English

• Mercury is considered a major air pollutant because of its toxicity and bioaccumulation^{[1, 2]}. Currently, coal-fired power plant has been the largest single-known source of anthropogenic mercury emissions in the world. In China, coal constitutes about 70% of the total primary energy consumption^[3]. For controlling and reducing mercury emission, the Ministry of Environmental Protection of China issued a new pollution discharge standard in 2011 and specified the emission limits of mercury and its compounds from coal-fired flue gas^[4]. There are three forms of mercury in flue gas: elemental mercury vapor (Hg⁰), oxidized mercury (Hg²⁺) and particulate-bound mercury (Hg^p)^[5]. Among them, Hg⁰ is the most difficult to capture because of its high volatility and poor solubility in water. Thus, it is highly desirable to develop novel and efficient methods to remove Hg⁰ from flue gas in coal-fired power plant.

  During the past two decades, considerable efforts have been undertaken to remove Hg⁰. Some materials including activated carbon^[6] and metal oxide^[7] all have been widely explored. However, there were still some crucial drawbacks that restrained their practical applications, including the high cost and low adsorption and oxidation efficiency. Recently, photocatalytic oxidation (PCO) technique for Hg⁰ removal using TiO₂ as photocatalyst has been extensively investigated to convert Hg⁰ to Hg²⁺ in simulated flue gas^[8-11]. When TiO₂ was irradiated under ultraviolet (UV) light, the electrons would be photoexcited from the valence band (VB) to the conduction band (CB), generating electron-hole (e^--h⁺) pairs that can recombine or initiate redox reactions. The e^- could reduce O₂ adsorbed on the material surface to generate superoxide radicals (·O₂^-), while the h⁺ could be trapped by H₂O and OH^- to generate hydroxyl radicals (·OH). As reported, the photogenerated ·O₂^- and/or ·OH radicals with enough oxidizability displayed important roles in Hg⁰ oxidation process as follows^[8-12]:

  $ \cdot {\rm{OH + H}}{{\rm{g}}^{\rm{0}}} \to {\rm{HgOH}} $

  $ \cdot {\rm{OH + HgOH}} \to {\rm{HgO + }}{{\rm{H}}_{\rm{2}}}{\rm{O}} $

  $ \cdot {\rm{O}}_2^-{\rm{ + H}}{{\rm{g}}^{\rm{0}}}{\rm{ + 2}}{{\rm{H}}^{\rm{ + }}} \to {\rm{HgO + }}{{\rm{H}}_{\rm{2}}}{\rm{O}} $

  The photocatalytic oxidation of Hg⁰ using TiO₂ could be promising because of its low cost, high chemical stability and absence of secondary pollution. However, TiO₂ has a large band gap (3.2 eV) and can only be excited under UV light, which represents merely 5% of the sunlight. In addition, the high rate of recombination between photogenerated electrons and holes resulted in a limited photocatalytic activity and rate^[4]. To prevent the recombination between photogenerated electrons and holes, the modified TiO₂ materials such as CuO/TiO₂^[4], SiO₂-TiO₂^[13] and TiO₂-aluminum silicate fiber^[14] had been investigated and much higher Hg⁰ removal efficiencies were obtained under UV light than that of pure TiO₂. To extend the absorption spectra of photocatalysts in pollutant gas treatment and air purification, several newly visible-light-driven photocatalysts such as BiOIO₃^[15], BiOI^[16] and BiVO₄^[17] also have been synthesized and satisfactory results were obtained in removing Hg⁰ and NO under visible light. Our previous studies also observed desirable Hg⁰ removal efficiencies (more than 90%) using BiOBr, BiOI and Ag-based photocatalysts in a wet PCO reactor under fluorescent light (FSL) irradiation^{[18, 19]}. Furthermore, it was found that this wet PCO system can simultaneously remove Hg⁰ and SO₂ by adding Ca(OH)₂ powder into the reaction solution, which had not been reported by other researchers.

  Although visible-light-responsive photocatalysis is efficient under FSL light or sunlight, the reusability and separation of photocatalysts from the solution are troublesome. To surmount the disadvantages, a novel magnetic AgI-BiOI/CoFe₂O₄ composite was synthesized to remove Hg⁰ from flue gas in our previous study^[20]. The experimental results indicated that the as-prepared ternary AgI-BiOI/CoFe₂O₄ material not only exhibited superior Hg⁰ oxidation activities but also possessed good magnetic separation performances. Although considerable works of Hg⁰ removal over AgI-BiOI/CoFe₂O₄ composites had been carried out in the previous work, some issues such as photocatalyst thermal stability, the effects of solution pH and inorganic anions had not yet been investigated and fully understood. Thus, the aim of the present work was to systematically understand the influences of operational factors including photocatalyst thermal stability, temperature of reaction solution in reactor, reaction solution pH, catalyst dosage, inorganic anions, and concentrations of O₂, SO₂ and NO on Hg⁰ removal. The reaction products in supernatant fluid, the reusability and magnetism of photocatalyst after reaction were also explored. The research results could deepen our understanding of Hg⁰ removal over magnetic AgI-BiOI/CoFe₂O₄ hybrids using wet photocatalytic oxidation technology in coal-fired power plants.

  1.   Experimental

  1.1   Preparation of AgI-BiOI/CoFe₂O₄ composites

  AgI-BiOI/CoFe₂O₄ hybrid composites were prepared via a solvothermal and subsequent coprecipitation method. The detailed procedures were provided in the previous work^[20]. Based on the previous results, an optimal AgI_0.3BiOI/CoFe₂O₄ material where the mass ratios of AgI and CoFe₂O₄ to AgI-BiOI/CoFe₂O₄ were about 0.3 and 0.2, respectively was chosen as default photocatalyst in the present study, and for simplicity it was designated as AgI-BiOI/CoFe₂O₄.

  1.2   Photocatalytic performance measurement

  The performances of Hg⁰ removal over AgI-BiOI/CoFe₂O₄ hybrid were carried out on a laboratory-scale wet bubbling photocatalytic reactor. The schematic diagram of the experimental setup is shown in Figure 1.

  图 1

  

  图 1  Schematic diagram of the experimental setup

  Figure 1.  Schematic diagram of the experimental setup

  下载: 全尺寸图片 幻灯片

  A detailed description of the experimental system was given elsewhere^[19]. During the experiment, baseline flue gas (6% of O₂, 12% of CO₂, 55 μ g/m³ of Hg⁰ vapor and balance N₂), SO₂ (when used) and NO (when used) were employed as the simulated flue gas. The total flow rate of flue gas was about 1.50 L/min. The temperature of reaction solution in photocatalytic reactor (t) was set as 35 ℃. A common fluorescent lamp (FSL) with a power of 11 W was used as the light source. The concentrations of Hg⁰ were measured by a VM-3000 mercury analyzer and the Hg⁰ removal efficiency η (%) was calculated according to the following equation:

  $ \eta {\rm{ = (1-}}{\mathit{C}_{\rm{0}}}\mathit{/}{\mathit{C}_\mathit{i}}{\rm{)}} \times {\rm{100\% }} $

  where C₀ and C_i refer to the outlet and inlet concentrations of Hg⁰ (μ g/m³).

  1.3   Characterization of photocatalysts

  The thermal gravimetric and differential scanning calorimetry (TG-DSC) was performed on a SDT Q600 thermogravimetric analyzer (TA Instrument, USA). X-ray diffraction (XRD) tests were performed on a D8 advance diffractometer (Bruker, Germany). N₂ adsorption-desorption measurements were carried out on an Autosorb iQ surface area and porosity analyzer (Quantachrome, USA). The magnetism of the composite was analyzed on a MPMS XL-7 magnetometer (Quantum Design, USA). The ion products in spent solution were measured via a 792 Basic ion chromatograph (Metrohm AG, Switzerland).

  2.   Results and discussion

  2.1   Effect of thermal stability of photocatalyst

  The thermal stability was first detected using thermogravimetric analyzer and the results are given in Figure 2. As shown in Figure 2(a), no significant weight loss of AgI-BiOI/CoFe₂O₄ is recorded when the temperature is below 350 ℃. However, two peaks appear under 350 ℃ in the DSC curve (Figure 2(b)). The endothermic peak at 153 ℃ and the exothermal peak at 300 ℃ are assigned to the desorption of water and the decomposition of organic substances adsorbed in the sample, respectively. With the temperature increasing from 350 to 600 ℃, a significant weight loss of AgI-BiOI/CoFe₂O₄ and two exothermal peaks located at 358 and 506 ℃ are observed, indicating the decomposition or transformation of the samples.

  图 2

  

  图 2  TG (a) and DSC (b) curve for AgI-BiOI/CoFe₂O₄

  Figure 2.  TG (a) and DSC (b) curve for AgI-BiOI/CoFe₂O₄

  下载: 全尺寸图片 幻灯片

  To investigate the effect of thermal stability of photocatalyst on Hg⁰ removal performance, the photocatalyst was calcinated at 200, 400 and 600 ℃ in air for 5 h and then characterized by N₂ adsorption-desorption and X-ray diffraction (XRD) techniques. Table 1 summarizes the physical features of calcinated samples. Obviously, the BET surface area and total pore volume of the composite first slightly increase and then greatly decrease with the increase of calcination temperature. The XRD patterns of the composites with different calcination temperature are shown in Figure 3, when the calcination temperature is below 200 ℃, the two samples display almost the same diffraction peaks, which can be indexed to the hexagonal phase of AgI (JCPDS 09-0374), tetragonal phase of BiOI (JCPDS 73-2062) and signal phase of CoFe₂O₄ (JCPDS 22-1086)^[21]. When the calcination temperature further increases to 400 and 600 ℃, the diffraction peaks of BiOI disappear and some new diffraction peaks belonged to orthorhombic Bi₅O₇I (JCPDS 40-0548) are detected and strengthened^[22], which is in line with the TG-DSC analysis. The above phenomena clearly suggested that much higher calcination temperature would result in inferior pore structure of AgI-BiOI/CoFe₂O₄ and some transformations of substances in chemical structure.

  表 1

  

  表 1  Physical features of AgI-BiOI/CoFe₂O₄ hybrids calcinated at different temperatures

  Table 1.  Physical features of AgI-BiOI/CoFe₂O₄ hybrids calcinated at different temperatures

  下载: 导出CSV

  Sample	BET surface area A/(m2·g-1)	Total pore volume v/(cm3·g-1)
  Without calcination	20.7	0.066
  200 ℃ calcination	33.9	0.114
  400 ℃ calcination	15.1	0.056
  600 ℃ calcination	1.5	0.002

  图 3

  

  图 3  XRD patterns of the photocatalysts

  Figure 3.  XRD patterns of the photocatalysts

  下载: 全尺寸图片 幻灯片

  The activities of Hg⁰ removal over AgI-BiOI/CoFe₂O₄ with various calcination temperatures under FSL irradiation are shown in Figure 4.

  图 4

  

  图 4  Effect of calcination temperature on Hg⁰ removal efficiency

  Figure 4.  Effect of calcination temperature on Hg⁰ removal efficiency

  下载: 全尺寸图片 幻灯片

  It is obvious that with increasing calcination temperature from 200 to 600 ℃, the Hg⁰ removal efficiency drops from 92% to 74%. The loss in Hg⁰ removal efficiency can be ascribed to two reasons: one is as a result of the inferior pore feature of AgI-BiOI/CoFe₂O₄ calcinated at higher temperature (Table 1) and the other can be due to the transformation of active materials in AgI-BiOI/CoFe₂O₄ composites. It is reported that the band gap energy of Bi₅O₇I (2.94 eV) was much larger than that of BiOI (1.73 eV), its visible light absorption ability was lower than that of BiOI^[23], thereby resulting in lower generation of photogenerated electron-hole pairs and lower Hg⁰ removal efficiency.

  2.2   Effect of reaction solution temperature

  The effect of reaction solution temperature in reactor on Hg⁰ removal efficiency is shown in Figure 5. When reaction solution temperature increases from 15 to 75 ℃, the Hg⁰ removal efficiency increases first and then decreases. At lower reaction solution temperature, the increasement of reaction solution temperature can enhance the reaction rate by lowering the energy barrier, which would accelerate the generation of reactive species on the surface of photocatalyst. However, much higher reaction solution temperature would decrease the solubility of O₂ and Hg⁰ in solution and weaken the adsorption of Hg⁰ and O₂ on the photocatalyst surface, leading to a decrease in Hg⁰ removal under FSL irradiation^[14].

  图 5

  

  图 5  Effect of reaction solution temperature on Hg⁰ removal efficiency

  Figure 5.  Effect of reaction solution temperature on Hg⁰ removal efficiency

  下载: 全尺寸图片 幻灯片

  2.3   Effect of solution pH

  Figure 6 shows the effect of solution pH on Hg⁰ removal efficiency. It is observed that increasing solution pH from 1 to 7 has little impact on Hg⁰ removal, while when continuing to increase pH values from 7 to 10 and 13, Hg⁰ removal efficiencies slightly decreases from 97% to 93% and 80%, respectively, indicating a negative influence on Hg⁰ removal in alkaline condition. Our previous work^[20] indicated that superoxides radical (·O₂^-) and hydroxyl radicals (·OH) appeared in the solution under visible light irradiation and photogenerated holes (h⁺) and ·O₂^- were the main reactive species rather than ·OH in wet PCO of Hg⁰ system, which was quite different from the gas-phase PCO of Hg⁰ system in the literatures^{[8, 13]}. In the process of Hg⁰ oxidation, the following reactions can be happened:

  $ {{\rm{h}}^{\rm{ + }}}{\rm{ + }}{{\rm{H}}_{\rm{2}}}{\rm{O}} \to {{\rm{H}}^{\rm{ + }}}{\rm{ + }} \cdot {\rm{OH}} $

  $ {\rm{2}}{{\rm{h}}^{\rm{ + }}}{\rm{ + H}}{{\rm{g}}^{\rm{0}}} \to {\rm{H}}{{\rm{g}}^{{\rm{2 + }}}} $

  $ {{\rm{h}}^{\rm{ + }}}{\rm{ + O}}{{\rm{H}}^{\rm{-}}} \to \cdot {\rm{OH}} $

  图 6

  

  图 6  Effect of solution pH on Hg⁰ removal efficiency

  Figure 6.  Effect of solution pH on Hg⁰ removal efficiency

  下载: 全尺寸图片 幻灯片

  At lower pH value, the content of H⁺ in the reaction solution is high and the redox reaction of Eq. (5) would be inhibited, and the photogenerated holes can be largely retained to oxidize Hg⁰ (Eq. (6)). However, when the pH value is high (pH > 8), the amounts of h⁺ will decrease due to the presence of OH^- according to the reaction of Eq. (7), resulting in the reduction of Hg⁰ removal efficiency.

  2.4   Effects of inorganic anions

  Considering that some common anions such as NO₃^-, SO₄^2-, CO₃^2- and Cl^- would be dissolved in realistic reaction solution, which would influence on Hg⁰ removal during the wet PCO process, NaNO₃, Na₂SO₄, Na₂CO₃ and NaCl salts were employed to investigate the effects of NO₃^-, SO₄^2-, CO₃^2- and Cl^- on Hg⁰ removal. As shown in Figure 7, when 0.1 mol/L of NO₃^- is added into the reaction solution, the Hg⁰ removal efficiency decreases slightly. Previous work^[24] had shown that NO₃^- can scavenge ·OH radicals and generate some amounts of OH^- as shown in Eq. (8), causing a slight decrease in Hg⁰ removal.

  $ {\rm{NO}}_3^-{\rm{ + }} \cdot {\rm{OH}} \to \cdot {\rm{N}}{{\rm{O}}_{\rm{3}}}{\rm{ + O}}{{\rm{H}}^{\rm{-}}} $

  图 7

  

  图 7  Effect of anions in solution on Hg⁰ removal efficiency

  Figure 7.  Effect of anions in solution on Hg⁰ removal efficiency

  下载: 全尺寸图片 幻灯片

  The presence of SO₄^2- in solution also exerted a slightly inhibitory effect on Hg⁰ removal due to the consumptions of h⁺ and ·OH by SO₄^2- as follows^[24]:

  $ {\rm{SO}}_4^{2-}{\rm{ + }}{{\rm{h}}^{\rm{ + }}} \to \cdot {\rm{SO}}_4^- $

  $ {\rm{SO}}_4^{2-}{\rm{ + }} \cdot {\rm{OH}} \leftrightarrow \cdot {\rm{SO}}_4^-{\rm{ + O}}{{\rm{H}}^{\rm{-}}} $

  The reduction of Hg⁰ removal efficiency in the presence of CO₃^2- was ascribed to the scavenging effect of reactive species (Eqs. (11)and(12))^[25]. Moreover, the presence of CO₃^2- could generate a large number of OH^- by hydrolysis reaction (Eq. (13)) and the generation of OH^- had a negative effect on Hg⁰ removal as observed in Figure 6.

  $ {\rm{CO}}_3^{2-}{\rm{ + }} \cdot {\rm{OH}} \to \cdot {\rm{CO}}_3^-{\rm{ + O}}{{\rm{H}}^{\rm{-}}} $

  $ {\rm{CO}}_3^{2-}{\rm{ + }}{{\rm{h}}^{\rm{ + }}} \to \cdot {\rm{CO}}_3^- $

  $ {\rm{CO}}_3^{2-}{\rm{ + }}{{\rm{H}}_{\rm{2}}}{\rm{O}} \leftrightarrow {\rm{HCO}}_3^-{\rm{ + O}}{{\rm{H}}^{\rm{-}}} $

  From Figure 7, it is also observed that a little decrease in Hg⁰ removal appears in the presence of Cl^- as a result of the depletions of ·OH, H⁺ and h⁺ as shown below^{[24, 26]}:

  $ {\rm{C}}{{\rm{l}}^{\rm{-}}}{\rm{ + }} \cdot {\rm{OH}} \leftrightarrow \cdot {\rm{HOC}}{{\rm{l}}^{\rm{-}}} $

  $ \cdot {\rm{HOC}}{{\rm{l}}^{\rm{-}}}{\rm{ + }}{{\rm{H}}^{\rm{ + }}} \to {\rm{Cl}} \cdot {\rm{ + }}{{\rm{H}}_{\rm{2}}}{\rm{O}} $

  $ {\rm{C}}{{\rm{l}}^{\rm{-}}}{\rm{ + }}{{\rm{h}}^{\rm{ + }}} \to {\rm{Cl}} \cdot $

  During the Fenton-like oxidation process, the chlorine-containing radicals, such as ·HOCl^- and ·Cl generated from ·OH and Cl^- could also react with Hg⁰ to produced oxidized Hg²⁺ (Eqs. (17) and (18))^[27].

  $ {\rm{H}}{{\rm{g}}^{\rm{0}}}{\rm{ + }} \cdot {\rm{Cl}} \to {\rm{HgCl + }} \cdot {\rm{Cl}} \to {\rm{HgC}}{{\rm{l}}_{\rm{2}}} $

  $ {\rm{H}}{{\rm{g}}^{\rm{0}}}{\rm{ + 2}} \cdot {\rm{HOC}}{{\rm{l}}^{\rm{-}}} \to {\rm{HgC}}{{\rm{l}}_{\rm{2}}}{\rm{ + 2O}}{{\rm{H}}^{\rm{-}}} $

  By comparison, it was found that among the four anions, CO₃^2- ions showed the most effect on Hg⁰ removal under FSL irradiation.

  2.5   Effect of flue gas compositions

  2.5.1   Effect of O₂

  The effect of O₂ concentration on the removal of Hg⁰ was studied. As shown in Figure. 8, when the concentration of O₂ increases from 0 to 6%, more reactive ·O₂^- radicals are generated, which would promote Hg⁰ removal efficiency (Eqs. (19)and(20)). The catalyst in the absence of O₂ exhibits good Hg⁰ removal efficiency, which can be due to the presence of the adsorbed O₂ on the catalyst surface or the O₂ dissolved in the water^[28].

  $ {{\rm{O}}_{\rm{2}}}{\rm{ + }}{{\rm{e}}^{\rm{-}}} \to \cdot {\rm{O}}_2^- $

  $ \cdot {\rm{O}}_2^-{\rm{ + H}}{{\rm{g}}^{\rm{0}}}{\rm{ + 2}}{{\rm{H}}^{\rm{ + }}} \to {\rm{HgO + }}{{\rm{H}}_{\rm{2}}}{\rm{O}} $

  However, when O₂ concentration continuously increases to 12%, Hg⁰ removal efficiency slightly decreases. Excessive content of O₂ could result in the generation of hydrogen peroxide (H₂O₂) (Eq. (21)), reducing the yields of ·O₂^-[29]. Furthermore, the generated H₂O₂ could further react with electrons to produce ·OH and OH^- (Eq. (22)), leading to a decrease in Hg⁰ removal^[29].

  $ {{\rm{O}}_{\rm{2}}}{\rm{ + 2}}{{\rm{H}}^{\rm{ + }}}{\rm{ + 2}}{{\rm{e}}^{\rm{-}}} \to {{\rm{H}}_{\rm{2}}}{{\rm{O}}_{\rm{2}}} $

  $ {{\rm{H}}_{\rm{2}}}{{\rm{O}}_{\rm{2}}}{\rm{ + }}{{\rm{e}}^{\rm{-}}} \to \cdot {\rm{OH + O}}{{\rm{H}}^{\rm{-}}} $

  2.5.2   Effect of SO₂

  The effect of SO₂ concentration on Hg⁰ removal efficiency is given in the central region of Figure 8. When 600 mg/m³ of SO₂ is introduced into the reaction system, a great decrease in Hg⁰ removal efficiency from 98% to 62% appears. However, once SO₂ is cut off, the Hg⁰ removal efficiency restores to about 90%. Since SO₂ is soluble in water, the effect of SO₂ on Hg⁰ removal could be originated from two aspects: One for the dissolution of SO₂ in solution, and the other for the consumption of reactive species by SO₂. The detailed reactions can be as follows:

  图 8

  

  图 8  Effect of O₂, SO₂ and NO on Hg⁰ removal efficiency

  Figure 8.  Effect of O₂, SO₂ and NO on Hg⁰ removal efficiency

  下载: 全尺寸图片 幻灯片

  The dissolution reaction of SO₂ in reaction solution can lead to the generation of sulphite and bisulfite anions, which can readily react with ·OH, ·O₂^- and h⁺, affecting the removal of Hg⁰ as follows^{[24, 30]}:

  $ {\rm{S}}{{\rm{O}}_{\rm{2}}}{\rm{ + }}{{\rm{H}}_{\rm{2}}}{\rm{O}} \leftrightarrow {\rm{HSO}}_3^-{\rm{ + }}{{\rm{H}}^{\rm{ + }}} $

  $ {\rm{HSO}}_3^-\leftrightarrow {\rm{SO}}_3^{2-}{\rm{ + }}{{\rm{H}}^{\rm{ + }}} $

  $ {\rm{HSO}}_3^-{\rm{ + }} \cdot {\rm{OH}} \to \cdot {\rm{SO}}_3^-{\rm{ + }}{{\rm{H}}_{\rm{2}}}{\rm{O}} $

  $ {\rm{SO}}_3^{2-}{\rm{ + }} \cdot {\rm{OH + }}{{\rm{H}}^{\rm{ + }}} \to \cdot {\rm{SO}}_3^-{\rm{ + }}{{\rm{H}}_{\rm{2}}}{\rm{O}} $

  $ {\rm{SO}}_3^-{\rm{ + }} \cdot {\rm{OH}} \to {\rm{SO}}_4^{2-}{\rm{ + }}{{\rm{H}}^{\rm{ + }}} $

  $ {\rm{HSO}}_3^-{\rm{ + 2}}{{\rm{h}}^{\rm{ + }}}{\rm{ + }}{{\rm{H}}_{\rm{2}}}{\rm{O}} \to {\rm{SO}}_4^{2-}{\rm{ + 3}}{{\rm{H}}^{\rm{ + }}} $

  $ {\rm{HSO}}_3^-{\rm{ + }}{{\rm{H}}^{\rm{ + }}}{\rm{ + 2}} \cdot {\rm{O}}_2^-\to {\rm{SO}}_4^{2-}{\rm{ + }}{{\rm{O}}_{\rm{2}}}{\rm{ + }}{{\rm{H}}_{\rm{2}}}{\rm{O}} $

  Since the redox potentials of BiOI and AgI are higher than those of SO₄^2-/H₂SO₃ (0.172 eV) and SO₂/SO₄^2- (0.158 eV)^{[21, 28]}, the generated reactive species such as ·OH, ·O₂^- and h⁺ can be captured easily and consumed directly by SO₂ as described below, resulting in the underutilization of photocatalysts.

  $ {\rm{S}}{{\rm{O}}_{\rm{2}}}{\rm{ + 2}}{{\rm{h}}^{\rm{ + }}}{\rm{ + 2}}{{\rm{H}}_{\rm{2}}}{\rm{O}} \to {\rm{SO}}_4^{2-}{\rm{ + 4}}{{\rm{H}}^{\rm{ + }}} $

  $ {\rm{S}}{{\rm{O}}_{\rm{2}}}{\rm{ + }} \cdot {\rm{OH}} \to {\rm{HS}}{{\rm{O}}_{\rm{3}}} $

  $ {\rm{HS}}{{\rm{O}}_{\rm{3}}}{\rm{ + }} \cdot {\rm{OH}} \to {\rm{SO}}_4^{2-}{\rm{ + 2}}{{\rm{H}}^{\rm{ + }}} $

  $ {\rm{S}}{{\rm{O}}_{\rm{2}}}{\rm{ + 2}} \cdot {\rm{O}}_2^-\to {\rm{SO}}_4^{2-}{\rm{ + }}{{\rm{O}}_{\rm{2}}} $

  2.5.3   Effect of NO

  The effect of NO on Hg⁰ oxidation is displayed in the right region of Figure 8. The introductions of 600 mg/m³ of NO into the flue gas exhibit an obvious inhibitory effect on Hg⁰ removal, which could be due to the competition of NO with Hg⁰ for the active radicals on the catalyst surface^[14]. When NO is turned off, similar to SO₂, the Hg⁰ removal efficiency caused by NO also recovered. It was reported that the photocatalytic oxidation processes of NO usually underwent three states (NO → HNO₂, HNO₂ → NO₂, NO₂ → HNO₃)^{[28, 31]}. Thus, the pathway of NO oxidation could be described as follows:

  $ {\rm{NO + }} \cdot {\rm{OH}} \to {\rm{HN}}{{\rm{O}}_{\rm{2}}} $

  $ {\rm{HN}}{{\rm{O}}_{\rm{2}}}{\rm{ + }} \cdot {\rm{OH}} \to {\rm{N}}{{\rm{O}}_{\rm{2}}}{\rm{ + }}{{\rm{H}}_{\rm{2}}}{\rm{O}} $

  $ {\rm{N}}{{\rm{O}}_{\rm{2}}}{\rm{ + }} \cdot {\rm{OH}} \to {\rm{HN}}{{\rm{O}}_{\rm{3}}} $

  Furthermore, NO also can react with ·O₂^- and h⁺ as shown below:

  $ {\rm{NO + }} \cdot {\rm{O}}_2^-\to {\rm{NO}}_3^- $

  $ {\rm{NO + }}{{\rm{h}}^{\rm{ + }}}{\rm{ + }}{{\rm{H}}_{\rm{2}}}{\rm{O}} \to {\rm{HN}}{{\rm{O}}_{\rm{2}}}{\rm{ + }}{{\rm{H}}^{\rm{ + }}} $

  $ {\rm{2NO + 2}}{{\rm{h}}^{\rm{ + }}}{\rm{ + }}{{\rm{O}}_{\rm{2}}}{\rm{ + 2}}{{\rm{H}}_{\rm{2}}}{\rm{O}} \to {\rm{2NO}}_3^-{\rm{ + 4}}{{\rm{H}}^{\rm{ + }}} $

  It was clear that the competition of NO with Hg⁰ for the active radicals such as ·OH, ·O₂^- and h⁺ appeared on the surface of photocatalyst, leading a negative effect on Hg⁰ removal.

  2.6   Effect of catalyst dosage

  Figure 9 shows the effect of catalyst dosage on Hg⁰ removal efficiency.

  图 9

  

  图 9  Effect of catalyst dosage on Hg⁰ removal efficiency

  Figure 9.  Effect of catalyst dosage on Hg⁰ removal efficiency

  下载: 全尺寸图片 幻灯片

  It is found that without the aid of photocatalyst, only about 5% of Hg⁰ removal efficiency appears under FSL irradiation, which could be due to a small amount of UV light emitted by FSL^[32]. When 0.01 g of photocatalyst is added into 1 L reaction solution, Hg⁰ removal efficiency greatly increases to about 75% under FSL light. With the photocatalyst dosage further increasing from 0.01 g to 0.08 g, Hg⁰ removal efficiency significantly increases from 75% to 95% due to presence of large amounts of reactive radicals, while further adding the photocatalyst from 0.08 g to 0.20 g, only about 3% of Hg⁰ removal efficiency is improved. Because the residence time of flue gas through the reaction solution was stationary, Hg⁰ removal efficiency would be changed slightly even though the dosage of catalyst was enough and substantial reactive species were generated in the solution^[33]. In addition, Figure 9 also shows that the Hg⁰ removal efficiency decreases from 97% to 90% when 1.0 L reaction solution reduces to 0.5 L, indicating that the decrease in residence time between flue gas and reaction solution resulted in a lower Hg⁰ oxidation efficiency.

  2.7   Reusability and magnetism

  The recovery and reuse of AgI-BiOI/CoFe₂O₄ for Hg⁰ removal was examined throughout four consecutive cycles and the results are shown in Figure 10a. It can be clearly observed that about 8% of Hg⁰ removal efficiency is lost after four cycling runs, suggesting that AgI-BiOI/CoFe₂O₄ contains good potential for repeated use. Furthermore, the magnetic property of AgI-BiOI/CoFe₂O₄ after four successive runs is also measured by using vibrating sample magnetometer. As exhibited in Figure 10(b), the composites show a high saturation magnetization (24.4 emu/g) and a superior magnetic separation property (inset of Figure 10(b)), which provides a possibility for its multiple reutilization in the future.

  图 10

  

  图 10  (a) Cycling runs for photocatalytic oxidation of Hg⁰ over AgI-BiOI/CoFe₂O₄ and (b) magnetic hysteresis loops for AgI-BiOI/CoFe₂O₄ after four consecutive runs

  Figure 10.  (a) Cycling runs for photocatalytic oxidation of Hg⁰ over AgI-BiOI/CoFe₂O₄ and (b) magnetic hysteresis loops for AgI-BiOI/CoFe₂O₄ after four consecutive runs

  下载: 全尺寸图片 幻灯片

  2.8   Reaction products analysis

  To detect the mercury species in supernatant solution after reaction, the reduction test of reaction solution by SnCl₂ solution was carried out as follows: Firstly, 100 mL of supernatant was transferred into a gas-washing bottle, whose inlet kept open with air and the outlet connected with the VM-3000 mercury analyzer. The atmospheric air through the gas-washing bottle was sent into the mercury analyzer using a built-in gas pump. Then, after 10 min later, 50 mL of SnCl₂ solution (10%) was quickly added into the above solution from the inlet of the gas-washing bottle. The Hg⁰ concentration was recorded by the on-line mercury analyzer. Based on the above method, the reduction test of reaction solution in the presence of SO₂ or NO was implemented. It should be noticed that to avoid the effect of physical-adsorbed Hg⁰ dissolved in solution, the supernatant after reaction was first bubbled with air for 30 min before the reduction reaction. From Figure 11, it can be seen that no Hg⁰ is detected from the supernatant in the first 10 min, however, once SnCl₂ solution is added into the supernatant, the concentration of Hg⁰ sharply rises, suggesting that Hg⁰ in the reaction stream is completely oxidized and Hg²⁺ is the final oxidation product.

  图 11

  

  图 11  Detection of Hg⁰ from supernatant liquids by SnCl₂ solution

  Figure 11.  Detection of Hg⁰ from supernatant liquids by SnCl₂ solution

  下载: 全尺寸图片 幻灯片

  To understand the ion products in the supernatants after reaction with the presence of SO₂ and NO under fluorescent light, ion chromatography (IC) were employed to detect the possible species of SO₃^2-, SO₄^2-, NO₂^- and NO₃^-. As shown in Table 2, only SO₄^2- and NO₃^- are observed in the supernatants after reactions, which are attributed to the final oxidation products of SO₂ and NO by reactive species. Furthermore, to assess the stability of photocatalyst, an inductively coupled plasma mass spectrometer (ICP-MS) was utilized to determine the Co²⁺ concentration in the spent solution of SO₂ and NO. The results show that Co²⁺ concentration is only about 1.94 and 1.21 mg/L, respectively, indicating that a small amount of Co²⁺ would be separated out from the magnetic AgI-BiOI/CoFe₂O₄ photocatalyst in the presence of concentration of SO₂ or NO. Thus, the possible exposure of little amount of Co²⁺ to environment and the study of stability of AgI-BiOI/CoFe₂O₄ photocatalyst in acidic solution should be emphasized and carried out in the further study.

  表 2

  

  表 2  Reaction products of SO₂ and NO by AgI-BiOI/CoFe₂O₄

  Table 2.  Reaction products of SO₂ and NO by AgI-BiOI/CoFe₂O₄

  下载: 导出CSV

  Ion category	SO32-	SO42-	NO2-	NO3-
  Measured concentration	0	40.32 mg/L	0	15.50 mg/L

  3.   Conclusions

  In this work, wet photocatalytic oxidation removal of Hg⁰ from flue gas using AgI-BiOI/CoFe₂O₄ photocatalyst under FSL irradiation was investigated under different conditions. From the experimental results, the following conclusions can be drawn:

  AgI-BiOI/CoFe₂O₄ composites exhibited a poor thermal stability. BiOI presented in AgI-BiOI/CoFe₂O₄ would gradually transformed into Bi₅O₇I after calcinated at 400 ℃, and AgI-BiOI/CoFe₂O₄ hybrids significantly agglomerated at 600 ℃, leading to lower Hg⁰ removal activity.

  With increasing reaction solution temperature, Hg⁰ removal efficiency increased first and then decreased. Acid solution had little impact on Hg⁰ removal while alkaline solution can greatly decrease Hg⁰ removal efficiency.

  The presence of SO₂ had an inhibitory effect on Hg⁰ removal while the inhibition of NO on Hg⁰ removal was relatively small. The analysis of reaction products indicated that SO₄^2-, NO₃^- and Hg²⁺ were the final products of SO₂, NO and Hg⁰ after reaction.

• 1. [1]

     YANG S J, GUO Y F, YAN N Q, WU D Q, HE H P, XIE J K, QU Z, JIA J P. Remarkable effect of the incorporation of titanium on the catalytic activity and SO₂ poisoning resistance of magnetic Mn-Fe spinel for elemental mercury capture[J]. Appl Catal B:Environ, 2011, 10(3/4):  698-708.

  2. [2]

     ZHOU Q, DUAN Y F, HONG Y G, ZHU C, SHE M, ZHANG J, WEI H Q. Experimental and kinetic studies of gas-phase mercury adsorption by raw and bromine modified activated carbon[J]. Fuel Process Technol, 2015, 134:  325-332. doi: 10.1016/j.fuproc.2014.12.052

  3. [3]

     YOU C F, XU X C. Coal combustion and its pollution control in China[J]. Energy, 2010, 35(11):  4467-4472. doi: 10.1016/j.energy.2009.04.019

  4. [4]

     WU J, LI C E, ZHAO X Y, WU Q, QI X M, CHEN X T, HU T, CAO Y. Photocatalytic oxidation of gas-phase Hg⁰ by CuO/TiO₂[J]. Appl Catal B:Environ, 2015, 176/177:  559-569. doi: 10.1016/j.apcatb.2015.04.044

  5. [5]

     张华伟, 陈江艳, 赵可, 牛庆欣, 王力. Mn/Ce掺杂改性半焦对模拟煤气中单质汞的脱除性能研究[J]. 燃料化学学报, 2016,44,(4): 394-400. ZHANG Hua-wei, CHEN Jiang-yan, ZHAO Ke, NIU Qing-xin, WANG Li. Removal of vapor-phase elemental mercury from simulated syngas using semi-coke modified by Mn/Ce doping[J]. J Fuel Chem Technol, 2016, 44(4):  394-400.

  6. [6]

     HAN L N, HE X X, YUE C X, HU Y F, LI L N, CHANG L P, WANG H, WANG J C. Fe doping Pd/AC sorbent efficiently improving the Hg⁰ removal from the coal-derived fuel gas[J]. Fuel, 2016, 182:  64-72. doi: 10.1016/j.fuel.2016.05.046

  7. [7]

     WANG F M, LI G L, SHEN B X, WANG Y Y, HE C. Mercury removal over the vanadia-titania catalyst in CO₂-enriched conditions[J]. Chem Eng J, 2015, 263:  356-363. doi: 10.1016/j.cej.2014.10.091

  8. [8]

     JEON S H, EOM Y J, LEE T G. Photocatalytic oxidation of gas-phase elemental mercury by nanotitanosilicate fibers[J]. Chemosphere, 2008, 71(5):  969-974. doi: 10.1016/j.chemosphere.2007.11.050

  9. [9]

     WU J, LI X, REN J X, QI X M, HE P, NI B, ZHANG C, HU C Z, ZHOU J. Experimental study of TiO₂ hollow microspheres removal on elemental mercury in simulated flue gas[J]. J Ind Eng Chem, 2015, 32:  49-57. doi: 10.1016/j.jiec.2015.07.019

  10. [10]

      SNIDER G, ARIYA P. Photo-catalytic oxidation reaction of gaseous mercury over titanium dioxide nanoparticle surfaces[J]. Chem Phys Lett, 2010, 491(1/3):  23-28.

  11. [11]

      SHEN H Z, IE I R, YUAN C S, HUNG C H, CHEN W H, LUO J J, JEN Y H. Enhanced photocatalytic oxidation of gaseous elemental mercury by TiO₂ in a high temperature environment[J]. J Hazard Mater, 2015, 289:  235-243. doi: 10.1016/j.jhazmat.2015.02.033

  12. [12]

      袁媛, 张军营, 赵永椿, 王宇翔, 郑楚光. SO₂和NO浓度对TiO₂-硅酸铝纤维脱除元素汞的影响[J]. 燃料化学学报, 2012,40,(5): 630-635. YUAN Yuan, ZHANG Jun-ying, ZHAO Yong-chun, WANG Yu-xiang, ZHENG Chu-guang. Effects of SO₂ and NO on removal of elemental mercury using a TiO₂-aluminum silicate fiber[J]. J Fuel Chem Technol, 2012, 40(5):  630-635.

  13. [13]

      LI Y, WU C Y. Kinetic study for photocatalytic oxidation of elemental mercury on a SiO₂-TiO₂ nanocomposite[J]. Environ Eng Sci, 2007, 24(1):  3-12. doi: 10.1089/ees.2007.24.3

  14. [14]

      YUAN Y, ZHANG J Y, LI H L, LI Y, ZHAO Y C, ZHENG C G. Simultaneous removal of SO₂, NO and mercury using TiO₂-aluminum silicate fiber by photocatalysis[J]. Chem Eng J, 2012, 192(2):  21-28.

  15. [15]

      QI X M, GU M L, ZHU X Y, WU J, LONG H M, HE K, WU Q. Fabrication of BiOIO₃ nanosheets with remarkable photocatalytic oxidation removal for gaseous elemental mercury[J]. Chem Eng J, 2016, 285:  11-19. doi: 10.1016/j.cej.2015.09.055

  16. [16]

      DONG G H, HO W K, ZHANG L Z. Photocatalytic NO removal on BiOI surface:The change from nonselective oxidation to selective oxidation[J]. Appl Catal B:Environ, 2015, 168(168/169):  490-496.

  17. [17]

      OU M, ZHONG Q, ZHANG S L, NIE H Y, LV Z J, CAI W. Graphene-decorated 3D BiVO₄ superstructure:Highly reactive (040) facets formation and enhanced visible-light-induced photocatalytic oxidation of NO in gas phase[J]. Appl Catal B:Environ, 2016, 193:  160-169. doi: 10.1016/j.apcatb.2016.04.029

  18. [18]

      ZHANG A C, XING W B, ZHANG D, WANG H, CHEN G Y, XIANG J. A novel low-cost method for Hg⁰removal from flue gas by visible-light-driven BiOX (X=Cl, Br, I) photocatalysts[J]. Catal Commun, 2016, 87:  57-61. doi: 10.1016/j.catcom.2016.09.003

  19. [19]

      ZHANG A C, ZHANG L X, CHEN X Z, ZHU Q F, LIU Z C, XIANG J. Photocatalytic oxidation removal of Hg⁰ using ternary Ag/AgI-Ag₂CO₃ hybrids in wet scrubbing process under fluorescent light[J]. Appl Surf Sci, 2017, 392:  1107-1116. doi: 10.1016/j.apsusc.2016.09.116

  20. [20]

      ZHANG L X, ZHANG A C, LU H, SUN Z J, SHENG W, SUN L S, XIANG J. Magnetically separable AgI-BiOI/CoFe₂O₄ hybrid composites for Hg⁰ removal:Characterization, activity and mechanism[J]. RSC Adv, 2017, 7(50):  31448-31456. doi: 10.1039/C7RA04175F

  21. [21]

      CHENG H F, HUANG B B, DAI Y, QIN X Y, ZHANG X Y. One-Step synthesis of the nanostructured AgI/BiOI composites with highly enhanced visible-light photocatalytic performances[J]. Langmuir, 2010, 26(9):  6618-6624. doi: 10.1021/la903943s

  22. [22]

      YU C L, FAN C F, YU J C, ZHOU W Q, YANG K. Preparation of bismuth oxyiodides and oxides and their photooxidation characteristic under visible/UV light irradiation[J]. Mater Res Bull, 2011, 46(1):  140-146. doi: 10.1016/j.materresbull.2010.08.013

  23. [23]

      CAO J, LI X, LIN H L, XU B Y, LUO B D, CHEN S F. Low temperature synthesis of novel rodlike Bi₅O₇I with visible light photocatalytic performance[J]. Mater Lett, 2012, 76(6):  181-183.

  24. [24]

      RAUF M A, MARZOUKI N, KORBAHTI B K. Photolytic decolorization of Rose Bengal by UV/H₂O₂ and data optimization using response surface method[J]. J Hazard Mater, 2008, 159(2/3):  602-609.

  25. [25]

      SANTIAGO D E, ARANA J, GONZÁLEZ-D O, ALEMÁN-D M E, ACOSTA-D A C, FERNANDEZ-R C, PÉREZ-P J, DONA-R J M. Effect of inorganic ions on the photocatalytic treatment of agro-industrial wastewaters containing imazalil[J]. Appl Catal B:Environ, 2014, 156/157(3):  284-292.

  26. [26]

      CHEN J, HU Z, WANG D, GAO C, JI R. Photocatalytic mineralization of dimethoate in aqueous solutions using TiO₂:Parameters and by-products analysis[J]. Desalination, 2010, 258(1):  28-33.

  27. [27]

      ZHAO Y, HAO R L. Macrokinetics of Hg⁰ removal by a vaporized multicomponent oxidant[J]. Ind Eng Chem Res, 2014, 53(27):  10899-10905. doi: 10.1021/ie5009376

  28. [28]

      XIA D H, HU L L, HE C, PAN W Q, YANG T S, YANG Y C, SHU D. Simultaneous photocatalytic elimination of gaseous NO and SO₂ in a BiOI/Al₂O₃-padded trickling scrubber under visible light[J]. Chem Eng J, 2015, 279:  929-938. doi: 10.1016/j.cej.2015.05.097

  29. [29]

      KIM J, LEE C W, CHOI W. Platinized WO₃ as an environmental photocatalyst that generates OH radicals under visible light[J]. Environ Sci Technol, 2010, 44(17):  6849-6854. doi: 10.1021/es101981r

  30. [30]

      LIU Y X, ZHANG J, SHENG C D, ZHANG Y C, ZHAO L. Simultaneous removal of NO and SO₂ from coal-fired flue gas by UV/H₂O₂ advanced oxidation process[J]. Chem Eng J, 2010, 162(3):  1006-1011. doi: 10.1016/j.cej.2010.07.009

  31. [31]

      LASEK J, YU Y H, WU J C S. Removal of NO_x by photocatalytic processes[J]. J Photochem Photobiol C, 2013, 14(1):  29-52.

  32. [32]

      MCLARNON C R, GRANITE E J, PENNLINE H W. The PCO Process for photochemical removal of mercury from flue gas[J]. Fuel Process Technol, 2005, 87(1):  85-89. doi: 10.1016/j.fuproc.2005.07.001

  33. [33]

      LIU Y X, ZHANG J, SHENG C D, ZHANG Y C, ZHAO L. Preliminary study on a new technique for wet removal of nitric oxide from simulated flue gas with an ultraviolet (UV)/H₂O₂ process[J]. Energy Fuels, 2010, 24(9):  4925-4930. doi: 10.1021/ef1006325

• 

  Figure 1  Schematic diagram of the experimental setup

  下载: 全尺寸图片 幻灯片
  

  Figure 2  TG (a) and DSC (b) curve for AgI-BiOI/CoFe₂O₄

  下载: 全尺寸图片 幻灯片
  

  Figure 3  XRD patterns of the photocatalysts

  下载: 全尺寸图片 幻灯片
  

  Figure 4  Effect of calcination temperature on Hg⁰ removal efficiency

  conditions: baseline flue gas, t=35 ℃, initial pH=7, solution volume=1 L, FSL radiation intensity=11 W/L, catalyst dosage=200 mg/L

  下载: 全尺寸图片 幻灯片
  

  Figure 5  Effect of reaction solution temperature on Hg⁰ removal efficiency

  conditions: baseline flue gas, initial pH=7, solution volume=1 L, FSL radiation intensity=11 W/L, catalyst dosage=200 mg/L

  下载: 全尺寸图片 幻灯片
  

  Figure 6  Effect of solution pH on Hg⁰ removal efficiency

  conditions: baseline flue gas, t=35 ℃, solution volume=1 L, FSL radiation intensity=11 W/L, catalyst dosage=200 mg/L

  下载: 全尺寸图片 幻灯片
  

  Figure 7  Effect of anions in solution on Hg⁰ removal efficiency

  conditions: baseline flue gas, t=35 ℃, initial pH=7, solution volume=1 L, FSL radiation intensity=11 W/L, catalyst dosage=200 mg/L

  下载: 全尺寸图片 幻灯片
  

  Figure 8  Effect of O₂, SO₂ and NO on Hg⁰ removal efficiency

  error bars represent standard deviation of means (n=4) conditions: Hg⁰=55.0 μ g/m³, t=35 ℃, initial pH=7, solution volume=1 L, FSL radiation intensity=11 W/L, catalyst dosage=200 mg/L

  下载: 全尺寸图片 幻灯片
  

  Figure 9  Effect of catalyst dosage on Hg⁰ removal efficiency

  conditions: Hg⁰=55.0 μ g/m³, t=35 ℃, initial pH=7, FSL radiation intensity=11 W/L

  下载: 全尺寸图片 幻灯片
  

  Figure 10  (a) Cycling runs for photocatalytic oxidation of Hg⁰ over AgI-BiOI/CoFe₂O₄ and (b) magnetic hysteresis loops for AgI-BiOI/CoFe₂O₄ after four consecutive runs

  error bars represent standard deviation of means (n=3) conditions: Hg⁰=55.0 μ g/m³, t=35 ℃, initial pH=7, solution volume=1 L, FSL radiation intensity=11 W/L, catalyst dosage=200 mg/L

  下载: 全尺寸图片 幻灯片
  

  Figure 11  Detection of Hg⁰ from supernatant liquids by SnCl₂ solution

  下载: 全尺寸图片 幻灯片

  Table 1.  Physical features of AgI-BiOI/CoFe₂O₄ hybrids calcinated at different temperatures

  Sample	BET surface area A/(m2·g-1)	Total pore volume v/(cm3·g-1)
  Without calcination	20.7	0.066
  200 ℃ calcination	33.9	0.114
  400 ℃ calcination	15.1	0.056
  600 ℃ calcination	1.5	0.002

  下载: 导出CSV

  Table 2.  Reaction products of SO₂ and NO by AgI-BiOI/CoFe₂O₄

  Ion category	SO32-	SO42-	NO2-	NO3-
  Measured concentration	0	40.32 mg/L	0	15.50 mg/L

  下载: 导出CSV
•