Magnetically separable AgI–BiOI/CoFe₂O₄ hybrid composites for Hg⁰ removal: characterization, activity and mechanism

Abstract

A series of magnetically separable AgI–BiOI/CoFe₂O₄ hybrid composites were successfully synthesized via a solvothermal and subsequent coprecipitation method. The microstructure and magnetism of the materials were characterized by X-ray diffraction (XRD), N₂ adsorption–desorption, scanning electron microscopy (SEM), transmission electron microscopy (TEM), high-resolution transmission electron microscopy (HRTEM), X-ray photoelectron spectroscopy (XPS), UV-vis diffuse reflectance spectroscopy (DRS), photocurrent test, electron spin resonance (ESR) and vibrating sample magnetometer (VSM). The photocatalytic performance of AgI–BiOI/CoFe₂O₄ composites on Hg⁰ removal from simulated flue gas was carefully designed and evaluated under fluorescent light (FSL) irradiation. The results showed that AgI–BiOI/CoFe₂O₄ composites displayed superior photocatalytic activities because of the synergistic effects between AgI, BiOI, and CoFe₂O₄ under FSL irradiation. The optimal weight ratio between AgI and the total weight of AgI–BiOI/CoFe₂O₄ photocatalyst was 0.3. The presence of a small amount of SO₂ had a dramatic inhibition on Hg⁰ removal, while the inhibitory effect of NO on Hg⁰ removal could only be observed at a higher NO concentration. The trapping experiments indicated that photoinduced holes (h⁺) and superoxide radicals (˙O₂⁻) were the primary active substances in the AgI–BiOI/CoFe₂O₄ photocatalytic oxidation system. According to the experimental and characterization results, one plausible mechanism for enhanced Hg⁰ removal performance over AgI–BiOI/CoFe₂O₄ composites was proposed.

---

1. Introduction

Mercury has received great attention owing to its toxicity, persistence and bioaccumulation in recent years.^1,2 Coal-fired power plants are thought to be responsible for most of the anthropogenic mercury emissions into the environment.^3,4 In 2011, the State Environmental Protection Administration of China issued a new national standard to specify the emission limits of mercury and its compounds from coal-fired flue gas.⁵ In China, coal accounts for about 70% of the total primary energy consumption.⁶ Mercury from flue gases often exists as elemental mercury (Hg⁰), oxidized mercury (Hg²⁺) and particulate mercury (Hg^p).⁷ Compared with Hg²⁺ and Hg^p, Hg⁰ is hardly removed owing to its volatility and indissolubility.⁸ To meet the increasingly severe emission standards, it is urgent to effectively remove Hg⁰ from coal-fired power plants.

In recent years, wet catalytic oxidation technology including UV/fenton,⁹ UV/H₂O₂,¹⁰ H₂O₂/Na₂S₂O₈,¹¹ NaClO₂,¹² and gas-phase photocatalytic oxidation with TiO₂,¹³ SiO₂–TiO₂,¹⁴ CuO/TiO₂,¹⁵ TiO₂–aluminum silicate fiber,¹⁶ and BiOIO₃,¹⁷ as photocatalysts have been explored extensively to remove Hg⁰ from flue gas. Most of these technologies can effectively remove Hg⁰ with the aid of UV energy.^9,13–17 Under UV light irradiation, the produced reactive species such as hydroxyl radicals (˙OH) and hydroperoxyl radicals (˙HO₂) possessing higher oxidability would oxidize Hg⁰ into Hg²⁺. This technique is promising since it could be carried out at low temperatures. However, the method needs lots of UV light irradiation and H₂O₂ or with the assistance of TiO₂. Moreover, TiO₂ itself has a large band gap (about 3.2 eV), which could only be excited under UV light to produce some useful reactive substances. The high cost of supplying successive UV light in coal-fired power plant limits its practical application. To solve this problem, it is urgent to fabricate a novel photocatalyst that can generate reactive species under visible light or even sunlight irradiation.

To date, a lot of novel Ag-based and Bi-based photocatalytic materials have been employed as efficient visible-light-responsive photocatalysts, such as Ag₂CO₃,¹⁸ Ag₃PO₄,¹⁹ Ag–AgI/Al₂O₃,²⁰ BiOX (X = Cl, Br, I),²¹ AgI/BiOI,²² and AgI/Bi₅O₇I.²³ During the photocatalytic oxidation processes, large amounts of reactive substances such as holes (h⁺), superoxide radicals (˙O₂⁻) and ˙OH were generated. Hg⁰ could be possibly oxidized by these radicals due to their superior oxidation ability. To realize the above inference, the activity of Hg⁰ removal using visible-light-driven Ag/AgI–Ag₂CO₃ photocatalyst was investigated tentatively in our previous studies.²⁴ The results indicated that this photocatalytic oxidation technology can effectively remove Hg⁰ under fluorescent light. Moreover, this technology can combine with wet flue gas desulfurization system (WFGD) to simultaneously remove Hg⁰ and SO₂ at relatively low temperature, which brings out us a new insight into Hg⁰ removal from flue gas.

Although photocatalysis is efficient under fluorescent light, however, the applications of this type of powder photocatalysts also face a major bottleneck such as separation and cyclic utilization.^25,26 In consideration of the easy separation of powder photocatalysts from reaction solution, a useful strategy to integrate Ag-based or Bi-based composites with magnetic nanoparticles such as Fe₃O₄,²⁷ CoFe₂O₄,²⁸ ZnFe₂O₄,²⁹ NiFe₂O₄,³⁰ and SrFe₁₂O₁₉,³¹ was proposed. Among the magnetic materials, CoFe₂O₄ has been widely employed in photocatalytic degradation of organic pollutants due to its small band gap energy and moderate saturation magnetization.^28,32 Moreover, it also was reported that there were significant synergistic effects of individual components in CoFe₂O₄-based nanocomposites. For example, when active components such as BiOBr,³³ BiVO₄,³⁴ ZnO,³⁵ CdS,³⁶ and Ag/Ag₃VO₄,³⁷ were combined with CoFe₂O₄ respectively and employed in photocatalytic degradation of organic pollutants, the hybrid composites all showed enhanced photocatalytic activities than single component under visible light irradiation, and more importantly, they could be used for many times and recovered easily by an external magnetic force.

Thus, to reduce the cost of photocatalyst and obtain better photocatalytic activity with easy recovery, in this work, a series of novel magnetically separable AgI–BiOI/CoFe₂O₄ hybrids with lower silver content were synthesized via a solvothermal and subsequent coprecipitation method to remove Hg⁰ under fluorescent light (FSL) irradiation, which has not yet been researched in the literatures. As was expected, the as-synthesized ternary AgI–BiOI/CoFe₂O₄ composites displayed higher Hg⁰ removal activity than single component CoFe₂O₄ and BiOI under fluorescent light. The cyclic experiment showed that the optimal hybrid had stable photocatalytic activity and was relatively stable in crystal structures in the photocatalytic process. The synthesized ternary composite proved to be promising for wet photocatalytic oxidation removal of Hg⁰ from flue gas.

2. Experimental

2.1. Synthesis of CoFe₂O₄ nanoparticles

The CoFe₂O₄ composites were synthesized via a solvothermal method.³⁷ All chemical agents employed were of analytical grade. Typically, 1.24 g of Co(NO₃)₂·6H₂O and 3.44 g of Fe(NO₃)₃·9H₂O were dissolved into 40 mL deionized water to obtain a mixed solution with magnetic stirring for about 60 min. Then, 30 mL of 1 M NaOH solution was added dropwise into the above mixed solution with stirring for another 60 min. Subsequently, the suspension was placed into a Teflon-lined autoclave (100 mL) and heated at 180 °C for 12 h. The resultant product was allowed to cool down to ambient temperature and then washed with distilled water several times. Finally, the obtained CoFe₂O₄ (abbreviated as CFO) composites were dried under vacuum at 60 °C for 12 h.

2.2. Synthesis of AgI–BiOI/CoFe₂O₄ microparticles

AgI–BiOI/CoFe₂O₄ samples were fabricated via a modified coprecipitation method,^38,39 in which the mass ratio of CoFe₂O₄ (CFO) and total mass of the photocatalyst was fixed at 20%. Firstly, 4.82 g of Bi(NO₃)₃·5H₂O and 0.36 g of AgNO₃ were dissolved into 200 mL aqueous solution containing 20 mL acetic acid. Secondly, 1.00 g of as-prepared CFO powder was added into the above solvent with continuous stirring for 60 min. Then, 100 mL of 0.12 M KI solution was added to the above suspension. Finally, after stirring for another 60 min and placed for 12 h, the resulting precipitate was washed with ethanol water mixture for 3–5 times, filtered and dried at 60 °C for 24 h, respectively. The obtained composites were designated as AgI_0.1BiOI/CFO for simplicity, where the value of 0.1 was the weight ratio between AgI and the total weight of the photocatalyst. Equally, other different photocatalysts were synthesized via the same procedure and named as BiOI/CFO, AgI/CFO, AgI_0.2BiOI/CFO, AgI_0.3BiOI/CFO, AgI_0.4BiOI/CFO, and AgI_0.5BiOI/CFO. For comparison, pure AgI and BiOI were also synthesized via the same method without the addition of CoFe₂O₄.

2.3. Characterization of photocatalysts

X-ray powder diffraction (XRD) analyses were performed on a D8 advance diffractometer (Bruker, Germany). N₂ adsorption–desorption tests were carried out on an Autosorb iQ surface area and porosity analyzer (Quantachrome, USA) and the photocatalysts were degassed at 150 °C in vacuum for 8 h before the nitrogen adsorption isotherm was constructed. Scanning electron microscope (SEM) was carried out on a field emission scanning electron microscope (NoVa™ Nano SEM 430, FEI Company). Transmission electron microscope (TEM) and high-resolution transmission electron microscope (HRTEM) analyses were conducted on a Tecnai G20 microscopy (FEI, USA). X-ray photoelectron spectroscopy (XPS) experiments were performed on an Escalab 250xi spectrometer (Thermo Fisher, USA). UV-vis diffuse reflection spectra (DRS) were recorded on an UV2401 spectrophotometer (Shimadzu, Japan). The free radicals were detected on an ER200-SRC electron spin resonance spectrometer (ESR, Bruker, Germany). The magnetism of the composite was analyzed on a MPMS XL-7 magnetometer (Quantum Design, USA).

2.4. Photoelectrochemical measurement

The photoelectrochemical performances of the composites were conducted on a CHI 760E electrochemical work-station (Chenhua, China) in a standard three electrode system using fabricated photo-anode as the working electrode. A Pt wire and a saturated calomel electrode were respectively utilized as the counter and reference electrode. The 0.5 M Na₂SO₄ solution (pH = 5.8) purged with N₂ was utilized as the electrolyte. A 300 W PLS-SXE300 Xe illuminator with a UV-cutoff filter (λ ≥ 420 nm) was used as the light source.

2.5. Photocatalytic activity test

The photocatalytic performances of as-prepared composites were implemented on a laboratory-scale wet bubbling photocatalytic reactor at ambient temperature. A detailed description about the experimental setup and process was provided in our previous work.²⁴ In the experiment, the simulated flue gas was composed of 6% of O₂, 12% of CO₂, 55 μg m⁻³ of Hg⁰ vapor, and balance N₂. The total flow rate was approximately 1.50 L min⁻¹. 0.2 g of photocatalyst and 1 L of deionized water were uniformly mixed as the reaction solution and the reaction temperature was set as 45 °C. During the test, Hg⁰ concentration data was collected by an online VM-3000 mercury analysis instrument. The Hg⁰ removal efficiency η (%) was calculated according to the following equation:

η = (1 − C_out/C_in) × 100 (1)

where C_out and C_in represent the outlet and inlet Hg⁰ concentrations (μg m⁻³), respectively.

3. Results and discussion

3.1. Photocatalytic activity

3.1.1. Effect of AgI content. The photocatalytic performances of Hg⁰ by as-prepared materials were studied under fluorescent light (FSL) irradiation and the corresponding results were shown in Fig. 1. It can be observed that under FSL irradiation, pure CFO and BiOI could only remove about 6% and 42% of Hg⁰ in 60 min, respectively, whereas approximately 68% of Hg⁰ was removed by BiOI/CFO. This comparison suggested that BiOI had relatively poor Hg⁰ removal ability, but the introduction of magnetic CFO onto BiOI could efficiently enhance its photocatalytic performance. The above phenomena can be explained that because of a small band gap of CoFe₂O₄ and a high e⁻–h⁺ pair recombination rate, pure CFO can hardly oxidize Hg⁰ to Hg²⁺. However, when BiOI was combined with CFO, a more efficient charge separation from BiOI/CFO can be achieved since the introduction of CFO can enhance the photoluminescent intensity of metal oxides.^33,40

Fig. 1 Effect of AgI content on Hg⁰ removal efficiency.

Furthermore, when different amounts of AgI were doped with BiOI/CFO, all ternary AgI–BiOI/CFO composites displayed much higher Hg⁰ removal efficiency than BiOI/CFO and the photocatalytic performances of ternary AgI–BiOI/CFO composites depended on AgI content. The AgI_0.3BiOI/CFO exhibited the highest Hg⁰ removal efficiency (99%) when the weight ratio between AgI and AgI–BiOI/CFO was increased from 10% to 50%. It was reported that the composites with multifunctional components could combine individual advantages to achieve a high photocatalytic performance.⁴¹ Thus, a much better Hg⁰ removal efficiency could be obtained for AgI–BiOI/CFO hybrids.

3.1.2. Effect of FSL illumination. Fig. 2 shows the effect of FSL illumination on Hg⁰ removal efficiency. It was clear that only about 5% and 35% of Hg⁰ removal efficiencies were observed only with FSL irradiation and only with AgI_0.3BiOI/CFO, respectively, indicating that Hg⁰ cannot be easily removed without FSL irradiation or photocatalyst. However, once both AgI_0.3BiOI/CFO composite and FSL irradiation were employed simultaneously in the reaction system, Hg⁰ removal efficiency dramatically increased to about 99%. The enhanced Hg⁰ removal activity could be evidently due to the synergistic effect of AgI_0.3BiOI/CFO and FSL irradiation.

Fig. 2 Effect of FSL irradiation on Hg⁰ removal efficiency.

3.1.3. Effect of flue gas composition. In the coal-fired power plants, actual exhaust gas inevitably contained some concentrations of SO₂ and NO, which may exert inhibitory or promotional impacts on Hg⁰ removal. Fig. 3 shows the effects of SO₂ and NO concentrations on Hg⁰ removal. When 200 ppm of SO₂ was introduced into the reaction stream in the ranges of 30–60 and 90–120 min, greater inhibitory effects on Hg⁰ removal appeared. However, when 100 and 200 ppm of NO were added, there was almost no loss in Hg⁰ removal efficiency, but a decrement of Hg⁰ removal efficiency from 93% to 75% was observed when NO concentration further increased to 600 ppm. Moreover, once SO₂ and NO were cut off, the removal efficiency of Hg⁰ restored to about its initial value, suggesting that the structure of the photocatalyst had not been destroyed by SO₂ and NO. According to previous reports,^9,42 the inhibition could be ascribed to the consumption of photogenerated reactive species by SO₂ and NO.

Fig. 3 Effect of SO₂ and NO on Hg⁰ removal efficiency.

3.2. Photocatalyst characterization

3.2.1. XRD and N₂ adsorption–desorption. The XRD patterns of CoFe₂O₄ (CFO), BiOI, AgI, and AgI–BiOI/CoFe₂O₄ samples with differing AgI content were shown in Fig. 4. The diffraction peaks shown in Fig. 4a–c were assigned to the signal phase of CoFe₂O₄ (JCPDS 22-1086), tetragonal BiOI (JCPDS 73-2062) and hexagonal AgI (JCPDS 09-0374), respectively.^22,43,44 By comparison, the diffraction intensities of BiOI in BiOI/CFO (Fig. 4d) and AgI in AgI/CFO (Fig. 4e) greatly decreased, suggesting the presence of CoFe₂O₄. It was clear that the diffraction patterns of AgI intensified gradually with the increasing of AgI/(AgI–BiOI/CFO) weight ratio (Fig. 4f–j), and the peak intensities of BiOI accordingly weakened. Moreover, two weak diffraction peaks of CoFe₂O₄ located at 2θ = 35.51° and 42.67° were observed from XRD patterns of Fig. 4f–j, indicating the coexistence of AgI, BiOI, and CoFe₂O₄ in AgI–BiOI/CoFe₂O₄ samples.

Fig. 4 XRD patterns of the photocatalysts.

Table 1 summarizes the physical features of selected composites. Obviously, CFO showed the largest BET surface area, total pore volume, and average pore diameter among the three samples. When BiOI itself or both BiOI and AgI were combined with CFO, the BET surface areas and total pore volumes of BiOI/CFO and AgI_0.3BiOI/CFO were greatly decreased, which indicated that BiOI and AgI were successfully doped with CFO.

Table 1 Physical features of the selected composites

Sample	BET surface area (m^2 g^−1)	Total pore volume (cm^3 g^−1)	Average pore diameter (nm)
CFO	55.07	2.61 × 10^−1	20.52
BiOI/CFO	21.28	7.85 × 10^−2	20.49
AgI0.3BiOI/CFO	20.74	6.56 × 10^−2	17.28

---

3.2.2. SEM, TEM and HRTEM. The morphology features of selective composites were analyzed by SEM, TEM and HRTEM. As shown in Fig. 5, CFO exhibited a rough irregular surface, while when AgI was combined with CoFe₂O₄, many AgI granules with smooth surface (see red circular marks) appeared. As for BiOI/CFO, the petal-like BiOI particles (see blue circular mark) were observed (Fig. 5c). From Fig. 5d, it can be seen that the petal-like BiOI particles in AgI_0.3BiOI/CFO composites became dispersed and somehow smaller than BiOI/CFO after the introduction of AgI. To provide the evidence that AgI and BiOI particles were successfully mixed with magnetic CFO, the TEM and HRTEM images of CFO and AgI_0.3BiOI/CFO were investigated. The TEM images shown in Fig. 5e and f indicated that the CoFe₂O₄ grains were uniformly distributed on the surface of AgI_0.3BiOI/CFO photocatalyst. The lattice spacing belonged to (311) plane of CFO was observed (Fig. 5g), while as for AgI_0.3BiOI/CFO (Fig. 5f), three different lattice spacings were observed. The lattice spacings of 0.25, 0.27 and 0.30 nm corresponded to the (311) plane of CFO, (102) plane of AgI and (102) plane of BiOI, respectively, which further indicated that BiOI and AgI were successfully doped with CFO.^36,45

Fig. 5 SEM images of (a) CFO, (b) AgI/CFO, (c) BiOI/CFO and (d) AgI_0.3BiOI/CFO, TEM images of (e) CFO and (f) AgI_0.3BiOI/CFO, and HRTEM images of (g) CFO and (h) AgI_0.3BiOI/CFO.

3.2.3. XPS. To obtain the surface compositions and chemical status of the as-prepared samples, XPS technique was employed and the corresponding results were displayed in Fig. 6. Fig. 6a exhibits the XPS spectra of Bi 4f. For BiOI/CFO, the peaks located at 164.49 and 159.15 eV were belonged to Bi 4f_5/2 and Bi 4f_7/2,⁴⁶ while the peaks of Bi 4f_5/2 and Bi 4f_7/2 from AgI_0.3BiOI/CFO shifted to higher binding energy. Fig. 6b shows the XPS spectra of I 3d. The peaks in BiOI/CFO centered at 630.33 and 618.93 eV corresponded to the spectrums of I 3d_3/2 and I 3d_5/2,⁴⁷ while for AgI/CFO and AgI_0.3BiOI/CFO, their peaks of I 3d_3/2 and I 3d_5/2 all shifted to higher binding energies. The peaks belonged to Ag 3d_3/2 and Ag 3d_5/2 was observed in Fig. 6c. It was noted that the peaks of Ag 3d in AgI_0.3BiOI/CFO have a red-shift to higher binding energy compared with these of AgI/CFO.

Fig. 6 XPS spectra of the photocatalysts: (a) Bi 4f, (b) I 3d, (c) Ag 3d, (d) Co 2p, and (e) Fe 2p.

The XPS spectra of Co 2p and Fe 2p were displayed in Fig. 6d and e, respectively. The peaks of both Co 2p and Fe 2p in BiOI/CFO, AgI/CFO and AgI_0.3BiOI/CFO were weaker than that of CFO due to the interposition of CFO into the composites and relatively lower content of CFO. In Fig. 6d, the peaks centered at 797.0 eV (with a satellite peak at 804.50 eV) and 781.8 eV (with a satellite peak at 787.1 eV) were ascribed to Co 2p_1/2 and Co 2p_3/2, which corresponded to the characteristic peaks of Co²⁺ ions in CoFe₂O₄.³⁷ Also, it was observed that the satellite peak of Co 2p_1/2 in BiOI/CFO and AgI_0.3BiOI/CFO shifted greatly to higher binding energies. In Fig. 6e, the photoionization region of Fe 2p yielded Fe 2p_3/2 binding energies of 712.32 eV and Fe 2p_1/2 binding energies of 725.84 eV, which is consistent with the Fe 2p binding energy for CoFe₂O₄.^37,43 According to previous works,^48,49 all the shifts of binding energy in the XPS spectra can be ascribed to the intense interaction between closely contacted phases of AgI, BiOI and CFO, suggesting the successful syntheses of AgI/CFO, BiOI/CFO and AgI_0.3BiOI/CFO.

3.2.4. DRS. The optical performances of selective composites were performed on a UV-vis spectrophotometer. From Fig. 7a, it was observed that CFO with black surface displayed an almost constant absorption capacity in the wavelength range of 300–800 nm, while pure AgI and BiOI exhibited the absorption edges of around 475 and 670 nm, respectively. When CFO was coupled with AgI and BiOI, the absorption edges of AgI/CFO and BiOI/CFO all shifted to shorter wavelengths than those of pure AgI and BiOI. However, the absorption strengths of AgI/CFO and BiOI/CFO increased in visible region in comparison to AgI and BiOI, indicating the synergistic effect between CoFe₂O₄ and AgI or BiOI. When BiOI/CFO was decorated with AgI, the absorption intensity of AgI_0.3BiOI/CFO decreased greatly compared to BiOI/CFO because the band gap of AgI was larger than BiOI.²² In the photocatalytic process, the component that can absorb more energy was beneficial for the photocatalytic activity of Hg⁰ removal to some extent, but the main factor could be the electron–hole (e⁻–h⁺) pairs separation efficiency of the component.^50,51

Fig. 7 (a) UV-vis diffuse reflectance spectra of selective samples; (b) plots of (αhν)² versus hν for AgI, CFO and plot of (αhν)² versus hν for BiOI.

According to the absorption spectra in Fig. 7a, the band gap values of the semiconductors can be measured by the equation as follows:

αhν = A(hν − E_g)^n/2 (2)

where α, h, ν, A, and E_g are the absorption coefficient, Planck constant, light frequency, a proportionality constant, and the band gap, respectively. The value of n depended on the types of photon transition in semiconductors (n = 1 for direct transition and n = 4 for indirect transition). Thus, the value of n is 1 for direct gap semiconductor AgI and CoFe₂O₄ while 4 for indirect gap semiconductor BiOI.^23,37 Hence, as Fig. 7b shows, the band gap energies (E_g) of AgI, BiOI and CoFe₂O₄ were estimated to be 2.56, 1.76 and 1.33 eV, respectively. The conduction band (CB) and valence band (VB) values of a semiconductor were measured by the following empirical equations:²²

E_CB = X − E^C − 0.5E_g (3)

E_VB = E_CB + E_g (4)

where E_CB and E_VB represent the CB and VB potentials, respectively; X is the electronegativity (the X values for AgI, BiOI and CoFe₂O₄ are 5.354, 5.941, and 5.815 eV, respectively^22,37); E^C represents the energy of free electrons on the hydrogen scale (4.5 eV) and E_g refers to the band gap energy of the semiconductor. Thus, the E_CB and E_VB of the composites were accordingly calculated and summarized in Table 2.

Table 2 Band gap, conduction and valence band of the samples

Sample	Band gap, Eg (eV)	Conduction band, ECB (eV)	Valence band, EVB (eV)
AgI	2.56	−0.43	2.13
BiOI	1.76	0.56	2.32
CoFe2O4	1.33	0.65	1.98

---

3.2.5. Photocurrent analysis and ESR. The above DRS analysis can be further supported by prompt photocurrent response of the composites under visible light irradiation. In general, the higher the e⁻–h⁺ pairs separation efficiency, the higher the photocurrent produced, thereby leading to a much higher photocatalytic activity. Fig. 8 compares the prompt photocurrent responses of different samples and a good reproducibility of the photocurrent was observed in several on/off cycling tests. The photocurrent of BiOI/CFO was steadier than those of AgI/CFO and AgI_0.3BiOI/CFO, implying that there was little recombination process within the whole irradiation period. In comparison to BiOI/CFO and AgI/CFO, AgI_0.3BiOI/CFO photocatalyst exhibited much higher photocurrents density. This suggested that the e⁻–h⁺ pairs separation performance in the photo-electrochemical process of AgI_0.3BiOI/CFO had been promoted, which could be ascribed to the appropriate energy band matching between AgI, BiOI and CoFe₂O₄.

Fig. 8 Prompt photocurrent responses of the samples under visible light irradiation.

To explore the reactive substances produced during the photocatalytic oxidation process, ESR technology was utilized and the corresponding result was shown in Fig. 9. Under visible light irradiation, the characteristic peaks of DMPO–˙OH were detected in the suspension of AgI_0.3BiOI/CFO, while no such signal appeared in the dark.⁵² Furthermore, when AgI_0.3BiOI/CFO was dispersed in methanol, DMPO–˙O₂⁻ signals also only appeared under visible light. In the photocatalytic process, ˙OH, ˙O₂⁻ and holes played vital roles in pollutant removal.^24,53–55 Thus, the ESR results provided a strong indication that the photogenerated carriers in AgI_0.3BiOI/CFO possibly had not only strong redox ability but also survived long enough to react with the absorbed O₂ and OH⁻ or H₂O to generate reactive ˙O₂⁻ and ˙OH.⁵⁵

Fig. 9 ESR spectra for DMPO–˙OH and DMPO–˙O₂⁻ of AgI_0.3BiOI/CFO.

3.2.6. Magnetic performance. Fig. 10 shows the magnetic hysteresis loops of CFO and AgI_0.3BiOI/CFO. CFO evidently displayed a strong magnetic response to the varied magnetic field. Since AgI_0.3BiOI/CFO only contained 20 wt% of CFO and the components of AgI and BiOI have no magnetism, the magnetic saturation (Ms) value of AgI_0.3BiOI/CFO (14.5 emu g⁻¹) was smaller than that of CFO (61.9 emu g⁻¹). However, from the inset of Fig. 10, it found that AgI_0.3BiOI/CFO demonstrated a relatively strong enough magnetic separation performance and could be employed as a promising magnetic photocatalyst in the future.

Fig. 10 Magnetization curves of CFO and AgI_0.3BiOI/CFO.

3.3. Mechanism study

The trapping experiments for reactive radicals were performed to clarify the functions of photogenerated e⁻–h⁺ pairs during Hg⁰ removal process. It should be noticed that prior to turning on the light source, 15 mL of isopropyl alcohol (IPA), 0.8 g of ethylene diamine tetraacetic acid disodium salt (EDTA–2Na) and 0.8 g of benzoquinone (BQ) were added along with 0.2 g AgI_0.3BiOI/CFO composite into the reaction liquid as the scavenger of ˙OH, h⁺ and ˙O₂⁻, respectively.²⁴ As shown in Fig. 11, the Hg⁰ removal efficiency of AgI_0.3BiOI/CFO decreased by about 7% in the presence of IPA when compared to that of without any scavenger, indicating that ˙OH radicals had little impact on Hg⁰ removal. In contrast, an obvious reduction and a drastic inhibition were observed after the addition of EDTA–2Na and BQ, respectively, implying that h⁺ and ˙O₂⁻ radicals were the primary reactive substances for Hg⁰ removal.

Fig. 11 Effect of different scavengers on Hg⁰ removal.

Based on above discussion, a schematic illustration of energy bands matching among AgI, BiOI and CoFe₂O₄ and possible ways of charges transfer were depicted in Fig. 12. Under visible light or FSL irradiation, the photoexcited e⁻–h⁺ pairs would be produced due to lower band gap energies of AgI (2.56 eV) and BiOI (1.76 eV). Since the CB edge of BiOI was more positive than that of AgI, the electrons on the CB of AgI would transfer to the CB of BiOI. Because the VB potential of AgI was more negative than that of BiOI correspondingly, the holes on the VB of BiOI would migrate to the VB of AgI.^22,47 For BiOI and AgI, the electrons on the CB potential of AgI can be trapped by adsorbed O₂ to generate ˙O₂⁻ since the CB potential of AgI was more negative than single electron reduction potential of O₂ (−0.046 eV vs. NHE).⁴⁷ The OH⁻ or H₂O can be oxidized by holes to yield ˙OH radicals since the VB potential of BiOI and AgI was more positive than ˙OH/OH⁻ (1.99 eV vs. NHE).^22,24 Moreover, since the CB potential (0.65 eV) and VB potential (1.98 eV) of CoFe₂O₄ was between −0.046 and 1.99 eV, the e⁻–h⁺ pairs did not have enough redox capacity to generate reactive species such as ˙OH and ˙O₂⁻ and were prone to recombine,³⁸ thus only 6% of Hg⁰ removal efficiency was observed for pure CFO. However, due to lower VB potential of CFO than that of AgI and BiOI, the holes on the VBs of AgI and BiOI would transfer to the VB of CFO,³³ thereby facilitating an effective separation of e⁻–h⁺ pairs and the presence of some amounts of reactive ˙OH. This might be the reason why binary BiOI/CFO and AgI/CFO composite achieved a higher Hg⁰ removal efficiency than single component. Taken together, the recombination of photogenerated e⁻–h⁺ pairs over AgI_0.3BiOI/CFO could be greatly suppressed in such a way and much more electrons and holes could react with O₂, OH⁻ to generate ˙O₂⁻ and ˙OH radicals. In view of photo-electrochemistry, the products such as ˙OH, ˙O₂⁻ and h⁺ have enough oxidation capability to oxidize Hg⁰. Among them, the ˙O₂⁻ and h⁺ radicals were confirmed to play crucial roles in Hg⁰ removal.

Fig. 12 Schematic mechanism of AgI_0.3BiOI/CFO for enhanced Hg⁰ removal.

3.4. Stability of photocatalyst

To understand the recyclability and durability of composite AgI_0.3BiOI/CFO, four successive runs of photocatalytic experiments were performed under FSL irradiation. As shown in Fig. 13, AgI_0.3BiOI/CFO exhibited a slight decline in Hg⁰ removal efficiency with the recycling tests increasing. About 87% of Hg⁰ removal efficiency was obtained after four successive runs, suggesting that AgI_0.3BiOI/CFO was stable during the photocatalytic oxidation process. Furthermore, the inset of Fig. 13 clearly demonstrated that AgI_0.3BiOI/CFO after four successive runs still had a good magnetic separation performance. To further investigate the structure change of as-prepared sample after Hg⁰ removal test, AgI_0.3BiOI/CFO after four successive tests were also characterized by XRD technique. As revealed in Fig. 14, the crystal structure of used AgI_0.3BiOI/CFO was almost identical to its fresh sample except two new peaks at 56.94° and 62.68° belonged to CoFe₂O₄,³⁷ indicating a relatively stable crystal structure of the composite in the photocatalytic process.

Fig. 13 Hg⁰ removal of consecutive experiments by AgI_0.3BiOI/CFO.

Fig. 14 XRD patterns of AgI_0.3BiOI/CFO before and after four successive runs.

3.5. Simultaneous removal of SO₂ and Hg⁰

Considering that the current mature SCR technology is effective to remove NO and has been widely adopted by power plant, the simultaneous removal of SO₂ and Hg⁰ over AgI–BiOI/CFO composite under alkaline condition was preliminary explored. As shown in Fig. 15, when 200 ppm of SO₂ was introduced into the reaction system, Hg⁰ removal efficiency has little inhibition and still maintained at 85% with the solution pH adjusted to 11 by adding NaOH solution. Also, nearly 98% of SO₂ removal efficiency was observed, suggesting that a certain degree of alkaline solution can not only restrain the inhibition of SO₂ on Hg⁰ removal but also can effectively remove SO₂ from flue gas.

Fig. 15 Simultaneous removal of Hg⁰ and SO₂ over AgI_0.3BiOI/CFO.

4. Conclusions

A series of magnetically separable ternary AgI–BiOI/CoFe₂O₄ photocatalysts with different AgI contents were successfully prepared by a solvothermal and sequent coprecipitation method. The as-prepared ternary AgI–BiOI/CoFe₂O₄ composites showed greatly enhanced Hg⁰ removal activity under FSL irradiation, especially AgI_0.3BiOI/CFO. AgI_0.3BiOI/CFO demonstrated much stable Hg⁰ oxidation ability and enough magnetic property to be reused in the photocatalytic process. The DRS and photocurrent analyses indicated that the enhanced Hg⁰ removal efficiency over AgI_0.3BiOI/CFO could be ascribed to the synergistic effect between AgI, BiOI and CoFe₂O₄ under FSL irradiation, which could lead to an efficient e⁻–h⁺ pairs separation and much more photogenerated reactive species. The trapping experiment result indicated the ˙O₂⁻ and h⁺ radicals were the primary reactive substances contributing to the oxidation of Hg⁰. This work can improve the insight into the wet photocatalytic oxidation process of Hg⁰ using magnetic photocatalyst under FSL irradiation and serve as a guide in using magnetic photocatalyst in the field of flue gas treatment.

Acknowledgements

This study was financially supported by the National Natural Science Foundation of China (No. 51676064, 51306046, and 51576086), the Fundamental Research Funds for the Universities of Henan Province (No. NSFRF140204) and the Young Core Instructor Project in the Higher Education Institutions of Henan Province (No. 2016GGJS-038).

References

1. A. P. Dastoor and Y. Larocque, Atmos. Environ., 2004, 38, 147–161.
2. E. G. Pacyna, J. M. Pacyna, K. Sundseth, J. Munthe, K. Kindbom, S. Wilson, F. Steenhuisen and P. Maxson, Atmos. Environ., 2010, 44, 2487–2499.
3. X. Zhang, B. X. Shen, S. W. Zhu, H. Xu and L. H. Tian, J. Hazard. Mater., 2016, 320, 556–563.
4. S. J. Zhao, Z. Qu, N. Q. Yan, Z. Li, W. F. Zhu, J. pan, J. F. Xu and M. D. Li, RSC Adv., 2015, 5, 30841–30850.
5. State Environmental Protection Administration of China (SEPA), Emission Standard of Air Pollution for Thermal Power Plants, GB 13223-2011, Beijing, 2011.
6. C. F. You and X. C. Xu, Energy, 2010, 35, 4467–4472.
7. K. C. Galbreath and C. J. Zygarlicke, Fuel Process. Technol., 2000, 65, 289–310.
8. R. K. Srivastava, N. Hutson, B. Martin, F. Princiotta and J. Staudt, Environ. Sci. Technol., 2006, 40, 1385–1393.
9. F. M. Zhan, C. T. Li, G. M. Zeng, S. S. Tao, Y. J. Xiao, X. Zhang, L. K. Zhao, J. Zhang and J. F. Ma, Chem. Eng. J., 2013, 232, 81–88.
10. Y. X. Liu, J. Zhang and J. F. Pan, Energy Fuels, 2014, 28, 2135–2143.
11. Y. Zhao, R. L. Hao, P. Zhang and S. H. Zhou, Fuel, 2014, 136, 113–121.
12. N. D. Hutson, R. Krzyzynska and R. K. Srivastava, Ind. Eng. Chem. Res., 2008, 47, 5825–5831.
13. H. Q. Wang, S. Y. Zhou, L. Xiao, Y. J. Wang, Y. Liu and Z. B. Wu, Catal. Today, 2011, 175, 202–208.
14. Y. Li and C.-Y. Wu, Environ. Sci. Technol., 2006, 40, 6444–6448.
15. J. Wu, C. E. Li, X. Y. Zhao, Q. Wu, X. M. Qi, X. T. Chen, T. Hu and Y. Cao, Appl. Catal., B, 2015, 176–177, 559–569.
16. Y. Yuan, J. Y. Zhang, H. L. Li, Y. Li, Y. C. Zhao and C. G. Zheng, Chem. Eng. J., 2012, 192, 21–28.
17. X. M. Qi, M. L. Gu, X. Y. Zhu, J. Wu, H. M. Long, K. He and Q. Wu, Chem. Eng. J., 2016, 285, 11–19.
18. G. P. Dai, J. G. Yu and G. Liu, J. Phys. Chem. C, 2012, 116, 15519–15524.
19. Y. P. Bi, S. X. Ouyang, N. Umezawa, J. Y. Cao and J. H. Ye, J. Am. Chem. Soc., 2011, 133, 6490–6492.
20. C. Hu, T. W. Peng, X. X. Hu, Y. L. Nie, X. F. Zhou, J. H. Qu and H. He, J. Am. Chem. Soc., 2010, 132, 857–862.
21. X. Zhang, Z. H. Ai, F. L. Jia and L. Z. Zhang, J. Phys. Chem. C, 2008, 112, 747–753.
22. H. F. Cheng, B. B. Huang, Y. Dai, X. Y. Qin and X. Y. Zhang, Langmuir, 2010, 26, 6618–6624.
23. M. Cui, J. X. Yu, H. J. Lin, Y. Wu, L. H. Zhao and Y. M. He, Appl. Surf. Sci., 2016, 387, 912–920.
24. A. C. Zhang, L. X. Zhang, X. Z. Chen, Q. F. Zhu, Z. C. Liu and J. Xiang, Appl. Surf. Sci., 2017, 392, 1107–1116.
25. G. C. Xi, B. Yue, J. Y. Cao and J. H. Ye, Chem.–Eur. J., 2011, 17, 5145–5154.
26. W. Wu, C. Z. Jiang and V. A. L. Roy, Nanoscale, 2015, 7, 38–58.
27. T. W. Ng, L. S. Zhang, J. S. Liu, G. H. Huang, W. Wang and P. K. Wong, Water Res., 2016, 90, 111–118.
28. Y. G. Xu, T. Zhou, S. Q. Huang, M. Xie, H. P. Li, H. Xu, J. X. Xia and H. M. Li, RSC Adv., 2015, 5, 41475–41483.
29. W. Q. Zhang, M. Wang, W. J. Zhao and B. Q. Wang, Dalton Trans., 2013, 42, 15464–15474.
30. A. Ren, C. B. Liu, Y. Z. Hong, W. D. Shi, S. Lin and P. Li, Chem. Eng. J., 2014, 258, 301–308.
31. T. P. Xie, L. J. Xu, C. L. Liu, J. Yang and M. Wang, Dalton Trans., 2014, 43, 2211–2220.
32. Y. S. Fu, H. Q. Chen, X. Q. Sun and X. Wang, Appl. Catal., B, 2012, 111–112, 280–287.
33. R. Jiang, H.-Y. Zhu, J.-B. Li, F.-Q. Fu, J. Yao, S.-T. Jiang and G.-M. Zeng, Appl. Surf. Sci., 2016, 364, 604–612.
34. S. Duangjam, K. Wetchakun, S. Phanichphant and N. Wetchakun, Mater. Lett., 2016, 181, 86–91.
35. P. Sathishkumar, N. Pugazhenthiran, R. V. Mangalaraja, A. M. Asiri and S. Anandan, J. Hazard. Mater., 2013, 252–253, 171–179.
36. S. Singh and N. Khare, Mater. Lett., 2015, 161, 64–67.
37. L. Q. Jing, Y. G. Xu, S. Q. Huang, M. Xie, M. Q. He, H. Xu, H. M. Li and Q. Zhang, Appl. Catal., B, 2016, 199, 11–22.
38. L. Kong, Z. Jiang, T. C. Xiao, L. F. Lu, M. O. Jones and P. P. Edwards, Chem. Commun., 2011, 47, 5512–5514.
39. A. R. Upreti, Y. Li, N. Khadgi, S. Naraginti and C. Zhang, RSC Adv., 2016, 6, 32761–32769.
40. H. G. Kim, P. H. Borse, J. S. Jang, E. D. Jeong, O. S. Jung, Y. J. Suhd and J. S. Lee, Chem. Commun., 2009, 39, 5889–5891.
41. J. C. Colmenares and R. Luque, Chem. Soc. Rev., 2014, 43, 765–778.
42. D. H. Xia, L. L. Hu, C. He, W. Q. Pan, T. S. Yang, Y. C. Yang and D. Shu, Chem. Eng. J., 2015, 279, 929–938.
43. B. Cai, M. G. Zhao, Y. Ma, Z. Z. Ye and J. Y. Huang, ACS Appl. Mater. Interfaces, 2015, 7, 1327–1333.
44. D. A. Reddy, J. Choi, S. Lee, R. Ma and T. K. Kim, RSC Adv., 2015, 5, 67394–67404.
45. J. L. Liang, C. Shan, X. Zhang and M. P. Tong, Chem. Eng. J., 2015, 279, 277–285.
46. H. Liu, W. R. Cao, Y. Su, Y. Wang and X. H. Wang, Appl. Catal., B, 2012, 111–112, 271–279.
47. J. Cao, Y. J. Zhao, H. L. Lin, B. Y. Xu and S. F. Chen, J. Solid State Chem., 2013, 206, 38–44.
48. J.-C. Wang, H.-C. Yao, Z.-Y. Fan, L. Zhang, J.-S. Wang, S.-Q. Zang and Z.-J. Li, ACS Appl. Mater. Interfaces, 2016, 8, 3765–3775.
49. T. Yan, H. Y. Liu, P. C. Gao, M. Sun, Q. Wei, W. G. Xu, X. D. Wang and B. Du, New J. Chem., 2015, 39, 3964–3972.
50. H. L. Wang, L. S. Zhang, Z. G. Chen, J. Q. Hu, S. J. Li, Z. H. Wang, J. S. Liu and X. C. Wang, Chem. Soc. Rev., 2014, 43, 5234–5244.
51. A. L. Linsebigler, G. Lu and J. T. Yates, Chem. Rev., 1995, 95, 735–758.
52. R. F. Dong, B. Z. Tian, C. Y. Zeng, T. Y. Li, T. T. Wang and J. L. Zhang, J. Phys. Chem. C, 2013, 117, 213–220.
53. F. Dong, Y. J. Sun, M. Fu, Z. B. Wu and S. C. Lee, J. Hazard. Mater., 2012, 219–220, 26–34.
54. Z. H. Ai, W. K. Ho, S. C. Lee and L. Z. Zhang, Environ. Sci. Technol., 2009, 43, 4143–4150.
55. M. Sun, D. Z. Li, Y. B. Chen, W. Chen, W. J. Li, Y. H. He and X. Z. Fu, J. Phys. Chem. C, 2009, 113, 13825–13831.

---