CeO2–TiO2 Catalysts for Catalytic Oxidation of Elemental Mercury in

ARTICLE pubs.acs.org/est

CeO2
TiO2 Catalysts for Catalytic Oxidation of Elemental Mercury in Low-Rank Coal Combustion Flue Gas Hailong Li,†,‡ Chang-Yu Wu,*,‡ Ying Li,§ and Junying Zhang† †

State Key Laboratory of Coal Combustion, Huazhong University of Science and Technology, Wuhan, Hubei 430074, China Department of Environmental Engineering Sciences, University of Florida, Gainesville, Florida 32611, United States § Department of Mechanical Engineering, University of Wisconsin-Milwaukee, Milwaukee, Wisconsin 53211, United States ‡

bS Supporting Information ABSTRACT: CeO2
TiO2 (CeTi) catalysts synthesized by an ultrasoundassisted impregnation method were employed to oxidize elemental mercury (Hg0) in simulated low-rank (sub-bituminous and lignite) coal combustion flue gas. The CeTi catalysts with a CeO2/TiO2 weight ratio of 1
2 exhibited high Hg0 oxidation activity from 150 to 250 °C. The high concentrations of surface cerium and oxygen were responsible for their superior performance. Hg0 oxidation over CeTi catalysts was proposed to follow the Langmuir
Hinshelwood mechanism whereby reactive species from adsorbed flue gas components react with adjacently adsorbed Hg0. In the presence of O2, a promotional effect of HCl, NO, and SO2 on Hg0 oxidation was observed. Without O2, HCl and NO still promoted Hg0 oxidation due to the surface oxygen, while SO2 inhibited Hg0 adsorption and subsequent oxidation. Water vapor also inhibited Hg0 oxidation. HCl was the most effective flue gas component responsible for Hg0 oxidation. However, the combination of SO2 and NO without HCl also resulted in high Hg0 oxidation efficiency. This superior oxidation capability is advantageous to Hg0 oxidation in low-rank coal combustion flue gas with low HCl concentration.

’ INTRODUCTION Coal combustion is currently the largest single-known source of anthropogenic mercury emissions in the United States. Because of the extreme toxicity, persistence, and bioaccumulation of methyl mercury transformed from emitted mercury,1 by April 2010 more than 20 U.S. states had proposed or adopted regulations to limit mercury emissions from coal-fired power plants.2 The U.S. Environmental Protection Agency (EPA) is also including mercury in the federal emissions standards for power plants, which will be finalized by November 2011.3 Implementation of these mercury regulations requires effective control technologies to be developed. Various technologies for controlling mercury emissions from coal-fired power plants such as sorbent injection and catalytic oxidation combined with wet flue gas desulfurization (WFGD) have been investigated. However, there is no single best technology that can be applied broadly.1 The efficacy of a mercury control method depends largely on the form and species of mercury.4 Mercury in coal combustion derived flue gas is present in three forms: elemental mercury (Hg0), oxidized mercury (Hg2+), and particulate-bound mercury (Hgp).4 Hgp can be captured by particulate matter (PM) control devices such as electrostatic precipitators (ESP) and fabric filters (FF). Water-soluble Hg2+ is readily captured in WFGD systems.4 Hg2+ can also be adsorbed on fly ash and subsequently collected in PM control devices. In contrast, Hg0 vapor is most likely to escape from existing air pollution control r 2011 American Chemical Society

devices because it is highly volatile and nearly insoluble in water.5 As such, Hg0 is the dominant mercury species emitted to the atmosphere. Consequently, catalysts capable of significant conversion (>80%) of Hg0 to Hg2+ would have tremendous value6 because Hg2+ can be removed simultaneously with PM and acid gases in ESP/FF and WFGD, respectively. Several metal oxides such as V2O5, Fe2O3, CuO, Cr2O3, Mn2O3, NiO, and MoO3 have been extensively investigated as potential Hg0 oxidation catalysts.6
8 V2O5, as the active component of selective catalytic reduction (SCR) catalysts, is among the most effective metal oxides for Hg0 oxidation.9 The presence of HCl greatly improves Hg0 oxidation over V2O5 based catalysts,10,11 which is believed to proceed through the following overall equation9,12 Hg 0 þ 2HCl þ

1 O2 T HgCl2 þ H2 O 2

ð1Þ

Obviously, oxygen is necessary for Hg0 oxidation by HCl. Besides gas-phase O2, stored oxygen on catalysts including lattice oxygen, chemisorbed, or weakly bonded oxygen are active for oxidation processes. Therefore, catalysts with larger oxygen storage capacity Received: March 7, 2011 Accepted: July 19, 2011 Revised: July 17, 2011 Published: July 19, 2011 7394

dx.doi.org/10.1021/es2007808 | Environ. Sci. Technol. 2011, 45, 7394–7400

Environmental Science & Technology could facilitate Hg0 oxidation. Furthermore, these catalysts probably could maintain Hg0 oxidation process under oxygendeficient environments such as low excess-air combustion flue gas. Although V2O5 based catalyst is active for Hg0 oxidation, its effectiveness depends on HCl concentration of the flue gas.13 Only less than 30% Hg0 oxidation was achieved over V2O5 based SCR catalyst under sub-bituminous coal combustion flue gas with low concentrations of HCl.14 To achieve higher Hg0 oxidation rates in coal combustion flue gas with lower HCl concentration, a catalyst which can effectively oxidize Hg0 in such scenarios is of great value. Additionally, the loss of V2O5 during synthesis and application of V2O5 based catalysts is harmful to the environment and human health.15 When we consider these limitations of V2O5 based catalysts, CeO2, which is abundant, nontoxic, and inexpensive, emerges as an attractive candidate. CeO2 has been extensively studied as a support,16 promoter,17 and active species18,19 for selective catalytic reduction of NOx due to its large oxygen storage capacity and unique redox couple Ce3+/Ce4+ with the ability to shift between CeO2 and Ce2O3 under oxidizing and reducing conditions, respectively.20 Within the redox shift between Ce3+ and Ce4+, labile oxygen vacancies and bulk oxygen species with relatively high mobility can be easily generated,20 and these are active for oxidation processes. Furthermore, CeO2 promotes the oxidation of NO to NO221 which was reported to be more effective than NO for Hg0 oxidation.22 Similarly, HCl would probably be transformed to active chlorine species over CeO2 based catalysts even without gas-phase O2. In addition, CeO2 based catalysts were reported to have resistance to water vapor,19 the presence of which inhibits Hg0 oxidation.22 Accordingly, we hypothesize that Hg0 can be effectively oxidized over CeO2 based catalysts even in extremely adverse conditions such as deficiency of O2 and low-rank coal combustion flue gas with low concentration of HCl or even without HCl. However, pure CeO2 is thermo-unstable.20,23 Introduction of Ti4+ into the CeO2 lattice could enhance the thermal stability by forming CeO2
TiO2 (CeTi) mixed oxide,23 which has been developed recently as a promising SCR catalyst for NOx removal.18 To date, no research on using CeTi catalysts for Hg0 oxidation has been reported. In this work, CeTi catalysts prepared by an ultrasound-assisted impregnation method were employed to oxidize Hg0 in simulated low-rank coals (sub-bituminous and lignite) combustion flue gas. The mechanisms involved in Hg0 oxidation were identified. Effects of catalyst composition and individual flue gas components on Hg0 oxidation were investigated as well. The ultimate goal is to develop a nontoxic catalyst with high oxygen storage capacity and no HCl dependence for Hg0 oxidation in flue gas.

’ EXPERIMENTAL SECTION Preparation of Catalysts. The CeTi catalysts were synthesized by an impregnation method enhanced by ultrasound. The catalysts are abbreviated by way of CexTi, where Ce represents CeO2, Ti represents TiO2, and x, in the range of 0.5 to 2, represents the CeO2/TiO2 mass ratio. Please refer to Synthesis of CeTi Catalysts in the Supporting Information for further details. Characterization of Catalysts. Please refer to Material Characterization Methods in the Supporting Information. Catalytic Activity Measurement. To evaluate the performance of CeTi catalysts on Hg0 oxidation, a bench-scale experimental system was built as shown in Figure S1. In each test, 1 g of CeTi catalyst was loaded in a Pyrex reactor, which was placed in a

ARTICLE

temperature controlled tubular furnace to control the reaction temperature. All individual flue gas components were from cylinder gases and were precisely controlled by mass flow controllers, with a total flow rate of 1 L 3 min
1. Water vapor was generated using a heated water bubbler. A Dynacal Hg0 permeation device (VICI Metronics) was used to provide a constant feed of Hg0 concentration (∼50 μg 3 m
3). The relatively high Hg0 concentration was employed to reduce the experimental errors caused by the sensitivity of the mercury analyzer and to allow experiments to be completed within a reasonable time scale. A RA-915+ mercury analyzer (OhioLumex) coupled with a mercury speciation conversion system was employed to measure Hg0 and Hg2+ concentrations. The specifications of the mercury speciation conversion system are described in the Supporting Information. Interferences on Hg0 measurement by the empty reactor and flue gas components were verified to be negligible. Even so, before proceeding to the mercury analyzer, acid gas and H2O were removed from the sample flow. Four sets of experiments were conducted, and their conditions are summarized in Table S1. Set I experiments aimed at determining the optimal CeO2/TiO2 ratio and operating temperature. The catalytic Hg0 oxidation activities over CeTi catalysts with different CeO2/TiO2 ratios were evaluated under simulated flue gas (4% O2, 8% H2O, 12% CO2, 10 ppm HCl, 300 ppm NO, 400 ppm SO2, and about 50 μg 3 m
3 Hg0) representing those from burning low-rank coals with space velocity about 60,000 h
1 and at reaction temperatures from 120 to 400 °C. At each selected temperature, Hg0 concentrations downstream from the catalysts were recorded after the process had reached equilibrium, which was defined as having fluctuation of Hg0 concentration less than 5% for more than 30 min. In Set II, the effect of individual flue gas components on Hg0 oxidation and the reaction pathways were studied. Experiments were conducted on the optimal Ce1.5Ti catalyst in the presence of individual flue gases (balanced in N2 or O2 plus N2) at 200 °C. Set III experiments were designed to identify the possible mechanism involved in Hg0 oxidation using HCl pretreated Ce1.5Ti (30 ppm HCl balanced in N2 passed through Ce1.5Ti at 200 or 300 °C for 1 h; then, the catalyst was flushed by pure N2 gas flow at the same temperature for 30 min). In set IV, the possibility of effective Hg0 oxidation without HCl was explored at 200 °C. The interaction between SO2 and NO on Hg0 oxidation was investigated as well. In Sets I, II, and IV, three replicates were conducted, and the mean values and standard deviations were reported. At the beginning of each test, the gas stream bypassed the reactor and the inlet gas was sampled to ensure a stable Hg0 feed concentration (Hgin0). Then, gas flow passed through the catalysts was taken from the reactor exit to get the outlet Hg0 concentration (Hgout0). At the end of each test, the mercury analyzer was switched to the reactor inlet to verify the Hgin0. The capacities of CeTi catalysts to adsorb Hg0 were observed to be negligible (less than 30 min was needed for Hg0 saturation) even at room temperature. Moreover, to avoid possible bias due to Hg0 adsorption, in Sets I, II, and IV, CeTi catalysts were first saturated in 50 μg 3 m
3 Hg0 plus N2 gas flow at room temperature. Therefore, the decrease of Hg0 concentration across the catalysts was attributed to Hg0 oxidation. Accordingly, the definition of Hg0 oxidation efficiency (Eoxi) over CeTi catalysts is as follows Eoxi ð%Þ ¼ 7395

0 ΔHg 0 Hgin0
Hgout ¼  100% Hgin0 Hgin0

ð2Þ

dx.doi.org/10.1021/es2007808 |Environ. Sci. Technol. 2011, 45, 7394–7400

Environmental Science & Technology

ARTICLE

’ RESULTS AND DISCUSSION Characterization of Catalysts. The BET surface areas of pure TiO2, pure CeO2, and the CeTi catalysts are listed in Table S2. Surface areas of all CeTi catalysts were about 60 m2 3 g
1. Pure TiO2 exhibited the lowest surface area of 56.99 m2 3 g
1, while pure CeO2 exhibited the highest surface area of 81.48 m2 3 g
1. No clear trend between surface area and CeO2/TiO2 ratio was observed. The XRD patterns of different catalysts are shown in Figure S2. For pure CeO2, only strong diffraction peaks of cubic CeO2 were observed in the XRD pattern. For pure TiO2, both anatase TiO2 and rutile TiO2 were detected, with anatase being the dominating phase. In all CeTi catalysts, antase TiO2 was also detected to be the predominant form, and the intensity of peaks attributed to anatase TiO2 decreased with the increase of CeO2 content, indicating that CeO2 and TiO2 interacted with each other in these catalysts.18 Cubic CeO2 phase was observed in all CeTi catalysts and became more apparent with the increase of CeO2 loading. This is in line with another study showing that cubic CeO2 was observed when the mass ratio of CeO2/TiO2 exceeded 0.2 in CeTi catalysts prepared by an impregnation method.19 Moreover, the CeO2 peaks became much wider and weaker with the increase of TiO2 content, indicating that crystalline size of CeO2 particles decreased with the increase of TiO2 content. This was probably due to the incorporation of Ti4+ into CeO2 lattice, because the radius of the Ti4+ ions (0.068 nm) is smaller than that of the Ce4+ ions (0.094 nm).23 The XPS spectra of Ce 3d for different catalysts are shown in Figure S3(A). The peaks labeled u are due to 3d3/2 spin
orbit states, and those labeled v are the corresponding 3d5/2 states.20 The u/v, u2/v2, and u3/v3 doublets represent the 3d104f0 state of Ce4+, while the doublet labeled u1/v1 represents the 3d104f1 initial electronic state corresponding to Ce3+.24 For pure CeO2, only the characteristic peaks attributed to Ce4+ were observed. In all CeTi catalysts, the peaks attributed to Ce4+ were still predominant. However, as shown in Figure S3(B), the small peak of v1 evidenced the presence of Ce3+ over CeTi catalysts. The presence of the Ce3+ could create charge imbalance, vacancies, and unsaturated chemical bonds on the catalyst surface,25 which lead to the increase of chemisorbed oxygen on the surface.17 The XPS spectra of Ce 3d, Ti 2p, and O 1s for different catalysts were fitted to a Gaussian model, and the surface atomic concentrations of the three elements were calculated accordingly. The results are shown in Table S2. For O 1s spectra, the peaks at less than 527.7
530.0 eV could be attributed to lattice oxygen (OA).18,25,26 Two shoulder peaks at the higher binding energy side likely belong to chemisorbed oxygen and/or weakly bonded oxygen species (OB) and oxygen in hydroxyl and/or surface adsorbed water (OC).18,26 As shown, the total oxygen concentration on the CeTi catalyst surface was much higher than that of pure TiO2 or pure CeO2. The higher surface oxygen concentration was attributed to the Ce3+ related chemisorbed and/or weakly bonded oxygen, which are the most active oxygen and play an important role in oxidation reactions.25 The atomic ratio of Ce/Ti was more than two times higher than the corresponding ratio calculated from preparation, indicating the richness of surface Ce. Ce1.5Ti with the highest surface Ce concentration showed the best Hg0 oxidation performance as will be demonstrated in a later section. Further increase of the CeO2/TiO2 ratio from 1.5 to 2.0 resulted in no increase of surface Ce concentration probably due to the formation of more cubic CeO2.

Figure 1. Hg0 oxidation efficiency over CeTi catalysts under simulated flue gas as a function of temperature.

Performance of CeTi Catalysts. Hg0 oxidation efficiencies over CeTi catalysts with different CeO2/TiO2 ratios as a function of temperature are shown in Figure 1. No obvious Hg0 oxidation was observed over pure TiO2, and the Eoxi in the entire temperature range was less than 10%. This is in line with an earlier study in which about 10% of Hg0 introduced was oxidized to Hg2+ over TiO2.27 The addition of CeO2 resulted in significant enhancement of Hg0 oxidation activity, e.g. about 50% Hg0 oxidation over Ce0.5Ti at 250 °C. Further increase of CeO2/TiO2 ratio to 1.5 yielded more Hg0 oxidation. Ce1.5Ti showed the highest Hg0 catalytic oxidation activity, with Eoxi higher than 90% from 200 to 250 °C. Further increase of CeO2/TiO2 ratio to 2.0 slightly lowered Eoxi. Pure CeO2 also exhibited some Hg0 oxidation capacity but much less than Ce1.5Ti, especially at low temperatures (120
250 °C). Above results demonstrate the synergy for Hg0 oxidation when CeO2 and TiO2 are combined. This is similar to other investigations18,19 where CeTi catalysts were more effective than pure TiO2 and pure CeO2 for selective catalytic reduction of NOx by NH3. For all CeTi catalysts, Eoxi increased with temperature from 120 to 250 °C and then decreased dramatically when temperature further increased from 250 to 400 °C. Since TiO2 is essentially inactive for the oxidation of Hg0,27 Hg0 oxidation observed was attributed to the activity of CeO2 supported on TiO2. The appearance of Ce3+ on the catalyst surface resulted in more surface chemisorbed oxygen,17 which was probably responsible for the excellent Hg0 oxidation performance of CeTi catalysts. It should be noted that the composition of simulated flue gas was in the range of those from burning lowrank coals and the space velocity of 60,000 h
1 was much higher than typical space velocities (2000
4000 h
1)28 in power plant SCR reactors. Even though the contacting conditions in fix-bed is better than that of the actual SCR reactor, the results still imply that applications of the CeTi catalysts are likely beneficial to Hg0 oxidation for coal-fired power plants, because both higher HCl concentration in high-rank coal combustion flue gases and lower space velocity have been reported to facilitate Hg0 conversion.10,11 Moreover, high activities of CeTi catalysts at low temperatures (150
250 °C) allow them to be placed downstream of PM control devices, where deactivation by exposure to high concentrations of fly ash is avoided. 7396

dx.doi.org/10.1021/es2007808 |Environ. Sci. Technol. 2011, 45, 7394–7400

Environmental Science & Technology

ARTICLE

where Cl* denotes an active chlorine species for oxidizing Hg0. With the aid of gas-phase O2, 10 ppm HCl resulted in Eoxi of 100%, indicating CeTi catalysts are extremely effective for Hg0 heterogeneous oxidation by HCl, since homogeneous Hg0 oxidation is hindered by a very high energy barrier.5 Ce3+ species on the CeTi catalyst surface was at least partly responsible for the high Hg0 oxidation activity, because the surface chemisorbed oxygen attributed to Ce3+ has been reported to be the most active oxygen.25,31 The presence of gas-phase O2 replenished the consumed chemisorbed oxygen, regenerated the lattice oxygen, and hence maintained the high surface oxygen concentration. Therefore, more HCl can be oxidized to form active chlorine species in the presence of gas-phase O2. Hg0 was thought to first react with active chlorine species forming HgCl, which was further oxidized to HgCl2 by another active chlorine species.32 The possible heterogeneous reactions over CeTi catalysts are proposed to be as follows 2HCl þ O f 2Cl þ H2 O ð4Þ

Figure 2. Effect of individual flue gas components and Hg0 oxidation without HCl at 200 °C.

Effect of Individual Flue Gas Components. Individual flue gas components balanced in pure N2 or N2 plus O2 were used to study their effect on Hg0 oxidation and the reaction pathways. The results are summarized in Figure 2. Effect of O2. Eoxi over Ce1.5Ti at 200 °C under pure N2 gas flow was observed to be 27.4%, which is 4 times larger than the Eoxi observed over V2O5
SiO2 catalyst under similar flue gas conditions.22 This is also superior to the performance of a V2O5 based catalyst reported in another study, where no obvious adsorption and oxidation of Hg0 was observed when a gas stream of Hg0/N2 passed through the catalyst.29 The loss of Hg0 on the Ce1.5Ti under pure N2 atmosphere was due to the reaction between Hg0 with stored oxygen (including lattice oxygen and chemisorbed oxygen). No obvious increase of Eoxi was detected when 4% gas-phase O2 was introduced to the gas flow or even when O2 concentration further increased to 20%. This insensitivity indicates CeTi catalyst has large oxygen storage capacity, and only a small fraction of the stored oxygen has been consumed in Hg0 oxidation during our experimental period. Effect of HCl. HCl is the most important flue gas species responsible for Hg0 oxidation since the main oxidized mercury species in coal combustion flue gas exists as HgCl2.14 However, in the absence of gas-phase O2, HCl not only cannot oxidize Hg0 but also greatly suppresses Hg0 adsorption on V2O5 based catalysts.13,30 In a reducing environment, Ce4+/Ce3+ shift over CeTi catalysts can release oxygen, which is necessary for Hg0 oxidation by HCl. As shown in Figure 2, HCl exhibited a promotional effect on Hg0 oxidation over CeTi catalyst. Ten ppm HCl balanced in N2 resulted in Eoxi of 48.5%, which is higher than the 27.4% Hg0 oxidation under pure N2 condition. Eoxi of 81.9% was observed when HCl concentration further increased to 30 ppm. In the absence of gas-phase O2, the reaction between HCl and CeTi catalysts occurs through

2HCl þ 2CeO2 T Ce2 O3 þ H2 O þ 2Cl

ð3Þ

Cl þ Hg 0 f HgCl

ð5Þ

HgCl þ Cl f HgCl2

ð6Þ

where O* represents chemisorbed or lattice oxygen on the surface of CeTi catalysts, which can be replenished or regenerated by gas-phase O2, respectively. The entire reaction can be written as follows Hg 0 þ 2HCl þ

1 CeTi O2 sf HgCl2 þ H2 O 2

ð7Þ

Effect of NO. Promotional effect of NO on Hg0 oxidation was

observed over CeTi catalyst. 100 ppm NO resulted in Eoxi of 37.7%. Increasing NO concentration to 300 ppm did not change Eoxi substantially. The presence of Ce3+ on CeTi catalysts was reported to promote the oxidation of NO to NO2,18 and NO2 was demonstrated to have a promotional effect on Hg0 oxidation over catalyst22 and fly ash.33 The Eoxi observed in the presence of NO was just slightly larger than that in its absence. This demonstrates that oxygen on CeTi catalysts is almost as effective as NO2 for Hg0 oxidation. In our short-term experiment, oxygen stored on the CeTi catalyst was enough for NO and Hg0 oxidation. Therefore, the addition of 4% O2 into flue gas containing 300 ppm NO only resulted in a minor increase of Eoxi. It should be noted that Hg2+ was observed in the gas flow downstream of the CeTi catalyst when flue gas contained NO, indicating that NO participated in the Hg0 oxidation reaction and volatile mercuric compounds, like Hg(NO3)2, were formed. The pathway summarized below is likely responsible for Hg0 oxidation in the presence of NO 1 CeTi O2 sf NO2 2

ð8Þ

Hg 0 þ NO2 f HgO þ NO

ð9Þ

NO þ

Hg 0 þ 2NO2 þ O2 f HgðNO3 Þ2

ð10Þ

Effect of SO2. In the absence of O2, Hg0 oxidation was greatly

inhibited by SO2. Almost no Hg0 oxidation was observed in gas flow containing 400 or 1200 ppm SO2 balanced in N2. It is very 7397

dx.doi.org/10.1021/es2007808 |Environ. Sci. Technol. 2011, 45, 7394–7400

Environmental Science & Technology

ARTICLE

Figure 3. Desorption of Hg0 from Ce1.5Ti by SO2.

likely that Hg0 oxidation over CeTi catalysts follows the Langmuir
Hinshelwood mechanism, where reactive species on catalyst surface react with adjacently adsorbed Hg0 to form Hg2+.26,29 If SO2 competes with Hg0 for active sites, it can greatly inhibit Hg0 adsorption and subsequent oxidation. The competitive adsorption of SO2 and Hg0 on the CeTi catalyst was demonstrated by a desorption experiment, the results of which are shown in Figure 3. Ce1.5Ti saturated by Hg0 at 200 °C under a flow of Hg0 balanced in N2 was used in this test. A spike of Hg0 was observed after cutting off Hg0 and adding 400 ppm SO2 at the same time, while Hg0 concentration decreased gradually after stopping Hg0 without adding SO2. The result demonstrates that both Hg0 and SO2 competed for the active sites, and the affinity between SO2 and CeTi catalyst was stronger than that between Hg0 and CeTi catalyst. In the gas flow containing 1200 ppm SO2 and 4% O2, Eoxi was observed to be 52.6%, which is higher than that without SO2. This indicates that SO2 has a promotional effect on Hg0 oxidation with the aid of O2. In the presence of gasphase O2, abundant chemisorbed oxygen was generated due to Ce3+ related charge imbalance. SO2 was oxidized by chemisorbed oxygen to form SO3, which constituted new chemisorption sites for Hg0 and could react with Hg0 to produce HgSO4.34 The reaction process is as follows SO2 þ

1 CeTi O2 sf SO3 2

Hg 0 þ SO3 þ

1 CeTi O2 sf HgSO4 2

ð11Þ ð12Þ

Effect of H2O. H2O has been reported to inhibit Hg0 oxidation

and removal over metal or metal oxide based catalysts due to competitive adsorption.22,32 Inhibitive effect of H2O on Hg0 oxidation was also observed over the CeTi catalyst. However, compared with our previous research on V2O5 based catalyst22 and another research on gold based catalyst,32 the reduction of Eoxi over CeTi catalyst was minor. This result indicates that the CeTi catalyst has good resistance to H2O. This is very beneficial for application in actual flue gas atmosphere, where H2O is inevitable. Identification of Hg0 Oxidation Mechanism. Several mechanisms including the Deacon process, the Mars-Maessen mechanism, the Langmuir
Hinshelwood mechanism, and the

Figure 4. Hg0 breakthrough curves over HCl pretreated Ce1.5Ti (A: effect of pretreatment temperature; B: effect of reaction temperature).

Eley
Rideal mechanism have been proposed for heterogeneous Hg0 oxidation.6 However, none has been verified as the dominant mechanism for catalytic Hg0 oxidation. It is believed that the mechanism involved depends on catalyst type, flue gas atmosphere, and even reaction temperature. Ce1.5Ti pretreated with HCl was chosen to identify the mechanism involved in Hg0 oxidation since HCl is the most effective flue gas component for Hg0 oxidation over CeTi catalyst, and CeO2, being the active component of CeTi catalysts, can adsorb and react with HCl to form active oxychloride.35 As shown in Figure 4(A), HCl pretreatment had a promotional effect on Hg0 oxidation at 200 °C. After passing through Ce1.5Ti pretreated at 200 °C, normalized Hg0 concentration swiftly dropped to about 0.3, which was much lower than that observed over nonpretreated Ce1.5Ti. However, the outlet Hg0 concentration quickly climbed back to the same level as that observed over nonpretreated Ce1.5Ti. This demonstrates that only a limited amount of HCl reacted with Ce1.5Ti catalyst to form active chlorine species. Ce1.5Ti pretreated at 300 °C showed better performance compared with Ce1.5Ti pretreated at 200 °C, indicating CeTi catalysts are more reactive with HCl at higher temperature. This also implies that the chlorine species on the CeTi catalyst was stable at 7398

dx.doi.org/10.1021/es2007808 |Environ. Sci. Technol. 2011, 45, 7394–7400

Environmental Science & Technology 300 °C. Otherwise, the significant loss of Hg0 would not be observed over the Ce1.5Ti pretreated at 300 °C since most of the chlorine species would have been decomposed during the flushing process in the pretreatment. This deduction is in line with other research that cerium oxychloride, which was probably responsible for Hg0 oxidation over CeTi catalyst, is thermally stable even at 777 °C.35 To study the effect of temperature on reactions between Hg0 and active chlorine species, Ce1.5Ti was first pretreated at 300 °C by HCl and then used for Hg0 oxidation at different temperatures under N2 atmosphere. As illustrated in Figure 4(B), Hg0 oxidation was facilitated at lower temperature, which is more suitable for Hg0 adsorption. Since reaction rates generally increase as temperature increases, more Hg0 oxidation should be observed at higher temperature if gas-phase Hg0 can react with active chlorine species over HCl pretreated Ce1.5Ti. Accordingly, it can be deduced that Hg0 oxidation over CeTi catalysts probably occurs through active surface species reacting with adsorbed Hg0, the process of which is the Langmuir
Hinshelwood mechanism. With the Langmuir
Hinshelwood mechanism, it is possible to explain the Hg0 oxidation performance over CeTi catalysts at different temperatures as shown in Figure 1. At low temperatures (120 to 250 °C), Hg0 adsorption occurred over CeTi catalysts (demonstrated by Hg0 desorption over CeTi catalysts after adsorption of Hg0 under pure N2 atmosphere), and Eoxi increased along with the increase of temperature since CeTi catalysts are more active for generating reactive species at higher temperature. The significant reduction of Eoxi with temperature increase from 250 to 400 °C was due to the inhibition of physical adsorption of Hg0 by high temperature. Hg0 Oxidation without HCl. Besides HCl, NO and SO2 were found to have promotional effects on Hg0 oxidation over the CeTi catalyst in the presence of O2. Set IV experiments were designed to reveal the possibility of effective Hg0 oxidation without HCl. The results are plotted in Figure 2. In the absence of O2, introduction of 400 ppm SO2 into gas flow containing 300 ppm NO resulted in a reduction of Eoxi from 38.3% to 6.5%. The competitive adsorption between Hg0 and SO2 was responsible for this deactivation. Further increase of SO2 concentration to 1200 ppm yielded no further inhibitive effect. In the presence of 4% O2, an addition of 400 ppm SO2 into gas flow containing 300 ppm NO increased Eoxi from 45.6% to 64.7%. With the aid of O2, the combination of 300 ppm NO and 1200 ppm SO2 resulted in Eoxi of 99.9%, implying that Hg0 can be effectively oxidized over CeTi catalysts without HCl. As demonstrated, application of CeTi catalysts is of tremendous value for Hg0 oxidation under low-rank coal combustion flue gas, which usually comes with low concentration of HCl. As shown in Figure 2, 300 ppm NO and 4% O2 resulted in Eoxi of 45.6%. Meanwhile 52.4% Hg0 oxidation was observed under 1200 ppm SO2 plus 4% O2. In the presence of O2, the promotional effect of NO and SO2 on Hg0 oxidation can be added together, since Eoxi of 99.9% under 300 ppm NO and 1200 ppm SO2 is almost equal to the sum of Eoxi’s under 300 ppm NO and 1200 ppm SO2, respectively. Even though SO2 was reported to have an inhibitive effect on the formation of nitrate36 which is active for Hg0 oxidation, no adverse effect of SO2 on Hg0 oxidation by NO was observed in the presence of O2. The inhibitive effect of SO2 in the absence of O2 was due to SO2 constraining the physical adsorption of Hg0, which is crucial for Hg0 oxidation through the Langmuir
Hinshelwood mechanism.

ARTICLE

’ ASSOCIATED CONTENT

bS

Supporting Information. Synthesis of catalysts, material characterization methods, mercury speciation conversion system, two tables, and three figures. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Phone: 352-392-0845. Fax: 352-392-3076. E-mail: cywu@ufl.edu.

’ ACKNOWLEDGMENT This project was partially supported by the National Key Basic Research and Development Program (973) (No. 2010CB227003) and the China Scholarship Council (CSC). We thank VICI Metronics, Inc. for supplying the Hg permeation device. ’ REFERENCES (1) Pavlish, J. H.; Sondreal, E. A.; Mann, M. D.; Olson, E. S.; Galbreath, K. C.; Laudal, D. L.; Benson, S. A. Status review of mercury control options for coal-fired power plants. Fuel Process. Technol. 2003, 82, 89–165. (2) Milford, J. B.; Pienciak, A. After the clean air mercury rule: prospects for reducing mercury emissions from coal-fired power plants. Environ. Sci. Technol. 2009, 43, 2669–2673. (3) U.S. Environmental Protection Agency, Air Toxics Standards for Utilities. http://www.epa.gov/ttn/atw/utility/utilitypg.html (accessed 26.05.11). (4) Zhuang, Y.; Thompson, J. S.; Zygarlicke, C. J.; Pavlish, J. H. Development of a mercury transformation model in coal combustion flue gas. Environ. Sci. Technol. 2004, 38, 5803–5808. (5) Galbreath, K. C.; Zygarlicke, C. J. Mercury speciation in coal combustion and gasification flue gases. Environ. Sci. Technol. 1996, 30, 2421–2426. (6) Presto, A. A.; Granite, E. J. Survey of catalysts for oxidation of mercury in flue gas. Environ. Sci. Technol. 2006, 40, 5601–5609. (7) Lee, W.; Bae, G.-N. Removal of elemental mercury (Hg(0)) by nanosized V2O5/TiO2 catalysts. Environ. Sci. Technol. 2009, 43, 1522–1527. (8) Li, J. F.; Yan, N. Q.; Qu, Z.; Qiao, S. H.; Yang, S. J.; Guo, Y. F.; Liu, P.; Jia, J. P. Catalytic oxidation of elemental mercury over the modified catalyst Mn/alpha-Al2O3 at lower temperatures. Environ. Sci. Technol. 2010, 44, 426–431. (9) Kamata, H.; Ueno, S-i.; Sato, N.; Naito, T. Mercury oxidation by hydrochloric acid over TiO2 supported metal oxide catalysts in coal combustion flue gas. Fuel Process. Technol. 2009, 90, 947–951. (10) Niksa, S.; Fujiwara, N. A predictive mechanism for mercury oxidation on selective catalytic reduction catalysts under coal-derived flue gas. J. Air Waste Manage. Assoc. 2005, 55, 1866–1875. (11) Senior, C. L. Oxidation of mercury across selective catalytic reduction catalysts in coal-fired power plants. J. Air Waste Manage. Assoc. 2006, 56, 23–31. (12) 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, 8136–8141. (13) Eswaran, S.; Stenger, H. G. Understanding mercury conversion in selective catalytic reduction (SCR) catalysts. Energy Fuels 2005, 19, 2328–2334. (14) Cao, Y.; Gao, Z.; Zhu, J.; Wang, Q.; Huang, Y.; Chiu, C.; Parker, B.; Chu, P.; Pan, W. P. Impacts of halogen additions on mercury oxidation, in a slipstream selective catalyst reduction (SCR), reactor when burning sub-bituminous coal. Environ. Sci. Technol. 2008, 42, 256–261. 7399

dx.doi.org/10.1021/es2007808 |Environ. Sci. Technol. 2011, 45, 7394–7400

Environmental Science & Technology

ARTICLE

(15) Worle-Knirsch, J. M.; Kern, K.; Schleh, C.; Adelhelm, C.; Feldmann, C.; Krug, H. F. Nanoparticulate vanadium oxide potentiated vanadium toxicity in human lung cells. Environ. Sci. Technol. 2007, 41, 331–336. (16) Casapu, M.; Kr€ocher, O.; Elsener, M. Screening of doped MnOx-CeO2 catalysts for low-temperature NO-SCR. Appl. Catal., B 2009, 88, 413–419. (17) Chen, L.; Li, J. H.; Ge, M. F. Promotional effect of ce-doped V2O5-WO3/TiO2 with low vanadium loadings for selective catalytic reduction of NOx by NH3. J. Phys. Chem. C 2009, 113, 21177–21184. (18) Gao, X.; Jiang, Y.; Zhong, Y.; Luo, Z.; Cen, K. The activity and characterization of CeO2-TiO2 catalysts prepared by the sol-gel method for selective catalytic reduction of NO with NH3. J. Hazard. Mater. 2010, 174, 734–739. (19) Xu, W.; Yu, Y.; Zhang, C.; He, H. Selective catalytic reduction of NO by NH3 over a Ce/TiO2 catalyst. Catal. Commun. 2008, 9, 1453–1457. (20) Reddy, B. M.; Khan, A.; Yamada, Y.; Kobayashi, T.; Loridant, S.; Volta, J. C. Structural characterization of CeO2-TiO2 and V2O5/CeO2TiO2 catalysts by raman and XPS techniques. J. Phys. Chem. B 2003, 107, 5162–5167. (21) Long, R. Q.; Yang, R. T. The promoting role of rare earth oxides on Fe-exchanged TiO2-pillared clay for selective catalytic reduction of nitric oxide by ammonia. Appl. Catal., B 2000, 27, 87–95. (22) Li, Y.; Murphy, P. D.; Wu, C.-Y.; Powers, K. W.; Bonzongo, J-C. J. Development of silica/vanadia/titania catalysts for removal of elemental mercury from coal-combustion flue gas. Environ. Sci. Technol. 2008, 42, 5304–5309. (23) Luo, M.; Chen, J.; Chen, L.; Lu, J.; Feng, Z.; Li, C. Structure and redox properties of CexTi1-xO2 solid solution. Chem. Mater. 2000, 13, 197–202. (24) Mullins, D. R.; Overbury, S. H.; Huntley, D. R. Electron spectroscopy of single crystal and polycrystalline cerium oxide surfaces. Surf. Sci. 1998, 409, 307–319. (25) Yang, S.; Zhu, W.; Jiang, Z.; Chen, Z.; Wang, J. The surface properties and the activities in catalytic wet air oxidation over CeO2TiO2 catalysts. Appl. Surf. Sci. 2006, 252, 8499–8505. (26) Eom, Y.; Jeon, S. H.; Ngo, T. A.; Kim, J.; Lee, T. G. Heterogeneous mercury reaction on a selective catalytic reduction (SCR) catalyst. Catal. Lett. 2008, 121, 219–225. (27) 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, 2441–2444. (28) Laudal, D. L. T., J. S.; Pavlish, J. H.; Brickett, L; Chu, P.; Srivastava, R. K.; Lee, C. W.; Kilgroe, J. D. Mercury speciation at power plants using SCR and SNCR control technologies. 3rd International Air Quality Conference, Arlington, VA, September, 2002. (29) He, S.; Zhou, J.; Zhu, Y.; Luo, Z.; Ni, M.; Cen, K. Mercury oxidation over a vanadia-based selective catalytic reduction catalyst. Energy Fuels. 2008, 23, 253–259. (30) Straube, S.; Hahn, T.; Koeser, H. Adsorption and oxidation of mercury in tail-end SCR-DeNOx plants - Bench scale investigations and speciation experiments. Appl. Catal., B 2008, 79, 286–295. (31) Lin, S. S.; Chen, C. L.; Chang, D. J.; Chen, C. C. Catalytic wet air oxidation of phenol by various CeO2 catalysts. Water Res. 2002, 36, 3009–3014. (32) Zhao, Y. X.; Mann, M. D.; Pavlish, J. H.; Mibeck, B. A. F.; Dunham, G. E.; Olson, E. S. Application of gold catalyst for mercury oxidation by chlorine. Environ. Sci. Technol. 2006, 40, 1603–1608. (33) Norton, G. A.; Yang, H.; Brown, R. C.; Laudal, D. L.; Dunham, G. E.; Erjavec, J. Heterogeneous oxidation of mercury in simulated post combustion conditions. Fuel 2003, 82, 107–116. (34) Fan, X.; Li, C.; Zeng, G.; Gao, Z.; Chen, L.; Zhang, W.; Gao, H. Removal of gas-phase element mercury by activated carbon fiber impregnated with CeO2. Energy Fuels. 2010, 24, 4250–4254. (35) Onstott, E. I. Cerium dioxide as a recycle reagent for thermochemical hydrogen production by splitting hydrochloric acid into the elements. Int. J. Hydrogen Energy 1997, 22, 405–408. (36) Xu, W.; He, H.; Yu, Y. Deactivation of a Ce/TiO2 catalyst by SO2 in the selective catalytic reduction of NO by NH3. J. Phys. Chem. C 2009, 113, 4426–4432. 7400

dx.doi.org/10.1021/es2007808 |Environ. Sci. Technol. 2011, 45, 7394–7400