxyl oxygen and H2O oxygen, which increased the percentage of lattice oxygen. The XPS results of Ce 3d for 6%Ce-6%MnOx/Ti-PILC are shown in Figure 4(B). The bands labeled u1 and v1 represent Ce3+ cations, and the peaks labeled u, u2, u3, v, v2 and v3 represent Ce4+ cations.17,38 Both Ce3+ and Ce4+ existed in the fresh 6%Ce-6%MnOx/Ti-PILC catalyst. Based on the peak areas, Ce4+ oxide was the primary form, which is considered to be beneficial for Hg0 oxidation.39 The presence of Ce3+ species could lead to a charge imbalance, vacancies, and unsaturated chemical bonds.40 After Hg0 saturation adsorption at 150 °C, the ratio of Ce4+/Ce3+ decreased from 1.7 to 0.9 compared to the fresh catalyst, which indicated a reduction of Ce4+ during Hg0 oxidation. In addition, this observation revealed the possibility of the reaction between cation vacancies and Hg0. The ratio of Ce4+/Ce3+ increased to 2.2 when the temperature reached 250 °C, which may be due to the oxidation of Ce3+ by O2 at high temperatures. The XPS spectra for Mn 2p are shown in Figure 4(C). For the fresh 6%Ce-6%MnOx/Ti-PILC catalyst, the peaks observed at 644.5 and 641.7 eV correspond to Mn4+ and Mn3+, respectively, and no Mn2+ peak was observed. For the catalyst

2CeO2 → Ce2O3 + [O]

(5)

2MnO2 → Mn2O3 + [O]

(6)

Mn2O3 → 2MnO + [O]

(7)

[O] + Hg → HgO

(8)

Because of the lattice oxygen ([O]) provided by the metal transfer, the Hg0 oxidation by lattice oxygen (eq 8) could be performed in the absence of O2 in the flue gas, as shown in Figure 2 (C). Many researchers have proposed that active oxygen can be generated from adsorbed oxygen by cerium species.18,35,37 In the presence of O2 in the flue gas, O2 would supply the metal oxides with oxygen to ensure that Hg0 oxidation can be continued. The transfer of O2 to lattice oxygen in metal oxides is as follows (eqs 9-11): Ce2O3 + 1/2O2 → 2CeO2

7896

(9)

Mn2O3 + 1/2O2 → 2MnO2

(10)

2MnO + 1/2O2 → Mn2O3

(11)

dx.doi.org/10.1021/es5007719 | Environ. Sci. Technol. 2014, 48, 7891−7898

Environmental Science & Technology

Article

For the hydroxyl oxygen (OH), the Hg0 oxidation is described in eq 12. The production of hydroxyl oxygen on the surface of the catalyst is described in eq 13. OH + Hg → HgO + H

(12)

H + 1/2O2 → OH

(13)

12JCZDJC29300) and the Marine Science and Technology Project from Tianjin Marine Burean (No. KJXH2013-05).

■

(1) Brauer, M.; Amann, M.; Burnett, R. T.; Cohen, A.; Dentener, F.; Ezzati, M.; Henderson, S. B.; Krzyzanowski, M.; Martin, R. V.; Van Dingenen, R.; van Donkelaar, A.; Thurston, G. D. Exposure assessment for estimation of the global burden of disease attributable to outdoor air pollution. Environ. Sci. Technol. 2011, 46 (2), 652−660. (2) Zhang, A.; Zheng, W.; Song, J.; Hu, S.; Liu, Z.; Xiang, J. Cobalt manganese oxides modified titania catalysts for oxidation of elemental mercury at low flue gas temperature. Chem. Eng. J. 2014, 236, 29−38. (3) Glodek, A.; Pacyna, J. M. Mercury emission from coal-fired power plants in Poland. Atmos. Environ. 2009, 43 (35), 5668−5673. (4) Galbreath, K. C.; Zygarlicke, C. J. Mercury speciation in coal combustion and gasification flue gases. Environ. Sci. Technol. 1996, 30 (8), 2421−2426. (5) Presto, A. A.; Granite, E. J. Survey of catalysts for oxidation of mercury in flue gas. Environ. Sci. Technol. 2006, 40 (18), 5601−5609. (6) Mei, Z.; Shen, Z.; Mei, Z.; Zhang, Y.; Xiang, F.; Chen, J.; Wang, W. The effect of N-doping and halide-doping on the activity of CuCoO4 for the oxidation of elemental mercury. Appl. Catal., B 2008, 78 (1−2), 112−119. (7) Gao, Y.; Zhang, Z.; Wu, J.; Duan, L.; Umar, A.; Sun, L.; Guo, Z.; Wang, Q. A critical review on the heterogeneous catalytic oxidation of elemental mercury in flue gases. Environ. Sci. Technol. 2013, 47 (19), 10813−10823. (8) Cai, J.; Shen, B.; Li, Z.; Chen, J.; He, C. Removal of elemental mercury by clays impregnated with KI and KBr. Chem. Eng. J. 2014, 241, 19−27. (9) Kamata, H.; Ueno, S.-i.; Naito, T.; Yukimura, A. Mercury Oxidation over the V2O5(WO3)/TiO2 commercial SCR catalyst. Ind. Eng. Chem. Res. 2008, 47 (21), 8136−8141. (10) Guo, P.; Guo, X.; Zheng, C.-g. Computational insights into interactions between Hg Species and α-Fe2O3 (001). Fuel 2011, 90 (5), 1840−1846. (11) Qiao, S.; Chen, J.; Li, J.; Qu, Z.; Liu, P.; Yan, N.; Jia, J. Adsorption and catalytic oxidation of gaseous elemental mercury in flue gas over MnOx/alumina. Ind. Eng. Chem. Res. 2009, 48 (7), 3317− 3322. (12) Mei, Z.; Shen, Z.; Wang, W.; Zhang, Y. Novel sorbents of nonmetal-doped spinel Co3O4 for the removal of gas-phase elemental mercury. Environ. Sci. Technol. 2007, 42 (2), 590−595. (13) Pan, H. Y.; Minet, R. G.; Benson, S. W.; Tsotsis, T. T. Process for converting hydrogen chloride to chlorine. Ind. Eng. Chem. Res. 1994, 33 (12), 2996−3003. (14) Ji, L.; Sreekanth, P. M.; Smirniotis, P. G.; Thiel, S. W.; Pinto, N. G. Manganese oxide/titania materials for removal of NOx and elemental mercury from flue gas. Energy Fuels 2008, 22 (4), 2299− 2306. (15) Li, H.; Wu, C.-Y.; Li, Y.; Li, L.; Zhao, Y.; Zhang, J. Role of flue gas components in mercury oxidation over TiO2 supported MnOxCeO2 mixed-oxide at low temperature. J. Hazard. Mater. 2012, 243 (0), 117−123. (16) Li, J.; Yan, N.; Qu, Z.; Qiao, S.; Yang, S.; Guo, Y.; Liu, P.; Jia, J. Catalytic oxidation of elemental mercury over the modified catalyst Mn/α-Al2O3 at lower temperatures. Environ. Sci. Technol. 2010, 44 (1), 426−431. (17) Larachi, F. C.; Pierre, J.; Adnot, A.; Bernis, A. Ce 3d XPS study of composite CexMn1−xO2−y wet oxidation catalysts. Appl. Surf. Sci. 2002, 195 (1−4), 236−250. (18) Shen, B.; Wang, F.; Liu, T. Homogeneous MnOx−CeO2 pellets prepared by a one-step hydrolysis process for low-temperature NH3SCR. Powder Technol. 2014, 253, 152−157. (19) Shen, B.; Wang, Y.; Wang, F.; Liu, T. The effect of Ce−Zr on NH3-SCR activity over MnOx(0.6)/Ce0.5Zr0.5O2 at low temperature. Chem. Eng. J. 2014, 236, 171−180. (20) Zhang, X.; Shen, B.; Wang, K.; Chen, J. A contrastive study of the introduction of cobalt as a modifier for active components and

Based on the results from our studies, the hydroxyl oxygen and lattice oxygen from Ce4+→ Ce3+ and Mn3+ → Mn2+ contributed to the Hg0 oxidation at low temperatures (150 °C). Therefore, the mechanism for Hg0 oxidation at low temperatures (150 °C) is as follows: 2CeO2 → Ce2O3 + [O]

(5)

Mn2O3 → 2MnO + [O]

(7)

[O] + Hg → HgO

(8)

OH + Hg → HgO + H

(12)

It is important to note that the mechanism for Hg0 oxidation at 250 and 150 °C was different. The hydroxyl oxygen and lattice oxygen from Mn4+ → Mn3+ contributed to the Hg0 oxidation at high temperatures (250 °C) according to the above studies. Therefore, the mechanism for Hg0 oxidation at high temperatures (250 °C) is as follows: 2MnO2 → Mn2O3 + [O]

(6)

[O] + Hg → HgO

(8)

OH + Hg → HgO + H

(12)

It should be noticed that other components in coal-fired flue gas may have effects on the mechanism of Hg0 removal in the case of flue gas before flue gas desulfurization. Several researchers demonstrated that HCl in the flue gas will enhance the Hg0 oxidation of the catalyst dramatically via a Deacon process.5,13 SO2 will poison the Mn by sulfurization,41,42 which would deactivate the activity of the catalyst. The details of the Hg0 removal reactions under various flue gas components over Ce-MnOx/Ti-PILCs will be considered in our future study. From the above studies, it is known that the 6%Ce-6% MnOx/Ti-PILC catalyst exhibited good adsorption and oxidation ability in Hg0 capture. The 6%Ce-6%MnOx/TiPILC catalyst was demonstrated to be a good Hg0 adsorbent and catalytic oxidant in the absence of HCl in the flue gas.

■

ASSOCIATED CONTENT

S Supporting Information *

Additional information as noted in the text. This material is available free of charge via the Internet at http://pubs.acs.org.

■

REFERENCES

AUTHOR INFORMATION

Corresponding Author

*Phone: 86-22-23503219; e-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This project was financially supported by the National Natural Science Foundation of China (No. 51176077), the Research Fund for International Young Scientists (No. 51350110229), the Key Project of Natural Science Foundation of Tianjin (No. 7897

dx.doi.org/10.1021/es5007719 | Environ. Sci. Technol. 2014, 48, 7891−7898

Environmental Science & Technology

Article

supports of catalysts for NH3-SCR. J. Ind. Eng. Chem. 2013, 19, 1272− 1279. (21) Li, H.; Wu, C.-Y.; Li, Y.; Zhang, J. Superior activity of MnOxCeO2/TiO2 catalyst for catalytic oxidation of elemental mercury at low flue gas temperatures. Appl. Catal., B 2012, 111−112, 381−388. (22) Wang, P.; Su, S.; Xiang, J.; You, H.; Cao, F.; Sun, L.; Hu, S.; Zhang, Y. Catalytic oxidation of Hg0 by MnOx−CeO2/γ-Al2O3 catalyst at low temperatures. Chemosphere 2014, 101, 49−54. (23) Yang, R. T.; Chen, J. P.; Kikkinides, E. S.; Cheng, L. S.; Cichanowicz, J. E. Pillared clays as superior catalysts for selective catalytic reduction of nitric oxide with ammonia. Ind. Eng. Chem. Res. 1992, 31 (6), 1440−1445. (24) Kloprogge, J. T. Synthesis of smectites and porous pillared clay catalysts: A review. J. Porous Mater. 1998, 5 (1), 5−41. (25) Shen, B.; Yao, Y.; Chen, J.; Zhang, X. Alkali metal deactivation of Mn−CeOx/Zr-delaminated-clay for the low-temperature selective catalytic reduction of NOx with NH3. Microporous Mesoporous Mater. 2013, 180, 262−269. (26) Shen, B.; Ma, h.; Yao, Y. Mn-CeOx/Ti-PILCs for selective catalytic reduction of NO with NH3 at low temperature. J. Environ. Sci. 2012, 24 (3), 499−506. (27) Arfaoui, J.; Boudali, L. K.; Ghorbel, A.; Delahay, G. Effect of vanadium on the behaviour of unsulfated and sulfated Ti-pillared clay catalysts in the SCR of NO by NH3. Catal. Today 2009, 142 (3−4), 234−238. (28) Anirudhan, T. S.; Bringle, C. D.; Radhakrishnan, P. G. Heavy metal interactions with phosphatic clay: Kinetic and equilibrium studies. Chem. Eng. J. 2012, 200−202, 149−157. (29) Shen, B.; Yao, Y.; Ma, H. Ceria modified MnOx/TiO2-pillared clays catalysts for selective catalytic reduction of NO with NH3 at low temperature. Chin. J. Catal. 2011, 32 (12), 1803−1810. (30) Kamata, H.; Ueno, S.-i.; Naito, T.; Yamaguchi, A.; Ito, S. Mercury oxidation by hydrochloric acid over a VOx/TiO2 catalyst. Catal. Commun. 2008, 9 (14), 2441−2444. (31) Bergaya, F. The meaning of surface area and porosity measurements of clays and pillared clays. J. Porous Mater. 1995, 2 (1), 91−96. (32) Chae, H. J.; Nam, I.-S.; Ham, S. W.; Hong, S. B. Physicochemical characteristics of pillared interlayered clays. Catal. Today 2001, 68 (1−3), 31−40. (33) Li, X.; Lu, G.; Qu, Z.; Zhang, D.; Liu, S. The role of titania pillar in copper-ion exchanged titania pillared clays for the selective catalytic reduction of NO by propylene. Appl. Catal., A 2011, 398 (1−2), 82− 87. (34) Boudali, L. K.; Ghorbel, A.; Grange, P.; Figueras, F. Selective catalytic reduction of NO with ammonia over V2O5 supported sulfated titanium-pillared clay catalysts: Influence of V2O5 content. Appl. Catal., B 2005, 59 (1−2), 105−111. (35) Wang, X.; Zheng, Y.; Lin, J. Highly dispersed Mn−Ce mixed oxides supported on carbon nanotubes for low-temperature NO reduction with NH3. Catal. Commun. 2013, 37, 96−99. (36) Li, Z.; Deng, S.; Yu, G.; Huang, J.; Lim, V. C. As(V) and As(III) removal from water by a Ce−Ti oxide adsorbent: Behavior and mechanism. Chem. Eng. J. 2010, 161 (1−2), 106−113. (37) Deng, S.; Liu, H.; Zhou, W.; Huang, J.; Yu, G. Mn−Ce oxide as a high-capacity adsorbent for fluoride removal from water. J. Hazard. Mater. 2011, 186 (2−3), 1360−1366. (38) He, J.; Reddy, G. K.; Thiel, S. W.; Smirniotis, P. G.; Pinto, N. G. Ceria-modified manganese oxide/titania materials for removal of elemental and oxidized mercury from flue gas. J. Phys. Chem. C 2011, 115 (49), 24300−24309. (39) Wan, Q.; Duan, L.; He, K.; Li, J. Removal of gaseous elemental mercury over a CeO2−WO3/TiO2 nanocomposite in simulated coalfired flue gas. Chem. Eng. J. 2011, 170 (2−3), 512−517. (40) Liu, X.; Zhou, K.; Wang, L.; Wang, B.; Li, Y. Oxygen vacancy clusters promoting reducibility and activity of ceria nanorods. J. Am. Chem. Soc. 2009, 131 (9), 3140−3141.

(41) Shen, B.; Liu, T. Deactivation of MnOx-CeOx/ACF catalysts for low-temperature NH3-SCR in the presence of SO2. Acta Phys.-Chim. Sin. 2010, 26 (11), 3009−3016. (42) Shen, B.; Liu, T.; Zhao, N.; Yang, X.; Deng, L. Iron-doped MnCe/TiO2 catalyst for low temperature selective catalytic reduction of NO with NH3. J. Environ. Sci. 2010, 22 (9), 1447−1454.

7898

dx.doi.org/10.1021/es5007719 | Environ. Sci. Technol. 2014, 48, 7891−7898