Achieving Large-Capability Adsorption of Hg⁰ in Wet Scrubbing by Defect-Rich Colloidal Copper Sulfides under High-SO₂ Atmosphere

Abstract

This paper reports on a novel method to remove Hg⁰ in the wet scrubbing process using defect-rich colloidal copper sulfides for reducing mercury emissions from non-ferrous smelting flue gas. Unexpectedly, it migrated the negative effect of SO₂ on mercury removal performance, while also enhancing Hg⁰ adsorption. Colloidal copper sulfides demonstrated the superior Hg⁰ adsorption rate of 306.9 μg·g⁻¹·min⁻¹ under 6% SO₂ + 6% O₂ atmosphere with a removal efficiency of 99.1%, and the highest-ever Hg⁰ adsorption capacity of 736.5 mg·g⁻¹, which was 277% higher than all other reported metal sulfides. The Cu and S sites transformation results reveal that SO₂ could transform the tri-coordinate S sites into S₂²⁻ on copper sulfides surfaces, while O₂ regenerated Cu²⁺ via the oxidation of Cu⁺. The S₂²⁻ and Cu²⁺ sites enhanced Hg⁰ oxidation, and the Hg²⁺ could strongly bind with tri-coordinate S sites. This study provides an effective strategy to achieve large-capability adsorption of Hg⁰ from non-ferrous smelting flue gas.

1. Introduction

Mercury has been a contaminant of global concern due to its toxicity, persistence, and bioaccumulation since the implementation of the Minamata Convention on August, 2017 [1]. Among anthropogenic sources, the non-ferrous smelting sector is a major contributor, accounting for 14.9% of all worldwide emissions, as reported in the Global Mercury Assessment Report 2018 [2,3]. Therefore, it is urgent to efficiently remove Hg⁰ from non-ferrous smelting flue gas to abate the significant challenge of global mercury pollution control.

There are various forms of mercury in non-ferrous smelting flue gas, including elemental mercury (Hg⁰), oxidized mercury (Hg²⁺), and particulate mercury (Hg^p) [4,5,6]. In the existing flue gas purification system, Hg^p would be captured by the electrostatic precipitator (ESP) [7,8,9]. Hg²⁺ has strong water solubility and can be removed by wet scrubber, whereas the capture of Hg⁰ is difficult due to its strong stability, high volatility, and low solubility [10]. Wet methods for Hg⁰ removal promote the conversion of Hg⁰ to Hg²⁺, such as the Boliden–Norzink and advanced oxidation process, thereby improving mercury removal efficiency [11,12]. Hence, oxidation is considered as a vital way for Hg⁰ removal in the wet scrubbing of non-ferrous smelting flue gas [13].

Oxidative demercurization remains a major bottleneck in the application of wet scrubbing from non-ferrous smelting flue gas [14,15]. There is a large amount of reductive SO₂ (>6% vol) in the flue gas, which is thousands of times higher than Hg⁰ concentration [16,17]. Selective oxidation of Hg⁰ poses a great challenge since the standard oxidation potential of Hg⁰ (0.85 V) is higher than that of SO₂ (0.17 V) [18,19]. Consequently, SO₂ is preferentially oxidized over Hg⁰, bringing about the depletion of oxidants and making it difficult to oxidize Hg⁰. Therefore, the efficient oxidation of Hg⁰ from high-SO₂ flue gas is key to overcoming the bottleneck of Hg⁰ removal in wet scrubbing.

Transition metal sulfides (TMSs) have become a hot material for Hg⁰ capture due to the strong affinity of sulfur sites with Hg⁰ [20,21,22]. Previous studies had shown that SO₂ could react with defective S²⁻ on TMSs to generate highly active S_n²⁻, which might promote Hg⁰ oxidation [23]. O₂ dissolves into aqueous phase and forms dissolved oxygen, which has a high oxidation activity due to hydrogen bonding and Van Der Waals force with water molecules, providing an oxidative environment for TMSs to capture Hg⁰ [24]. Therefore, if TMSs are added to the wet scrubber, it is expected to eliminate the negative impact of SO₂ on the removal of Hg⁰, which could enhance the ability of TMSs to oxidize mercury, thereby greatly improving the adsorption capacity of mercury. However, the mechanism of SO₂ and O₂ in wet scrubbing for mercury capture by TMSs has not been reported.

Based on the above considerations, this paper involves a novel method of using metal sulfides in the wet scrubbing process for large-capability adsorption of Hg⁰ from high-SO₂ flue gas. The optimal mercury adsorbent was prepared by sulfide precipitation and optimized systematically. The morphology and structure of the adsorbents were characterized. The excellent Hg⁰ removal performance of colloidal copper sulfides (c-CuS) was examined in the flue gas wet scrubber. The effects of SO₂ and O₂ on Hg⁰ removal by c-CuS were investigated, and the mechanism of the activation of oxidizing sites (S₂²⁻ and Cu²⁺) by SO₂ and O₂ was proposed after identifying the structural changes of c-CuS under different atmospheres.

2. Experimental Section

2.1. Chemicals and Reagents

The chemicals in analytical grade were used in this study, including copper chloride (CuCl₂), copper nitrate (CuNO₃·6H₂O), Copper sulfate (CuSO₄·5H₂O), sodium sulfide (NaS·9H₂O), sodium hydroxide (NaOH), sodium chloride (NaCl), and nitric acid (HNO₃). They were purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China) All the agents were directly used without any further purification. Ultrapure water was used in all experiments unless otherwise stated.

2.2. Synthesis of TMSs

TMSs suspensions were synthesized by the double-jet liquid-phase sulfidation precipitation method. Simply, 50 mL of a 10 mmol·L⁻¹ metal chloride solution and an equal volume of 10 mmol·L⁻¹ Na₂S·9H₂O solution were added simultaneously to the beaker at a speed of 1.2 mL min⁻¹. These solutions were mixed by stirring at a speed of 380 rpm for 0.5 h, and then TMSs suspensions were obtained. c-CuS was synthesized using a single-jet liquid-phase sulfidation method. Then, 50 mL of Na₂S·9H₂O was rapidly added into a CuCl₂ solution of 1.0 mmol·L⁻¹, and then they were mixed by stirring. The experimental investigation on raw material concentration and ratio was undertaken to judiciously optimize the synthesis conditions.

2.3. Sample Characterization

These suspensions and colloids were filtered with nanopore filter membranes. The filtered residues were washed three times with ultrapure water and then dehydrated at −89 °C in a vacuum freeze-dryer for 8 h. The obtained filtered residues of TMSs suspensions and c-CuS were analyzed for crystalline structure by X-ray diffraction (XRD, Empyrean 2, PANalytic, Malvern, UK) with Cu-Kα radiation. High-resolution transmission electron microscopy (HRTEM, Titan G260-300, FEI, Lausanne, Switzerland) was used to observe the morphology of c-CuS particles. The scanning electron microscope (SEM, JSM-6360LV, Jeol, Tokyo, Japan) of c-CuS was characterized at the accelerating voltage of 200 kV after splashing a thin Au layer. The structure information of samples under different atmospheres was characterized by X-ray photoelectron spectroscopy (XPS, Escalab 250 Xi, Thermo Fisher Scientific, Waltham, MA, USA). The ultimate vacuum degree of the sample analysis room was 5 × 10⁻⁷ Pa, and the Al Kα X-ray was used as the excitation source with a power of 16 mA × 12.5 kV. Correction was made using C ls = 284.6 eV as the internal standard for electron binding energy. The fingerprint and structural properties of c-CuS were analyzed by three-dimensional excitation–emission matrix (3D-EEM, F7000, Hitachi, Ibaraki, Japan) fluorescence spectroscopy. 3D-EEM spectra were generated by scanning in the range of 200 nm to 600 nm, while excitation and emission sampling wavelengths were 10 nm, and the scanning speed was 12,000 nm·min⁻¹. Finally, temperature-programmed desorption (TPD) was employed to identify the form of mercury species adsorbed on c-CuS. The TPD tests were carried out under pure nitrogen at the flow rate of 500 mL·min^–1 from 50 °C to 600 °C at a rate of 5 °C·min⁻¹.

2.4. Hg⁰ Removal Test

An experiment was conducted in a bubbling reactor to remove Hg⁰. Pure N₂ was used as the carrier gas for the Hg⁰ generator (VICI Metronics) with a flow rate of 200 mL·min⁻¹. The simulated flue gas (SFG) containing N₂, SO₂, and O₂ at a flow rate of 300 mL·min⁻¹ was mixed with the Hg⁰ carrier gas. The mixture gas was then blown into a bubbling reactor containing the scrubbing solution. Then, the SFG was treated with 5 mol·L⁻¹ NaOH and quartz wool before the mercury detection to reduce the influence of SO₂ and water vapor. The mercury analyzer (RA-915M, Lumex Zeeman) was used to monitor the outlet Hg⁰ concentration. All data were obtained by averaging three measurement results. Finally, the remaining Hg⁰ in the tail gas was absorbed by KMnO₄ solution and activated carbon. The operating conditions in the experiment are listed in Table S1. The Hg⁰ removal efficiency and adsorption capacity were calculated according to Equations (1) and (2), respectively.

$$\eta =\frac{C_{in}-C_{out}}{C_{in}}\times 100\%$$

(1)

$$Q_t=1/M\times {\int }_{t1}^{t2}(C_{inlet}-C_{outlet})∆t\times u$$

(2)

where $\eta$ (%) represents Hg⁰ removal efficiency, and C_in and C_out (μg·m⁻³) are the Hg⁰ concentration at the inlet and outlet of the experimental system, respectively. $Q_t$ (mg·g⁻¹) is the adsorption capacity of Hg⁰, M (mg) is the mass of adsorbent, t₁ and t₂ (min) are the start and end times, Δt (min) is the scrubbing time, and u is the gas flow rate (m³·min⁻¹).

2.5. DFT Calculation

Perdew–Burke-Ernzerhof (PBE0) functional was used to characterize the exchange-correlation effects in the density functional theory (DFT) by CP2K [25,26]. The CuS (110) system was simulated using an orthorhombic box with dimensions of 16.44 × 13.15 × 34.49 Å3, and the dipole correction technique was applied in the XY orientation. BROYDEN-MIXING was used to optimize the geometries, while the DZVP-MOLOPT-SR-GTH basis set was used to describe the basis. A plane wave cutoff of 400 Ry was set, and Goedecker–Teter–Hutter (GTH) pseudopotentials were used to represent all electrons [27,28]. In addition, the D3 dispersion correction by Grimme et al. was employed, and the adsorption energy (E_ad) was calculated using Equation (3), based on the optimized structure.

E_ad = E_(Hg+CuS) − E_Hg − E_CuS

(3)

where E_Hg is the state energy of a free Hg atom, E_CuS represents the total energy of the CuS of different configurations, and E_(Hg+CuS) is the total energy of the Hg atoms being adsorbed on CuS.

3. Results and Discussion

3.1. Selection and Optimization of Transition Metal Sulfides for Hg⁰ Removal

Hg⁰ removal performance was evaluated by adding TMSs suspensions into a bubble reactor as the scrubbing solution. Figure 1a shows that TMSs suspensions had a good ability to capture Hg⁰ in wet scrubbing. Especially, copper sulfides suspension was the preferred Hg⁰ removal scrubbing agent, with the highest efficiency of 87.6%. Furthermore, the effect of regulating feeding methods on the dispersibility of detergent was investigated to improve the Hg⁰ removal performance of copper sulfides. Figure 1b displays that c-CuS prepared by single-jet was homodispersed in solution, and the Tyndall effect was evident, while the precipitation of suspensions copper sulfides (s-CuS) was visible. The removal efficiency gradually decreased from 65.8% to 46.4% due to the poor contact between gas and s-CuS, while the mercury removal efficiency of c-CuS could be maintained at 97.5 ~ 98.9%. Therefore, the c-CuS synthesized by single-jet has better Hg⁰ capture performance.

Optimization of the raw material concentration and ratio to synthesize copper sulfides is crucial for enhancing Hg⁰ removal. As shown in Figure 1c and Figure S1, when the concentration of c-CuS was 1/32 mmol·L⁻¹, the efficiency of Hg⁰ removal was only 65% due to the insufficient content of active components in the solution. Increasing the concentration of CuCl₂ and Na₂S fourfold to synthesize c-CuS could immediately improve the removal efficiency of Hg⁰ to 93%. When the concentrations were optimized to 1/2 mmol·L⁻¹ and 2 mmol·L⁻¹, the efficiencies were higher than 96.3%. Unfortunately, a large amount of precipitation appeared in the solution of 2 mmol·L⁻¹ concentration, which was not conducive to sufficient contact with Hg⁰. Finally, the Cu/S ratio of copper sulfides was optimized, as shown in Figure 1d. Copper sulfides had the best removal efficiency for Hg⁰, which was achieved by reaching 98.1% when the Cu:S ratio was 1:1. However, as shown in Figure S2, further increasing the S²⁻ ratio caused possible instability of colloid, which would have a negative effect on the Hg⁰ removal. In summary, the c-CuS was prepared with the concentration of 1/2 mmol·L⁻¹ and the Cu:S raw material ratio of 1:1, which was the optimal mercury capture agent.

3.2. Structural Characterizations of c-CuS

Figure 2 shows the XRD patterns of the obtained c-CuS and s-CuS filter residues. The phase was confirmed to be Cu₄(S₂)₂(CuS)₂ (JCPDS card NO. 74-1234), with overlapping diffraction peaks that could be divided into lattice planes such as (101), (102), (103), (110), and (116) [29]. The diffraction peaks of s-CuS were stronger and sharper, indicating a better crystallinity. According to Equation (4), the crystallinity of s-CuS was calculated to be 70.9%. Surprisingly, the crystallinity of the c-CuS sample was only about 3.4% with severely broadened diffraction peaks. It indicated that c-CuS might contain abundant defects, which was beneficial for exposing more active adsorption sites to capture Hg⁰ [15,30].

$$X_c=\frac{I_c}{I_c+I_a}\times 100\%$$

(4)

where I_c is the crystal diffraction intensity and I_a is the amorphous scattering intensity.

Figure 3 shows the XPS full spectrum of c-CuS. The peak analysis result indicated that the Cu:S atomic ratio was 0.98:1 on the surface of c-CuS filter residue (the O, C peak belongs to the peak of carrier sheet). The mismatched stoichiometric ratio suggested that rich sulfur sites were exposed. Figure 4a shows that c-CuS was composed of near-spherical nanoparticles with the diameter of about 8.0 nm, similar to those of SEM in Figure S3. HRTEM images in Figure 4b display many orange-marked vacancies and amorphous regions in the lattice, indicating that rich defect that could provide more active sites for mercury capture. The measured lattice spacing of 1.89 nm was consistent with the (110) plane of c-CuS. The broad diffraction rings in the selected-area electron diffraction (SAED) shown in Figure 4c further confirmed the existence of amorphous regions in c-CuS [31]. Figure 4d–f shows the high-angle annular dark-field image (HADDF), Cu element, and S element scanning images of c-CuS, respectively. It confirms that Cu and S were evenly distributed. In summary, c-CuS with abundant defect sites has the potential for high-activity mercury capture.

3.3. Hg⁰ Removal Performance of c-CuS under SO₂ and O₂ Atmospheres

SO₂ is typically considered to be a negative effects gas that inhibits Hg⁰ removal in industrial flue gas [23,32]. SO₂ resistance has a crucial impact on the mercury capture performance of adsorbents. SO₂-contained simulated flue gas was inputted into a bubbling reactor with 10 mL of c-CuS to investigate the effect of SO₂ on the Hg⁰ removal performance of c-CuS. As shown in Figure 5a, the removal efficiency was 80.1% in the absence of SO₂. The efficiency significantly increased to 95.1 ~ 97.4% under 3 ~ 9% vol SO₂. Furthermore, the performance of c-CuS for Hg⁰ removal was compared with and without SO₂ pretreatment, as shown in Figure 5b. The outlet concentration of Hg⁰ was about 10 μg·m⁻³ after passing through c-CuS solution pretreated without SO₂, but it decreased immediately to ~0 μg·m⁻³ after turning on SO₂. When c-CuS solution was pretreated with SO₂, the outlet Hg⁰ concentration was about 3 μg·m⁻³ after opening gas pipeline of Hg⁰. Meanwhile, the outlet concentration exhibited almost no fluctuation after supplying SO₂ and Hg⁰ simultaneously. This might mean that SO₂ did not occupy the adsorption sites, and SO₂ could interact with c-CuS to enhance adsorption with Hg⁰. The 3D-EEM was used to compare the fluorescence fingerprint spectra of c-CuS under N₂ + 6% vol SO₂ atmospheres. As shown in Figure 5c,d, the region II area of the sample treated with SO₂ was smaller than the region I area of c-CuS under N₂ atmosphere, confirming that SO₂ leads to the formation of more vacancies. The formation of vacancies meant that low-active S²⁻ had been converted to more active S₂²⁻ [33,34]. The specific conversion process of activity sites will be further studied in the following.

In Figure S4, it could be observed that the capture of mercury was almost unaffected under an atmosphere of 3 ~ 9% vol O₂. This was attributed to the fact that c-CuS solution could dissolve only 0.076 mmol·L⁻¹ oxygen, which was much higher than the amount of mercury and much lower than the O₂ proportion in the flue gas. Therefore, we conducted further investigations to explore the influence of dissolved oxygen on Hg⁰ oxidation. As displayed in Figure 6a, the dissolved oxygen concentration in c-CuS solution increased from 0.6 mg∙L⁻¹ to 7.6 mg∙L⁻¹. The redox potential of the c-CuS solution gradually increased from 166 mV to 222 mV, which meant enhancing the oxidation ability of the scrubbing solution. It led to a significant increase in the removal efficiency of mercury from 45.0% to 95.3%. As shown in Figure 6b, the results displayed that the Cu LMM auger spectral peaks included Cu²⁺ of binding energy at 568.1 eV and Cu⁺ at 565.8 eV [35]. The Cu⁺ spectral peak of c-CuS in oxygen-free water was visible, while it was weakened for c-CuS in oxic water, conversely indicating the oxidation of Cu⁺ to Cu²⁺ by O₂. It could be inferred that dissolved oxygen in an aqueous solution could increase the Cu²⁺ content of c-CuS and strengthen the oxidation of Hg⁰.

The Hg⁰ adsorption capacity and rate of c-CuS were investigated and calculated by penetration experiments. The prepared 10 mL c-CuS (containing 0.48 mg CuS particles) was input with 6% vol SO₂ + 6% vol O₂ flue gas containing an inlet Hg⁰ concentration of 890.0 µg·m⁻³. As shown in Figure 7, the outlet concentration of Hg⁰ gradually increased from 303.2 µg·m⁻³ to 835.5 µg·m⁻³ after 2400 min of scrubbing experiment. The Hg⁰ adsorption capacity of c-CuS was calculated to be 736.5 mg·g⁻¹, and the average adsorption rate was 306.9 μg·g⁻¹·min⁻¹. Figure 8 and Table S2 demonstrate the comparison of the adsorption capacity and rate of c-CuS with previously reported metal sulfides. c-CuS has the highest adsorption capacity and rate among the reported sulfide adsorbents. Compared to all other adsorbents, the Hg⁰ adsorption capacity of c-CuS is exceptionally high, exceeding them by an impressive 277%, such as nano-CuS, Co₃S₄, S/FeS₂, ZnS, and so on. Additionally, the Hg⁰ adsorption rate of c-CuS is much higher than that of other mineral sulfides, which is due to the abundant adsorption sites under the positive effect of SO₂ and O₂ [7,14,36,37,38,39,40].

3.4. Mechanism for Hg⁰ Adsorption

As mentioned above, SO₂ and O₂ have positive effects on c-CuS for Hg⁰ capture. To determine the mechanism of Hg⁰ adsorption under SO₂ and O₂ atmospheres, as shown in Figure 9, the XPS spectra of spent c-CuS samples were analyzed under different atmospheres. Shown in Figure 9a–d are the S 2p peaks of samples under N₂, 6% vol SO₂, 6% vol SO₂ + 6% vol O₂, and 6% vol SO₂ + 6% vol O₂ + Hg⁰, respectively. The forms of S on the c-CuS surface are multiple, such as tri-coordinated S²⁻_(CN=3), tetra-coordinated S²⁻_(CN=4), and sulfur–sulfur-coordinated S₂²⁻ [23]. After peak fitting, the peaks at 161.1 eV and 162.2 eV were assigned to S²⁻_(CN=3), while the peaks at 161.9 eV and 163.0 eV were attributed to S²⁻_(CN=4). The peaks at 163.2 eV and 164.4 eV were attributed to S₂²⁻, while the peaks at 167.9 eV and 168.9 eV belonged to SO₄²⁻ [37,38,41].

Under N₂ atmosphere, the ratio of S and Cu total amount (S_t, Cu_t) on c-CuS was found to be 1.01:1, with S 2p consisting of 26.8% S²⁻_(CN=3), 39.0% S²⁻_(CN=4), 33.5% S₂²⁻, and 0.7% SO₄²⁻. Upon treatment with 6% vol SO₂, the S_t:Cu_t ratio increased to 1.79:1, with a decrease in S²⁻_(CN=3) content to 10.4%, an increase in S₂²⁻ to 49.5%, and a slight change in other forms of S, indicating the formation of new S₂²⁻ by the combination of S²⁻_(CN=3) with SO₂. Additionally, the Cu 2p peak in Figure 10 shifted towards higher binding energy, indicating the reduction of a part of Cu²⁺ to Cu⁺ by SO₂. Similarly, under 6% vol SO₂ + 6% vol O₂ conditions, S_t:Cu_t decreased with the oxidation of Cu⁺ into Cu²⁺ by O₂ and inhibition of S²⁻_(CN=4) combination with SO₂ adsorbed on the c-CuS surface. The S²⁻_(CN=3) content decreased with an increase in S²⁻_(CN=4), which led to less S₂²⁻. The further supply of Hg⁰ resulted in a decrease in the ratio of S₂²⁻ and S²⁻_(CN=4), but an increase in S²⁻_(CN=3), indicating the transformation of S₂²⁻ and S²⁻_(CN=4) into S²⁻_(CN=3) after capture of Hg⁰. The results of Cu 2p shifting towards higher binding energy confirmed that Cu²⁺ was also an active site for Hg⁰ oxidation. It suggests that S₂²⁻ and Cu²⁺ serve as oxidation sites, while S²⁻_(CN=3) is the binding site with Hg after adsorption. The TPD results in Figure 11 indicated that the decomposition temperature of Hg on the c-CuS was about 195 ~ 220 °C, which suggested the presence of black HgS [38].

The experimental results were inconclusive about the main binding site of the adsorbed mercury. Therefore, the E_ad at each site on the c-CuS (110) crystal was analyzed at the molecular level using DFT calculations. The result in Figure 12 indicated that the tri-coordinate sulfur sites had the highest adsorption energy of −125.23 kJ·mol⁻¹. It means that the tri-coordinate sulfur site might be the main binding site for mercury, which is consistent with XPS results.

In summary, the mechanism for Hg⁰ capture under SO₂ and O₂ atmospheres could be described by Equations (5) ~ (10). After the adsorbing of Hg⁰ on the c-CuS surface, Cu²⁺ and S₂²⁻ sites oxidized Hg⁰_ad to Hg²⁺ and binded with S²⁻_(CN=3) to form HgS. As described in Equations (9) and (10), SO₂ and O₂ respectively activate S²⁻_(CN=3) and Cu⁺ transformed into S₂²⁻ and Cu²⁺ as Hg⁰ oxidation sites.

Hg⁰ → Hg⁰_ad

(5)

Hg⁰_ad + 2Cu²⁺ − S²⁻_(CN=4) → Hg²⁺ + 2Cu⁺ − S²⁻_(CN=3)

(6)

Hg²⁺ + S²⁻_(CN=3) → HgS

(7)

Hg⁰_ad + S₂²⁻ → HgS + S²⁻_(CN=3)

(8)

4Cu⁺ + O₂ + 4H⁺ → 4Cu²⁺ + 2H₂O

(9)

3S²⁻_(CN=3) + HSO₃⁻+ 3H⁺ → 2S₂²⁻ + 2H₂O

(10)

4. Conclusions

In this paper, the optimal conditions for preparing c-CuS with a Cu–S ratio of 1:1 and a concentration of 0.5 mmol·L⁻¹ were synthesized by a double-jet liquid-phase sulfidation precipitation method. Defect-rich c-CuS with low crystallinity was used as a scrubbing agent to remove mercury from flue gas in wet scrubbing. It migrated the negative effect of SO₂ on mercury removal performance, while also enhancing mercury capture. The Hg⁰ removal efficiency of c-CuS was 99.1% under 6% vol SO₂ + 6% vol O₂ atmosphere. The Hg⁰ adsorption capacity of c-CuS reached 736.5 mg·g⁻¹, and the average adsorption rate was 306.9 μg·g⁻¹·min⁻¹, which was far better than other reported metal sulfides adsorbents. Based on structural characterization and DFT calculation, it was found that SO₂ and O₂ can enhance the formation of Cu²⁺ and S₂²⁻ sites from Cu⁺ and S²⁻_(CN=3) to promote the oxidation of Hg⁰ to Hg²⁺, and then Hg²⁺ could strongly bind with S²⁻_(CN=3). However, reusability of the material should be developed in future studies.