A novel pre-oxidation method for elemental mercury removal utilizing a complex vaporized absorbent

Journal of Hazardous Materials 280 (2014) 118–126

Contents lists available at ScienceDirect

Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat

A novel pre-oxidation method for elemental mercury removal utilizing a complex vaporized absorbent Yi Zhao ∗ , Runlong Hao, Qing Guo School of Environmental Science and Engineering, North China Electric Power University, Baoding 071003, P.R. China

h i g h l i g h t s

g r a p h i c a l

a b s t r a c t

• An innovative liquid-phase complex absorbent (LCA) for Hg0 removal was prepared. • A novel integrative process for Hg0 removal was proposed. • The simultaneous removal efficiencies of SO2 , NO and Hg0 were 100%, 79.5% and 80.4%, respectively. • The reaction mechanism of simultaneous removal of SO2 , NO and Hg0 was proposed.

a r t i c l e

i n f o

Article history: Received 17 April 2014 Received in revised form 4 July 2014 Accepted 30 July 2014 Available online 7 August 2014 Keywords: Pre-oxidation Vaporized liquid-phase complex absorbent Hg0 removal Simultaneous removal of SO2 , NO and Hg0 Reaction mechanism of simultaneous removal of SO2 , NO and Hg0

a b s t r a c t A novel semi-dry integrative method for elemental mercury (Hg0 ) removal has been proposed in this paper, in which Hg0 was initially pre-oxidized by a vaporized liquid-phase complex absorbent (LCA) composed of a Fenton reagent, peracetic acid (CH3 COOOH) and sodium chloride (NaCl), after which Hg2+ was absorbed by the resultant Ca(OH)2 . The experimental results indicated that CH3 COOOH and NaCl were the best additives for Hg0 oxidation. Among the influencing factors, the pH of the LCA and the adding rate of the LCA significantly affected the Hg0 removal. The coexisting gases, SO2 and NO, were characterized as either increasing or inhibiting in the removal process, depending on their concentrations. Under optimal reaction conditions, the efficiency for the single removal of Hg0 was 91%. Under identical conditions, the efficiencies of the simultaneous removal of SO2 , NO and Hg0 were 100%, 79.5% and 80.4%, respectively. Finally, the reaction mechanism for the simultaneous removal of SO2 , NO and Hg0 was proposed based on the characteristics of the removal products as determined by X-ray diffraction (XRD), atomic fluorescence spectrometry (AFS), the analysis of the electrode potentials, and through data from related research references. © 2014 Published by Elsevier B.V.

1. Introduction Due to its high toxicity, volatility, and bioaccumulation [1], Hg0 emitted from coal-fired power plants has become a major concern in recent years. In 2005, the U.S. Environmental Protection Agency (EPA) issued the Clean Air Mercury Rule (CAMR) aimed at

∗ Corresponding author. Tel.: +86 312 7522343; fax: +86 312 7522192. E-mail address: [email protected] (Y. Zhao). http://dx.doi.org/10.1016/j.jhazmat.2014.07.061 0304-3894/© 2014 Published by Elsevier B.V.

reducing the Hg0 emissions from coal-fired power plants [2]. China, as the country having the greatest amount of mercury emissions in the world, has also taken many measures to control the Hg0 emissions from coal-fired power plants. Mercury found in coal-fired flue gases is often presented as three chemical forms: elemental mercury (Hg0 ), oxidized mercury (Hg2+ ) and particle bound mercury (Hgp ) [3]. The latter two can be partially removed by Wet Flue Gas Desulfurization (WFGD) systems, fabric filters (FF) or electrostatic precipitation (ESP) systems. While Hg0 , due to its high volatility and low solubility in water, is difficult to be collected through the

Y. Zhao et al. / Journal of Hazardous Materials 280 (2014) 118–126

use of existing Air Pollution Control Devices (APCDs), its removal has become an important focus of research in recent years. In recent years, there are a lot of methods, including adsorption, homogeneous oxidation and heterogeneous oxidation, have been extensively researched to remove Hg0 [4]. For adsorption method, fly ash [5], activated carbon, etc. [6,7], were used to effective remove Hg0 . The current state of the art technology for the capture of Hg0 from flue gases utilizes activated carbon injection (ACI) [8]. However, ACI has the disadvantages of very high costs (about $US 3,810 to $US 166,000 per pound of Hg0 ) [9] and secondary environmental problems. In addition, metallic sorbents, especially Au and Pd, are also potential alternatives for Hg adsorption as well as oxidation. Wilcox and Lim [10,11] have done a lot of works on Hg0 removal by gold-based absorbents, and obtained satisfactory results. And Sasmaz et al. [12] has determined the binding mechanism of Hg0 on noble metals to well understand the surface reactivity of Hg0 on noble metals. As for the oxidation method, the core of which is to rapidly convert the Hg0 into soluble Hg2+ , the methods of Hg0 oxidation include catalytic oxidation, liquid-phase oxidation, gas phase oxidation and light chemical oxidation. Among these methods, the catalyst used in the process of selective catalytic reduction (SCR) has the ability to oxidize Hg0 [13], especially in the presence of halogens [14]. Fan et al., [15] carried out an experiment of CeO2 /HZSM-5 catalytic oxidization of Hg0 , and found that the acidic site on the surface of the HZSM-5 exhibited a strong ability to adsorb Hg0 , with CeO2 as a key component. The SO2 and NO increased the Hg0 oxidation in the presence of O2 , while the H2 O vapor inhibited the Hg0 removal process. The highest removal efficiency of Hg0 was obtained when the reaction temperature was below 300 ◦ C and the loading of CeO2 was 6%. Hutson et al. [16] carried out an experiment on the simultaneous removal of SO2 , NOx and Hg0 from coal-fired flue gas, using a NaClO2 -enhanced wet scrubber, from which a novel method for the simultaneous removal of multi-pollutants was proposed. In addition, Fang et al., [17] adopted urea/KMnO4 as an absorbent to investigate the effects of various factors on the simultaneous removal of Hg0 , SO2 and NO. These factors included urea concentration, KMnO4 concentration, inlet Hg0 concentration, initial pH, reaction temperature, SO2 concentration, and NO concentration. Their experimental results indicated that the removal of NO and Hg0 depended primarily on the KMnO4 concentrations. Halogen-containing compounds have been employed as gas phase oxidants to oxidize Hg0 . Yan et al. [18,19] synthesized S2 Cl2 , SCl2 and S2 Br2 to carry out Hg0 oxidation, and found that Hg0 could be oxidized faster under the synergistic effect of fly ash. Jia et al. [20] found that Hg0 oxidation could be enhanced by UV light in the presence of CH4 , and the Hg0 removal efficiency was 65.5% under a 253.7 nm light. Nevertheless, these methods cannot be applied in industrial engineering applications, due to various defects such as low removal efficiency, the possibility of secondary environmental problems, and the excessive costs. To address these drawbacks, this work has studied a Fenton-Reagent based, liquid-phase complex absorbent (LCA) designed to remove Hg0 . It has been recognized that H2 O2 was a promising reagent for flue gas purification due to its superior environmental friendliness, and its lower price. However, prior attempts at Hg0 removal with H2 O2 have been unsatisfactory [21]. A similar phenomenon was also observed in the Hg0 oxidation by hydroxyl radicals derived from H2 O2 [22]. In order to improve the performance of the Fenton methodology for Hg0 removal, in this paper’s research, a vaporized liquid-phase complex absorbent (LCA) composed of a Fenton reagent, CH3 COOOH ( ) [23] and NaCl [18,19] was prepared to effectively oxidize the Hg0 . In addition, in order to make it possible that the low pH LCA can work together with the WFGD or CFB-FGD systems, we designed a two-stage-treatment process of the preoxidation combined with

119

the absorption to remove Hg0 , in which, the vaporized LCA initially oxidized Hg0 in a preoxidation device, and then the Hg2+ was absorbed by the followed alkaline industrial Ca(OH)2 in CFB-FGD system or CaCO3 slurry in the WFGD system. To the authors’ knowledge, there have been no reports in the field of Hg0 removal, on this type of application of the LCA, as well as the novel flue gas stagetreatment. For the development of this novel method, the optimal preparation conditions of the LCA and the best reaction conditions, were established based on investigations of the effects on the Hg0 removal efficiency: the different halogen additives, the pH of the LCA solution, the adding rate of the LCA, the reaction temperatures, and the initial concentrations of Hg0 , O2 , NO and SO2 . The experimental results indicated that the proposed methodology exhibited a satisfactory performance on the Hg0 removal, and on the simultaneous removal of SO2 , NO and Hg0 , as far as this application was concerned. Therefore, these research results have achieved not only substantial academic significance, but also offer important value for real-world applications. 2. Experimentation 2.1. Experimental apparatus The experiments were conducted on a fixed-bed system that was made up of simulated flue gas generation, LCA vaporization, integration of pre-oxidation and absorption, and tail gas detection, as shown in Fig. 1. Compressed gas cylinders (Fig. 1, 1–5) (North Special Gas Company, Baoding, China) generated simulated flue gas of N2 , SO2 , NO, O2 and CO2 , and the Hg0 was generated by a mercury osmotic tube (20 ng/min, VICI Metronics Co., USA) (Fig. 1, 8) heated in a thermostatic water bath (Fig. 1, 10) (HH-ZK2, Yuhua Instrumental Company, Gongyi, China) with 1 l/min of N2 as the carrier gas. Because the Hg0 concentrations in actual flue gas emitted from typical coal-fired power plants in China are mainly distributed about 20 ␮g/m3 [24], thus Hg0 concentration used in our experiments was determined as 20 ␮g/m3 . A peristaltic pump, number BT1001F from Longerpump in Baoding, China (Fig. 1, 17), was used to add the LCA into a custom-designed vaporization device (Fig. 1, 11) that was heated by a thermally controlled electric heater (Fig. 1, 12) (ZDHW, Zhongxingweiye Company, Beijing, China). In order to avoid Hg0 adsorption or oxidation across surfaces, the reactor was a cylindrical quartz tube (Fig. 1, 14) with a length of 30 cm and an inner diameter of 3.2 cm, heated by a tube-type resistance furnace (Fig. 1, 15) (DC-RB, Duchuang Technology Company, Beijing, China), the pipe lines used in the reaction system were made by Teflon and heated by heater bands. The tail gas was detected by a flue gas analyzer (Fig. 1, 21) (ECOM-J2KN, RBR Company, Germany) and an Atomic Fluorescence Mercury Detector (Fig. 1, 22) (QM201, SuzhouQingan Company, Suzhou, China). Internal temperatures of the vaporization device (Fig. 1, 11) and the reactor (Fig. 1, 14) were measured over time, by three digital regulators and three thermocouples (Fig. 1, 13) (XMTD, Baoding, China). 2.2. Reagents and LCA vaporization All reagents used were analytical reagents from the Kermel Company of Tianjin, China. The LCA was prepared with H2 O2 of 30.0% (w/w), PAA of 16.0% (w/w), FeSO4 ·7H2 O of 99–101% (w/w), HCl of 1.0 mol/l, NaCl of 99.5% (w/w), NaBr of 99.0% (w/w) and NaClO of 10% (w/w) The absorbent was Ca(OH)2 , and the dryer was anhydrous CaCl2. The method used for mercury sampling was OHM method recommended by the US EPA, in which, the oxidized mercury that escaped from reactor was absorbed by 1 mol/l of KCl solution, the mercury in tail gas was treated by 10% (v/v) H2 SO4 –4% (w/w) KMnO4 before being discharged to atmosphere.

120

Y. Zhao et al. / Journal of Hazardous Materials 280 (2014) 118–126

Fig. 1. Schematic diagram of the experimental apparatus: 1–5: N2 , CO2 , SO2 , NO, O2 gas cylinders; 6: flow meters; 7: buffer bottle; 8: mercury osmotic tube; 9: tee joint; 10: thermostat water bath; 11: vaporization device; 12: thermal control electric heater; 13: digital regulators; 14: reactor; 15: tube type resistance furnace;16: Ca-based absorbent; 17: peristaltic pump; 18: LCA solution; 19: KCl solution; 20: dry tower; 21: flue gas analyzer; 22: cold atom fluorescence mercury detector; 23: 10% (v/v) H2 SO4 –4% (w/w) KMnO4 solution.

During the preparation of the LCA, fresh solutions of H2 O2 , FeSO4 PAA and a halogen compound were added to the beaker utilizing a pipette and shaken adequately. The corresponding concentrations of each component in the LCA were 4.0, 4.0, 1.0 and X mol/l. For the vaporization of the LCA, the solution (Fig. 1, 18) was pumped by a peristaltic pump (Fig. 1, 17) and carried simultaneously by simulated flue gas into the vaporization device (Fig. 1, 11), where it was immediately vaporized. 2.3. Experimental procedures and analytical methods During the experiments, the SO2 , NO and N2 were metered through a mass flow controller (6) and mixed with the Hg0 in a buffer bottle (Fig. 1, 7) so that the SO2 , NO and Hg0 were diluted to the desired concentrations by the N2 , from which the simulated flue gas was formed. The oxidation and absorption reactions were carried out when the mixture of vaporized LCA and simulated flue gas had entered into the reactor (Fig. 1, 14) filled with Ca(OH)2 supported on glass wool. The removal efficiencies were calculated according to the concentrations of Hg0 , SO2 and NO before and after reaction (Eq. (1)). =

Cin − Cout × 100% Cin

However, it can be seen from Fig. 2 that the Hg0 removal efficiency is about 33.2% when the Fenton reagent is the same as the oxidant, indicating that the removal of Hg0 by Fenton methodology is unsatisfactory [22]. Previous works have shown that peroxyacetic acid (PAA) is a promoter for H2 O2 , and halogens are favorable to Hg0 oxidation [18,19]. Hence, the effects on the Hg0 removal process, of the PAA and various halogen compounds such as HCl, NaCl, NaClO and NaBr were investigated. It is found from Fig. 2 that when the LCA is made up of a Fenton reagent, PAA and HCl, the Hg0 removal efficiencies increase from 38.3% to 56.5% with the concentrations of HCl increase from 8 to 40 mmol/l. When the LCA is prepared with a Fenton reagent, the PAA and NaCl, the removal efficiencies increase from 57.2% to 82.6% as the concentrations of NaCl increasing from 8 to 40 mmol/l, demonstrating that Cl− is beneficial to Hg0

(1)

where  is the removal efficiency of Hg0 , Cin the inlet concentration and Cout is the outlet concentration. The solution alkalinity was tested with a pH meter (type PHS-3C, Leici, Shanghai, China). X-ray diffraction (type XRD, D8 ADVANCE type, BRUKER-AXS, Germany) was used to characterize the removal products in fresh and spent absorbent. An atomic fluorescence spectrometer (AFS, A-620 type, Beijing Ruili Company, China) was used to determine the concentration of oxidized mercury in the KCl solution. 3. Results and discussion 3.1. Effects of different halogen compounds on Hg0 removal The standard electrode potentials of H2 O2 (1.770 V) and •OH (2.800 V) [25] are known to be higher than that of Hg2+ /Hg0 (0.796 V), which means that Hg0 may be oxidized by H2 O2 and •OH.

Fig. 2. Effects of different additives on Hg0 removal efficiency. Hg0 concentration, 20 ␮g/m3 ; reaction temperature, 90 ◦ C; LCA solution pH, 1.1; adding rate of LCA, 150 ␮l/min; flue gas velocity, 1 l/min.

Y. Zhao et al. / Journal of Hazardous Materials 280 (2014) 118–126

100

H2O2 concentration (mol/L) 90

0

1

2

3

4

5

6

121

7

8

9

80

Hg removal efficiency (%)

70 60

40

0

50

60

40

20

0

Hg removal efficiency (%)

80

30

Effect of NaCl, H2O2 (4mol/L)+PAA (1mol/L)

0

Effect of H2O2, PAA (1mol/L)+NaCl (40mmol/L)

20 0

8

16

24

32

40

48

56

64

72

80

0

1

Fig. 3. Effect of NaCl concentration on Hg0 removal efficiency. Hg0 concentration, 20 ␮g/m3 ; reaction temperature, 90 ◦ C; LCA solution pH, 1.1; adding rate of LCA, 150 ␮l/min; flue gas velocity, 1 l/min.

oxidation, and the increase caused by the NaCl is larger than that of the HCl, which may be attributed to the generations of chlorinecontaining oxidants, such as HOCl•, Cl• and Cl2 [26–28]. On the other hand, a sodium salt observed on the wall of the reactor might play an important role in capturing Hg0 [29]. Surprisingly, the Hg0 removal efficiencies increase from 75.4% to 92.1% as the NaBr concentrations increasing from 4 to 8 mmol/l. Obviously, bromine was more favorable to Hg0 oxidation when compared with chlorine, and this result was consistent with the results of previous studies [18,30]. Meanwhile, the significant enhancement of the PAA on the Hg0 removal is also proved by comparing the experimental results obtained from the LCA composed of the Fenton reagent and NaCl with those obtained from the LCA composed of the Fenton reagent, the PAA and NaCl. The effect of the NaClO on the Hg0 removal shows that Hg0 removal efficiencies increase from 68.0% to 83.0% when the NaClO concentrations range from 8 to 40 mmol/l. The increase caused by the NaClO on the Hg0 removal was almost equal to that of NaCl. However, in the interest of saving costs, since the prices of NaBr at 21,000 ¥/t, and NaClO at 3850 ¥/t, are both far higher than that of NaCl at 260 ¥/t, the NaCl was considered to be the most appropriate as an additive to the LCA. In order to determine the optimal concentration of the NaCl in the LCA, the effect of the NaCl concentration on the Hg0 removal was investigated experimentally. As shown in Fig. 3, the removal efficiencies increase from 58.4% to 82.5% while the NaCl concentrations is various from 8 to 40 mmol/l, and thereafter the efficiencies remain constant at 83.0%. Hence, the optimal concentration of NaCl for the LCA was determined to be 40 mmol/l. The promotion on Hg0 oxidation resulting from NaCl addition was more significant when NaCl concentration was in the range of 8–40 mmol/l. This was due to that the molar ratio of chlorine-containing oxidants to Hg0 increased as NaCl concentration increasing in LCA, which was beneficial for Hg0 oxidation. However, the generated chlorinecontaining oxidants would be saturated compared with Hg0 , when NaCl concentration continued to increase to 40 mmol/l. Thus, the promotion of NaCl addition was obviously weakened. Furthermore, the effect of H2 O2 concentration on Hg0 removal was investigated. Fig. 3 shows that when H2 O2 concentration increases from 1 to 4 mol/l, Hg0 removal efficiency sharply

3

4

5

LCA pH value

88

NaCl addition (mmol/L)

2

Fig. 4. Effect of LCA solution pH on Hg0 removal efficiency. Hg0 concentration, 20 ␮g/m3 ; reaction temperature, 90 ◦ C; adding rate of LCA, 150 ␮l/min; flue gas velocity, 1 l/min.

increases from 32% to 83%, indicating that the more H2 O2 exist in the LCA, the stronger oxidizability can be obtained. Thereafter, Hg0 removal efficiency is invariant with H2 O2 concentration further increasing. The reason was that when H2 O2 was over the desirable concentration, the reaction rate would be a control step for Hg0 oxidation. 3.2. The effect of the LCA solution alkalinity on the Hg0 removal process It has been accepted that the solution’s alkalinity has a significant effect on the stability of the ferrous and the oxidation potentials of the H2 O2 and the PAA. The experiments at different LCA solution pH levels were thus conducted to verify the influences of pH on the Hg0 removal. As shown in Fig. 4, the removal efficiencies reach above 80.0% when the solution pH is less than 1, but the removal efficiencies decline linearly from 82.3% to 15.6% when the solution pH increases from 1 to 4 and even the removal efficiency decreases to 8.0% when the pH is 5. With respect to the significant inhibition on the Hg0 removal as the solution pH increasing, it was speculated there were three explanations for this phenomenon. (1) The H2 O2 was decomposed intensely at higher pH levels (Eq. (2)) [31,32], and the decomposition product, HO2 − could then further promote the H2 O2 decomposition (Eq. (3)) [33] and consume the •OH (Eq. (4)). Therefore, the oxidizability of the reaction system had decreased. (2) The oxidizing ability of the PAA was increased when it was catalyzed by ferrous in an acidic environment [23], and the acetic acid resulting from the reduction or the decomposition of the PAA, could then participate in the reaction between the H2 O2 and the PAA (Eq. (5)), which might inhibit the decomposition of H2 O2 and maintain the PAA concentration level. On the contrary, however, the oxidizing ability of the PAA actually decreased as the pH level increased. (3) The catalytic function of the ferrous, which was the most important catalyst responsible for generating the radicals derived from the H2 O2 and the PAA, would be reduced due to the precipitation of iron as the solution pH level increased. H2 O2 → H+ + HO− 2 H2 O2 + HO− 2

(2) −

→ OH + O2 + H2 O

− • HO• + HO− 2 → OH + HO2

(3) (4)

122

Y. Zhao et al. / Journal of Hazardous Materials 280 (2014) 118–126

90

90

Hg Removal Efficiency (%)

100

Hg removal efficiency (%)

100

80

70

0

0

70

80

60

50

50 0

100

200

300

400

500

60

600

Fig. 5. Effect of adding rate of LCA on Hg0 removal efficiency. Hg0 concentration, 20 ␮g/m3 ; reaction temperature, 90 ◦ C; LCA solution pH, 1.1; flue gas velocity, 1 l/min.

CH3 COOH + H2 O2 → CH3 COOOH + H2 O

Generally, the reaction rate for an oxidation reaction depends on the concentration of the oxidants, which rely on the adding rate of the oxidants. Hence, the effect of the adding rate of the LCA on the Hg0 removal was studied experimentally, as shown in Fig. 5. The removal efficiencies rise from 61.1% to 82.2% while the adding rates increase from 50 to 150 ␮l/min. The removal efficiency is constant while the adding rate is between 150 and 450 ␮l/min. Thereafter, a slight decrease occurs. The interpretation of this experimental phenomenon could be that the molar ratio of the gas phase oxidants to the Hg0 increased as the LCA adding rate increasing, which led to an increase in the oxidation rate of the Hg0 . However, as the LCA adding rate was further increased, the quenching of the •OH and the decomposition of the H2 O2 might have been accelerated due to the excessive oxidants in the simulated flue gases (Eqs. (5)–(8)) [34], resulting in a decrease of the removal efficiency. In consideration of cost savings, 150 ␮l/min was determined to be the best adding rate for the LCA. HO• + HO• → H2 O2

(6)

HO2 • + HO• → H2 O + O2

(7)

HO2 + HO2 → H2 O2 + O2

80

90

100

110

120

130

140

150

160

)

Fig. 6. Effect of reaction temperature on Hg0 removal efficiency. Hg0 concentration, 20 ␮g/m3 ; LCA solution pH, 1.1; adding rate of LCA, 150 ␮l/min; flue gas velocity, 1 l/min.

(5)

3.3. The effect of the adding rate of the LCA on the Hg0 removal

•

70

Reaction Temperature (

Adding rate of LCA (µl/min)

•

60

(8)

3.4. The effect of the reaction temperature on the Hg0 removal process In chemical reactions, the reaction rate and the stability of the reactants are affected by the reaction temperature. Therefore, the effect of the reaction temperature on the Hg0 removal was investigated. In a typical coal-fired power plant, the flue gas temperatures in the tail of the power boiler are approximately 800–900 ◦ C. After passing the economizer, the air pre-heater, and the dust collection device, the flue gas temperatures at the inlet (FGD-CFB) are between 120 and 170 ◦ C [35], and then at the outlet the temperature they are approximately 70 ◦ C. Thus, the investigations of temperature were in a range between 70 and 150 ◦ C. As shown in Fig. 6, the Hg0 removal efficiencies increase linearly

from 60.1% to 82.3% as the temperatures increasing from 70 to 90 ◦ C, and increase slowly from 82.0% to 91.0% in the range of 90–130 ◦ C. Thereafter, the removal efficiencies decrease sharply. The increase in the removal efficiency might have resulted from the evaporation rate of the LCA; the diffusion rate of the gas phase oxidants, and the chemical reaction rate between the reactants and the Hg0 becoming effectively accelerated within a temperature range of 70–90 ◦ C. However, the increase was gradually inhibited in the temperature range of 90–130 ◦ C, due to the decompositions of the H2 O2 and the PAA. Apparently, the inhibiting effect was larger than the promoting effect when the temperature was above 130 ◦ C, owing to fierce heat decompositions of the oxidants, evidently leading to a decrease of the removal efficiency. Therefore, the optimal reaction temperature was selected to be 130 ◦ C. In order to investigate the heat decompositions of the main oxidants, the thermal stability experiments on H2 O2 and PAA were carried out under the optimal conditions except for the temperature, in which, H2 O2 of 4 mol/l was pumped into the vaporization device at a rate of 150 ␮l/min in the temperature range of 70–150 ◦ C with the carrier gas as N2 at a flow rate of 1.0 l/min, and then the outlet O2 concentrations were detected for indirectly reflecting H2 O2 decomposition. The results showed that O2 concentrations were 0.1% for the temperature range of 70–95 ◦ C, 0.2% for 95–132 ◦ C, 0.3% for 132–139 ◦ C and 0.4% for 139–150 ◦ C, and the corresponding decompositions of H2 O2 were 5%, 10%, 15% and 20%. Meanwhile, a similar experiment on PAA of 1 mol/l was also carried out under the identical conditions except for the carrier gas flow of 250 ml/min. The results were that O2 concentrations were constant at 0 in almost all temperatures except for O2 concentration of 0.1% in the temperature range of 134–150 ◦ C, and the corresponding decomposition rate was 5%. Thus, it can be concluded that the decompositions of H2 O2 and PAA would be accelerated as the temperature increased, especially in the temperature range of 132–150 ◦ C, which could account for the phenomenon that the removal efficiency of Hg0 decreased obviously in Fig. 6. For the industrial engineering of this method, the selection of reasonable reaction temperature is necessary on promise of meeting the Hg emission standard, in order to avoid the excess decomposition at the high reaction temperature.

Y. Zhao et al. / Journal of Hazardous Materials 280 (2014) 118–126

123

100 1000

LCA added Fluorescence value Removal efficiency

400

40

Remove NO2 20

200

0

0

30

60

90

120

150

0

60

600

Hg removal efficiency (%)

Fluorescence value

80

NO2 input

800

0 180

Reaction time (min) Fig. 8. Transient response of NO2 during LCA oxidation Hg0 . Reaction temperature, 130 ◦ C; LCA solution pH, 1.1; adding rate of LCA, 150 ␮l/min; flue gas velocity, 1 l/min.

0

0

Fig. 7. Effects of initial concentrations of Hg , O2 , SO2 and NO on Hg removal efficiency. Reaction temperature, 130 ◦ C; LCA solution pH, 1.1; adding rate of LCA, 150 ␮l/min; flue gas velocity, 1 l/min.

3.5. Effects of initial concentrations of Hg0 , O2 , SO2 and NO on the Hg0 removal The experiments of Hg0 removal in the presence of SO2 , NO and O2 were carried out to investigate the effects on the Hg0 removal of the coexistence gases that are found in coal-fired flue gas. As shown in Fig. 7, as an increase of Hg0 concentration from 10 to 50 ␮g/m3 , the efficiencies of Hg0 removal decrease from 88% to 85%, which indicates that the effect of Hg0 concentration on Hg0 removal is insignificant at the higher Hg0 concentration. The results show that the proposed method is preferably adapted to different types of coal for removing Hg0 . Fig. 7 also shows that the Hg0 removal efficiencies are 87.1% and 88.4% with respect to the O2 concentrations of 2.0% and 8.0%. This indicated that the O2 exhibited a weak effect on Hg0 oxidation, which was consistent with the results reported by Niksa et al. [36]. Fig. 7 also describes the effect of SO2 on the Hg0 removal, in which it is found that the Hg0 removal efficiencies are 92.0%, 82.1% and 64.3% when the SO2 concentrations are 1100, 2200 and 3500 mg/m3 , respectively, revealing that the increase of the Hg0 oxidation occurs at lower SO2 concentration levels. It has been reported that SO3 resulting from the oxidation of SO2 was effective in Hg0 removal (Eq. (9)), and that H2 SO4 has a potential to oxidize Hg0 into mercuric sulfate (HgSO4 ) [37,38], which is the most stable form of mercury at temperatures below 320 ◦ C, on the basis of the Frandsen model. Hence, a small amount of SO2 was beneficial to the Hg0 removal process. Whereas inhibition was prominent when higher SO2 concentrations existed in simulated flue gas, this might be due to the competitive reaction between the Hg0 and the SO2 , with limited gas phase oxidants. An experiment to verify this phenomenon does confirm that the removal efficiencies of Hg0 increase sharply from 64.0% to 94.0% as the adding rates of the LCA increasing from 150 to 500 ␮l/min, at SO2 concentration of 3500 mg/m3 . Hg0 + SO3 + O2 → HgSO4

(9)

The effect of NO on the Hg0 removal is shown in Fig. 7. Hg0 removal efficiencies are 89%, 82% and 70% with respect to NO concentrations at 350, 550 and 700 mg/m3 , respectively. Obviously, the effect of NO on the Hg0 removal was similar to that of SO2 . Niksa

et al. [36] pointed out that the impact of NO on homogeneous Hg0 oxidation was surprisingly strong, but interestingly, the NO was found to either promote or inhibit homogeneous Hg0 oxidation, depending on its concentration level. It was also reported that Hg0 could be catalytically oxidized by NO2 on the surface of the catalyst [39], thus the authors of this paper speculated that NO2 produced by the oxidation of NO may catalyze Hg0 oxidation on the surface of the Ca (OH)2 . Additionally, NO2 , as a gas oxidant, can directly oxidize Hg0 (Eqs. (10) and (11)) [40]. The competitive reaction between the Hg0 and the NO at higher NO concentration levels is also proved experimentally: when the adding rates of the LCA increase from 150 to 500 ␮l/min, the Hg0 removal efficiencies sharply increase from 71% to 93%, at NO concentration of 700 mg/m3 . Hg0 + NO2 → HgO + NO

(10)

Hg0 + O2 + 2NO2 → Hg(NO3 )2

(11)

It has been known that the transient response experiment can reflect the importance of a reactant, so two transient response experiments were carried out by using LCA as the oxidant, also by using H2 O as a blank oxidant under the same conditions to observe the roles of NO2 and SO3 /H2 SO4 in the removal of Hg0 . It can been seen from Fig. 8 that in the presence of NO2 of 300 mg/m3 , Hg0 removal efficiency evidently increases by approximately 30% in the reaction time range of 60–80 min when H2 O is used, however, Hg0 of approximately 93% is removed when LCA is used as the oxidant in the reaction time range of 90–125 min. Thereafter, when NO2 is removed from the reaction system, a slight decrease of Hg0 removal is observed. The experimental results show that NO2 can promote Hg0 removal, and the combination of NO2 and LCA is more effective. The reasons had been explained by the previous works [39,40]. As shown in Fig. 9, in the presence of SO2 of 1000 mg/m3 and with H2 O injection as a blank oxidant, Hg0 removal efficiency is basically 0% in the reaction time range of 45–80 min. This indirectly demonstrates that Hg0 removal cannot be promoted when SO3 and H2 SO4 do not exist in the reaction system in absence of LCA. However, when LCA is injected in the reaction time range of 90 to 150 min, Hg0 removal efficiency is reached as high as 85%; and then as SO2 is removed from the reaction system in the reaction time range of 120–150 min, a slight decrease of Hg0 removal is observed. These phenomena demonstrated that Hg0 oxidation could be promoted by SO3 as well as H2 SO4 resulting from SO2 oxidation, which is consistent with the previous observations [37,38].

124

Y. Zhao et al. / Journal of Hazardous Materials 280 (2014) 118–126

100 1000

600

60

LCA added 40

400

Remove SO2 Fluorescence value Removal efficiency

200

0

30

0

20

60

90

120

0

H2O added

Hg removal efficiency (%)

Fluorescence value

80

SO2 input

800

0 150

Reaction time (min) Fig. 9. Transient response of SO2 during LCA oxidation Hg0 . Reaction temperature, 130 ◦ C; LCA solution pH, 1.1; adding rate of LCA, 150 ␮l/min; flue gas velocity, 1 l/min.

3.6. The simultaneous removal of SO2 , NO and Hg0 The optimal preparation conditions for the LCA, and the best reaction conditions, were determined according to the experimental data, in which the concentration ratios of H2 O2 , FeSO4 , PAA and NaCl were 7.8, 4.0, 4.4 and 40.0 mmol/l, with an LCA solution pH of 1.1, a reaction temperature of 130 ◦ C, and an LCA adding rate of 150 ␮l/min. Under optimal conditions, Hg0 removal efficiency of 91.0% was obtained when the Hg0 concentration was 20 ␮g/m3 . Additionally, experiments of the simultaneous removal of SO2 , NO and Hg0 were carried out with an adding rate for the LCA of 500 ␮l/min, and the SO2 concentration of 2200 mg/m3 , the NO concentration of 550 mg/m3 , and the Hg0 concentration of 20 ␮g/m3 . The experimental data shown in Table 1 indicated that this integrative method exhibited a stable and good performance for the simultaneous removal of multi-pollutants from the flue gas. In terms of costs, it was estimated that the molar ratio of effective oxidants and multi-pollutants in the flue gas was approximately 5, and through estimation, the cost for the simultaneous removal of SO2 , NO and Hg0 was 11,000–12,000 ¥/t. Data regarding the air pollution control equipment in coal-fired power plants revealed the costs of WFGD-SCR and ACI were about 10,000–13,000 ¥/t and or $US 3810–166,000 per pound, respectively. Obviously, the cost for the simultaneous removal of SO2 , NO and Hg0 is lower than the total cost for WFGD-SCR and ACI. The utilization of simultaneous removal technology can simplify Air Pollution Control Device Systems (APCDs), decrease the total area utilized, and reduce capital construction costs. This proposed methodology is thus quite desirable for future industrial engineering applications. 3.7. Reaction mechanism for the simultaneous removal of SO2 , NO and Hg0 Characterization of the removal products by XRD and AFS were conducted to reveal the reaction mechanisms for the simultaneous Table 1 Parallel tests of simultaneous removal SO2 , NO and Hg0 . 1 SO2 NO Hg0

100 78.8 78.5

2 100 76.8 75.3

3 100 80.5 81.2

4 100 81.7 83.2

5 100 80.3 80.1

6 100 78.9 83.9

Fig. 10. XRD patterns of the spent absorbent (II) and Ca(OH)2 (I).

removal of SO2 , NO and Hg0 . The powder X-ray diffraction patterns of the fresh and the spent Ca(OH)2 are shown in Fig. 10. By contrast, it could be found that CaSO4 (JCPDS file no. 45-0157), CaSO3 (JCPDS file no. 44-0516), Ca(NO3 )2 (JCPDS file no. 07-0204) and Ca(NO2 )2 (JCPDS file no. 28-0232) appear on the surface of samples, and the characteristic peaks of CaSO4 and Ca(NO3 )2 are stronger than those of CaSO3 and Ca(NO2 )2 . This demonstrated that the major removal products of desulfurization and denitrification were CaSO4 and Ca(NO3 )2 , and the minor ones were CaSO3 and Ca(NO2 )2 . Five samples taken from spent KCl solution after 30 min of reaction time were analyzed by AFS. As shown in Table 2, there is no Hg2+ in the scan blank, while Hg2+ is found in samples with an average concentration of 75.0 ng/l. Through estimation, the corresponding mass was found to be 22.5 ng, which meant that the Hg0 could be oxidized as Hg2+ by the vaporized LCA. In fact, approximately 600 ng of Hg0 entered the reactor from the mercury osmotic tube during the reaction period of 30 min, and it was determined by calculations that 546 ng of Hg0 was oxidized into Hg2+ , based on the Hg0 removal efficiency rate of 91%, as presented in Section 3.6. From the perspective of mass balance, the Hg0 has three possible pathways to take: (1) 54 ng of Hg0 escaped from the reactor and was absorbed by a solution of 10% (v/v) H2 SO4 and 4% (w/w) KMnO4 . (2) 523.5 ng of Hg2+ accounting for 95.9% (w/w) of the total oxidized mercury, was absorbed by Ca(OH)2 , indicating that Ca(OH)2 was an excellent absorbent for Hg2+ removal. (3) The other 22.5 ng of Hg2+ , accounting for 4.1% (w/w) of the total oxidized mercury, was absorbed by the KCl solution. Based on the characteristics of the removal products and the relevant references, the reaction mechanism for the simultaneous removal of SO2 , NO and Hg0 by LCA can be concluded to be as follows: the oxidants such as H2 O2 , PAA and •OH (Eqs. (12) and Table 2 Concentration of Hg2+ in the spent KCl solution (ng/l).*

Average 100 79.5 80.4

Standard error 0 2.44 8.41

Numbers

1

2

3

4

5

Average

Scan blank Samples

0 75.1

0 77.6

0 73.4

0 74.3

0 74.7

0 75.0

*

The concentration of KCl solution is 1.0 mol/l.

Y. Zhao et al. / Journal of Hazardous Materials 280 (2014) 118–126

125

(13)) might play a leading role in the oxidation of NO and SO2 (Eqs. (17)–(24)) because the concentration of those primary oxidants was found to be larger than those of the secondary oxidants, such as ClOH•− , Cl•, and Cl2 (Eqs. (14)–(16)). From an electrochemical perspective, the standard electrode potentials of H2 O2 (1.770 V), •OH (2.800 V) [25] and PAA (1.960 V) [41] are far higher than those of SO4 2− /H2 SO3 (0.172 V), SO4 2− /SO2 (0.158 V), NO2 /NO (1.049 V), NO3 - /NO (0.957 V), NO2 − /NO (−0.460 V) and NO3 − /NO2 − (0.835 V), which demonstrated the feasibility of an oxidation reaction of SO2 and NO. For the Hg0 removal, it seemed that those chlorine radicals generated from the hydroxyl radicals and chlorine ions (Eqs. (14)–(16)) were dominant in the Hg0 oxidation (Eqs. (25)–(28)), as discussed in Section 3.1, and which was also determined by previous researchers [26–28]. The reaction intermediates, as well as almost all the oxidation products such as HNO3 , HNO2 , H2 SO4 , H2 SO3 , NO2 , SO3 , Hg2 (OH)2 , HgCl and HgCl2 , were absorbed by the Ca(OH)2 (Eqs. (29)–(34)) at the reactor and then partially absorbed by the KCl solution, especially for Hg2+ . The possible reaction paths of simultaneous removal are described as followed. The active species generation (Eqs. (12)–(16)) [23,30–32]:

steel is considered as an effective one and has widely been used in the stacks of Chinese coal-fired power plants. Thus we advise that it may be employed as an inside lining of the preoxidation device to inhibit the halogenated compounds corrosion. Also, the residual halogenated compounds in gas phase can then be absorbed by the followed turbulent industrial Ca(OH)2 in the CFB-FGD reactor. Moreover, in consideration of the fact that the halogens did present in actual flue gas, we conceived an economic method that the Fenton/PAA can cooperate with halogens in flue gas to remove Hg0 , thus halogens used in this method will be saved. And we will do a depth research on this conception in the future. For the removal products, they can then be utilized in such products as wall panels, as the main material in steam bricks, in road pavements, in the reconstruction of highway embankments, and as a concrete additive instead of part of cement landfills [43]. Taking into consideration the extensive installations of WFGDs in Chinese coal-fired power plants, future research efforts will need to be focused on developing an integrative process through vaporized LCA pre-oxidation and the CaCO3 slurry absorption.

Fe2+ + H2 O2 → HO• + Fe3+ + OH−

(12)

4. Conclusions

Fe2+ + CH3 COOOH → HO • +Fe3+ + OH−

(13)

(1) A liquid-phase complex absorbent (LCA) made up of Fenton, PAA and NaCl was prepared, and the experiments of the simultaneous removal of SO2 , NO and Hg0 were carried out to develop a novel integrative method for simultaneously removing multipollutants from coal-fired flue gas. Under optimal experimental conditions, the highest Hg0 removal efficiency rate of 91% was obtained, while the simultaneous removal efficiencies of SO2 , NO and Hg0 were 100%, 79.5% and 80.4%, respectively. (2) According to the characteristics of the removal products determined by XRD and AFS, and the related references, the main products of the Hg0 removal were verified as Hg2 (OH)2 , HgCl and HgCl2 , and those of desulfurization and denitrification were identified as CaSO4 , Ca(NO3 )2 , CaSO3 and Ca(NO2 )2 . Combined with the analysis of the electrode potentials, the reaction mechanism for simultaneous removal of SO2 , NO and Hg0 were proposed.

HO•

−

+ Cl →

ClOH• −

(14)

ClOH• − + H+ → Cl• + H2 O

(15)

Cl•

(16)

+ Cl•

→ Cl2

The oxidation reaction (Eqs. (9)–(11), (17)–(28)) [23,30–32]: + 2NO + 3H2 O2 → 2NO− 3 + 2H + 2H2 O

(17)

NO + H2 O2 → NO2 + H2 O

(18)

+

SO2 + H2 O2 → 2H

+ SO2− 4

−

NO + CH3 COOO → NO2 + CH3 COO −

(19) − −

SO2 + CH3 COOO → SO3 + CH3 COO HO• HO•

+ NO → H

+

+ NO2 → H

+ NO− 2

+

+ NO− 3

+

HO• + SO2 → 2H

+ SO2− 4

(20) (21) (22) (23)

Acknowledgments (24)

2HO• + Hg0 → Hg(OH)2

(25)

2ClOH• − + Hg0 → HgCl2 + 2HO•

(26)

2Cl•

0

(27)

0

(28)

+ Hg + M → HgCl2 + M

2Cl2 + Hg → HgCl2 + 2Cl The absorption reactions (Eqs. (29)–(34)): Ca(OH)2 + Hg(OH)2 → HO–HgOH–OH· · ·Ca(OH)2

(29)

Ca(OH)2 + HgCl2 → Cl–HgOH–Cl· · ·Ca(OH)2

(30)

Ca(OH)2 + S(IV) → CaSO3 + H2 O

(31)

Ca(OH)2 + S(VI) → CaSO4 + H2 O

(32)

Ca(OH)2 + 2N(IV)/(III) → Ca(NO2 )2 + H2 O

(33)

Ca(OH)2 + 2N(V) → Ca(NO3 )2 + H2 O

(34)

Based on the principle of mass balance, the concentrations of the chlorinated oxidants in gas phase have been calculated. It is approximately 130 ppm that is basically same with the concentration of chlorine in the actual flue gas [42]. Hence, the potential corrosive hazards of halogenated compounds may exist in industrial application. For this, the anti-corrosion is necessary in the preoxidation device. Among of anti-corrosion materials, titanium

The authors appreciate the financial support by a grant from the National High Technology Research and Development Program of China (863 Program, No. 2013AA065403), Hebei Provincial Natural Science Foundation, P.R. China (B2011502027), Program for Changjiang Scholars and Innovative Research Team in University (IRT1127) and Zhejiang Provincial Engineering Research Center of Industrial Boiler & Furnace Flue Gas Pollution Control, Hangzhou, P.R. China (311202). References [1] Y. Wang, Y.X. Qi, Y.F. Li, J.J. Wu, X.J. Ma, C. Yu, L. Ji, Preparation and characterization of a novel nano-absorbent based on multi-cyanoguanidine modified magnetic chitosan and its highly effective recovery for Hg(II) in aqueous phase, J. Hazard. Mater. 260 (2013) 9–15. [2] US-EPA, T6560-50.-P, Standards of Performance for New and Existing Stationary Source: Electric Utility Steam Generating Units, 2005, 40 CFR Parts 60, 63, 72 and 75. [3] H.Q. Yang, Z.H. Xu, M.H. Fan, A.E. Bland, R.R. Judkins, Adsorbents for capturing Hg0 in coal-fired boiler flue gas, J. Hazard. Mater. 146 (2007) 1–11. [4] J. Wilcox, E. Rupp, S.C. Ying, D.H. Lim, A.S. Negreira, A. Kirchofer, F. Feng, K. Lee, Mercury adsorption and oxidation in coal combustion and gasification processes, Int. J. Coal Geol. 90 (2012) 4–20. [5] Y. Cao, Q.H. Wang, J. Li, J. Cheng, C.C. Chan, M. Cohron, W.P. Pan, Enhancement of Hg0 capture by the simultaneous addition of hydrogen bromide (HBr) and fly ashes in a slipstream facility, Environ. Sci. Technol. 43 (2009) 2812–2817. [6] E. Sasmaz, A. Kirchofer, A.D. Jew, A. Saha, D. Abram, T.F. Jaramillo, J. Wilcox, Mercury chemistry on brominated activated carbon, Fuel 99 (2012) 188–196.

126

Y. Zhao et al. / Journal of Hazardous Materials 280 (2014) 118–126

[7] J. Wilcox, E. Sasmaz, A. Kirchofer, Heterogeneous mercury reaction chemistry on activated carbon, J. Air Waste Manage. Assoc. 61 (2011) 418–426. [8] A.P. Jones, J.W. Hoffmann, D.N. Smith, T.J. Feeley, J.T. Murphy, DOE/NETL’s phase II Hg0 control technology field testing program: preliminary economic analysis of activated carbon injection, Environ. Sci. Technol. 41 (2007) 1365–1371. [9] S. Fabrizio, Simulation of Hg0 capture by activated carbon injection in incinerator flue gas. 1. In-duct removal, Environ. Sci. Technol. 35 (2001) 4367–4372. [10] D.H. Lim, J. Wilcox, Heterogeneous mercury oxidation on Au(111) from first principles, Environ. Sci. Technol. 47 (2013) 8515–8522. [11] D.H. Lim, S. Aboud, J. Wilcox, Investigation of adsorption behavior of mercury on Au(111) from first principles, Environ. Sci. Technol. 46 (2013) 7260–7266. [12] E. Sasmaz, S. Aboud, J. Wilcox, Hg binding on Pd binary alloys and overlays, J. Phys. Chem. C 113 (2009) 7813–7820. [13] A.S. Negreira, J. Wilcox, DFT study of Hg oxidation across vanadia-titania SCR catalyst under flue gas conditions, J. Phys. Chem. C 117 (2013) 1761–1772. [14] Z.B. Wu, B.Q. Jiang, Y. Liu, Effect of transition metals addition on the catalyst of manganese/titania for low-temperature selective catalytic reduction of nitric oxide with ammonia, Appl. Catal. B 79 (2008) 347–355. [15] X.P. Fan, C.T. Li, G.M. Zeng, X. Zhang, S.S. Tao, P. Lu, Y. Tan, D.Q. Luo, Hg0 removal from simulated flue gas over CeO2 /HZSM-5, Energy Fuel 26 (2012) 2082–2089. [16] N.D. Hutson, R. Krzyzynska, R.K. Srivastava, Simultaneous removal of SO2 , NOx , and Hg from coal flue gas using a NaClO2 -enhanced wet scrubber, Ind. Eng. Chem. Res. 47 (2008) 5825–5831. [17] P. Fang, C.P. Cen, X.M. Wang, Z.J. Tang, Z.X. Tang, D.S. Chen, Simultaneous removal of SO2 , NO and Hg0 by wet scrubbing using urea + KMnO4 solution, Fuel Process. Technol. 106 (2013) 645–653. [18] S.H. Liu, N.Q. Yan, Z.R. Liu, Z. Qu, H.P. Wang, S.H. Chang, C. Miller, Using bromine gas to enhance Hg0 removal from flue gas of coal-fired power plants, Environ. Sci. Technol. 41 (2007) 1405–1412. [19] Z. Qu, N.Q. Yan, P. Liu, Y.F. Guo, J.P. Jia, Oxidation and stabilization of elemental Hg0 from coal-fired flue gas by sulfur monobromide, Environ. Sci. Technol. 44 (2010) 3889–3894. [20] L. Jia, R. Dureau, V. Ko, E.J. Anthony, Oxidation of Hg0 under ultraviolet (UV) irradiation, Energy Fuel 24 (2010) 4351–4356. [21] J.J. Tokos, B. Hall, J.A. Calhoun, E.M. Prestbo, Homogeneous gas-phase reaction of Hg0 with H2 O2 , O3 , CH3 I and (CH3 )2 S: implications for atmospheric Hg cycling, Atmos. Environ. 32 (1998) 823–827. [22] B. Pal, P.A. Ariya, Gas-phase HO-initiated reactions of elemental Hg0 : kinetics, product studies, and atmospheric implications, Environ. Sci. Technol. 38 (2004) 5555–5566. [23] E.S. Hwang, J.N. Cash, M.J. Zabik, Ozone and hydrogen peroxyacetic acid treatment to reduce or remove EBDCs and ETU residues in a solution, J. Agric. Food Chem. 49 (2001) 5689–5694. [24] Y.J. Wang, Y.F. Duan, L.G. Yang, C.S. Zhao, X.L. Shen, M.Y. Zhang, Y.Q. Zhuo, C.H. Chen, Experimental study on mercury transformation and removal in coal-fired boiler flue gases, Fuel. Process. Technol. 90 (2009) 643–651.

[25] N.M. Modirshahla, A. Behnajady, Photooxidative degradation of malachite Green (MG) by UV/H2 O2 : influence of operational parameters and kinetic modeling, Dyes Pigm. 70 (2007) 54–59. [26] Y.G. Adewuyi, S.O. Owusu, Ultrasound-induced aqueous removal of nitric oxide from flue gases: effects of sulfur dioxide, chloride, and chemical oxidant, J. Phys. Chem. A 110 (2006) 11098–11107. [27] X.Y. Yu, J.R. Barker, Hydrogen peroxide photolysis in acidic aqueous solutions containing chloride ions. I. Chemical mechanism, J. Phys. Chem. A 107 (2003) 1313–1324. [28] M.H. Xua, Y.Q., C.G. Zheng, L.C. Li, J. Liu, Modeling of homogeneous mercury speciation using detailed chemical kinetics, Combust. Flame 132 (2003) 208–218. [29] E. Pitoniak, C.Y. Wu, D.W. Mazyck, K.W. Powers, W. Sigmund, Adsorption enhancement mechanisms of silica-titania nanocomposites for elemental Hg0 vapor removal, Environ. Sci. Technol. 39 (2005) 1269–1274. [30] J. Wilcox, T. Okano, Ab initio-based Hg0 oxidation kinetics via bromine at post combustion flue gas conditions, Energy Fuel 25 (2011) 1348–1356. [31] Q. Hua, C. Zhang, Z. Wang, Photodegradation of methyl tertbutyl ether (MTBE) by UV/H2 O2 and UV/TiO2 , J. Hazard. Mater. 154 (2008) 795–803. [32] F. Yuan, C. Hu, X.X. Hu, J.H. Qu, Degradation of selected pharmaceuticals in aqueous solution with UV and UV/H2 O2 , Water Res. 43 (2009) 1766–1774. [33] W.M. Wang, J. Song, X. Han, Schwertmannite as a new Fenton-like catalyst in the oxidation of phenol by H2 O2 , J. Hazard. Mater. 262 (2013) 412–419. [34] J.D. Laat, T.G. Le, Kinetics and modeling of the Fe(III)/H2 O2 system in the presence of sulfate in acidic aqueous solutions, Environ. Sci. Technol. 39 (2005) 1811–1818. [35] Y. Zhao, Y.H. Han, T.Z. Ma, T.X. Guo, Simultaneous desulfurization and denitrification from flue gas by ferrate(VI), Environ. Sci. Technol. 45 (2011) 4060–4065. [36] S. Niksa, J.J. Helble, N. Fujiwara, Kinetic modeling of homogeneous mercury oxidation: the importance of NO and H2 O in predicting oxidation in coal-derived systems, Environ. Sci. Technol. 35 (2001) 3701–3706. [37] S. Eswaran, H.G. Stenger, Understanding Hg0 conversion in selective catalytic reduction (SCR) catalysts, Energy Fuel 19 (2005) 2328–2334. [38] Y.S. Gao, Z. Zhang, J.W. Wu, L.H. Duan, A. Umar, L.Y. Sun, Z.H. Guo, Q. Wang, A critical review on the heterogeneous catalytic oxidation of elemental mercury in flue gases, Environ. Sci. Technol. 47 (2013) 10813–10823. [39] G.E. Dunham, E.S. Olson, S.J. Miller, Impact of flue gas constituents on carbon sorbents, in: Proceedings of the Air Quality II - International Conference on Mercury, Trace. Elem. Particul. Mat., 2000. [40] X.Y. Wen, C.T. Li, X.P. Fan, H.L. Gao, W. Zhang, L. Chen, G.M. Zeng, Y.P. Zhao, Experimental study of gaseous elemental Hg0 removal with CeO2 /␥-Al2 O3 , Energy Fuel 25 (2011) 2939–2944. [41] M. Kitis, Disinfection of wastewater with peracetic acid: a review, Environ. Int. 30 (2004) 47–55. [42] K.L. Liu, Y. Gao, R. John, et al., An investigation of mercury emission from FBC systems fired with high-chlorine coals, Energy Fuel (15) (2001) 1173–1180. [43] Y. Zhao, Y.H. Han, C. Chen, Simultaneous removal of SO2 and NO from flue gas using multicomposite active absorbent, Ind. Eng. Chem. Res. 51 (2012) 480–486.