Effect of the properties of MnO_x/activated carbon and flue gas components on Hg⁰ removal at low temperature

Abstract

Manganese oxide loaded on activated carbon (Mn/AC) was synthesized using an impregnation method, and its capacity for Hg⁰ removal in simulated flue gas was investigated at 120 °C. The Hg⁰ removal performance was significantly enhanced by manganese oxide. The effects of the Mn loading on Hg⁰ removal were evaluated. X-ray diffraction (XRD) and Brunauer–Emmett–Teller (BET) were employed to characterize the samples. In addition, the effects of individual flue gas components, including SO₂, NO and HCl, on the Hg⁰ removal performance over the Mn/AC sorbent were investigated. The results indicated that 10% Mn/AC was the optimal sorbent under the simulated flue gas conditions. The XRD and BET analyses indicated that the Mn₃O₄ crystal particles and surface area were the primary factors that contributed to Hg⁰ removal. The competitive adsorption and formation of Mn(SO₄)_x were the primary reasons that SO₂ inhibits Hg⁰ removal. In a pure N₂ atmosphere, a low concentration of NO decreased the Hg⁰ removal due to consumption of active oxygen species. However, a high concentration of NO promoted Hg⁰ removal due to the formation of N-containing active species that were generated on the sample surface. The Mn/AC sample remained highly active toward Hg⁰ removal in the presence of HCl due to reaction with active chlorine species.

---

Introduction

In China, mercury originating from coal combustion has become a significant environmental issue that has attracted increasing attention.^1–3 Therefore, as the largest producer and consumer of coal, China has an obligation to reduce mercury emissions that are derived from coal combustion.^4,5

The mercury removal capacity of various pollution control technologies is substantially dependent on its speciation in the flue gases.^6,7 Typically, the mercury emitted during coal combustion exists primarily as elemental mercury (Hg⁰), oxidized mercury (Hg²⁺) and particle-bound mercury (Hg_P). Hg²⁺ is soluble in water and can be removed by wet flue gas desulfurization systems (WFGDs). Along with fly ash particles, Hg_P can be efficiently removed by an electrostatic precipitator (ESP) or fabric filter (FF). However, Hg⁰ is difficult to remove with current pollution control utilities due to its relatively low water solubility (6 × 10⁻⁵ g L⁻¹ at 25 °C) and high vapour pressure (2.46 × 10⁻¹ Pa at 25 °C).⁸ Therefore, the development of Hg⁰ removal technology is key to controlling mercury emissions.

Adsorption and oxidation methods have been developed for effectively controlling low-concentration Hg⁰.^7,9–13 The removal of Hg⁰ via physical and chemical adsorption using activated carbon has been extensively studied in recent years and is widely applicable for Hg⁰ emission control in certain coal-fired power plants, especially for activated carbons modified by Cl, Br and I.^14–16 In addition, some researchers have attempted to remove Hg⁰ using metal oxides.^17,18 Noble metal catalysts and transition metal oxides have also been investigated for heterogeneous Hg⁰ removal.^19,20 Noble metals, such as Ag, Au, Pd and Pt, can effectively control Hg⁰.^21,22 In comparison to noble metals, inexpensive transition metal oxides, such as V₂O₅, MnO_x, Co₃O₄, CeO₂, FeO_x and CuO, exhibit high removal activity for Hg⁰.^4,23–26 However, the activity over some of these catalysts largely depends on the HCl concentration due to the fewer active sites derived from the smaller surface area of the supports. Therefore, it was predicted that metal oxides loaded on a support with a larger surface area should exhibit high performance during Hg⁰ removal. In addition, metal oxides placed upstream of the electrostatic precipitators or fabric filters may result in deactivation due to exposure to a high concentration of fly ash.²⁷ To minimize the effect of fly ash, metal oxides should be placed downstream from the electrostatic precipitators or fabric filters where the surrounding temperature is relatively low (i.e., typically less than 160 °C).¹⁶ Therefore, materials with a high Hg⁰ removal performance at low temperatures would be more promising for industrial applications and of great interest to researchers.

As previously reported, activated carbon (AC) is a good support candidate at low temperature for Hg⁰ removal due to its large surface area and various active functional groups that are located on the edges of the graphene layers.^17,28 Manganese oxides play an important role in a large number of important adsorption and catalytic reactions.^29–32 In addition, manganese oxides exhibit excellent mercury removal performance in a low temperature range.^30,33,34 Therefore, manganese oxides that are loaded on activated carbon may act as a competitive adsorbent/catalyst for Hg⁰ removal at low temperatures. However, few studies on the use of manganese oxides loaded on activated carbon for use in Hg⁰ removal have been reported.

The Hg⁰ removal performance was strongly affected by the multicomponent flue gases, especially HCl, SO₂ and NO. HCl can efficiently oxidize Hg⁰. Qiao et al.³⁵ reported that the addition of 20 ppm of HCl greatly enhances the removal of mercury over MnO_x/alumina catalysts. Li et al.³⁶ demonstrated that HCl is the most effective flue gas component that affects Hg⁰ removal. 10 ppm of HCl with the aid of 4% O₂ resulted in 100% Hg⁰ oxidation efficiency over the MnO_x–CeO₂ catalysts. However, this removal performance heavily depends on the HCl content. The effect of SO₂ on Hg⁰ removal over manganese oxide catalysts is not conclusive. Xie et al.¹⁷ reported that SO₂ deactivated the ability of the catalyst to remove Hg⁰ in the absence of O₂. However, SO₂ was beneficial during mercury capture in the presence of O₂ over the MnO_x–CeO₂/activated coke adsorbent. Wang et al.³⁷ demonstrated that the Hg⁰ removal performance was independent of the SO₂ concentration over the CuO–MnO₂–Fe₂O₃ catalysts. The influence of NO on Hg⁰ removal also varies over different single manganese oxides or binary manganese oxides catalysts.^38,39 In addition, the reaction mechanisms over Mn-based material have rarely been systematically studied.

Based on the above discussion, this study consisted of three parts. The first part focused on evaluating the effect of various metal oxides on Hg⁰ removal to determine the best active component to load on the activated carbon. The second part involved determining the optimal loadings of the active component as well as characterizing the samples to gain insight into the observed effect. The third part evaluated the performance of the optimal sample in the presence of common individual acidic flue gas components, and possible reaction mechanisms were revealed based on these experimental results. An improved understanding of the preparation parameters and the role of the flue gas components on Hg⁰ removal will enable us to optimize the operating conditions and lower the application cost.

Experimental section

Material and characterization

The ACs used in this experiment are commercial carbon obtained from the Chengde Jibei Yanshan Activated Carbon Plant in Hebei Province. Prior to their use, the ACs were crushed and sieved to 40–80 mesh (0.45–0.2 mm) followed by washing with distilled water to remove dust and drying for 24 h at 120 °C. The Mn/AC samples were prepared using an incipient-wetness impregnation method with manganese nitrate as the precursor. The dried AC sorbents were impregnated with the required amount of manganese nitrate, and then, the mixtures were placed in an ultrasonic cleaner for 1 h. Next, the samples were dried at 120 °C for 12 h followed by calcination at 300 °C for 3 h in N₂ and 200 °C for 1 h in O₂. Finally, the samples were cooled to room temperature and stored in a desiccator prior to use. The samples are referred to as x% MO_x/AC, where x represents the weight ratio of M to AC.

Nitrogen adsorption was performed at −196 °C using an automatic surface and porosity analyser (AutosorbiQ, Quantachrome). The surface area was calculated from the nitrogen adsorption isotherms using the BET equation. The total pore volume was determined based on the amount of nitrogen gas adsorbed at a relative pressure (p/p₀ = 0.99) to the liquid adsorbate volume. The micropore volume was calculated using the Horvath–Kawazoe (HK) method. The powder X-ray diffraction (XRD) patterns of the samples were performed on a powder diffractometer (Rigaku D/Max-RA) using Cu Kα radiation (40 kV and 150 mA).

Experimental device and removal capacity measurement

The removal performance evaluation of the samples had been described in detail elsewhere.⁴⁰ It consists of a fixed-bed reactor, a simulated flue gas feed system, a Hg⁰ vapour-generating unit and a continuous online mercury analyser. The evaluation tests were carried out in a fixed-bed quartz flow reactor (4 mm i.d.) that contained a 50 mg sample. The feed gases, which consisted of O₂, SO₂, NO and HCl balanced with N₂, were adjusted using mass flow controllers and then introduced into the reactor. The gas concentrations were monitored using an online Nicolet 6700-FTIR spectrophotometer equipped with a 2 dm³ gas cell. A constant quantity of Hg⁰ vapour was generated from a Hg⁰ permeation tube, which was sealed in a U-shaped glass tube that was immersed in a water bath maintained at 60 °C. The Hg⁰ concentrations were monitored continuously using online mercury analysers (RA-915M, Lumex), and the initial Hg⁰ concentration was maintained at 136.0 ± 4.0 μg m⁻³ with a total flow rate of 600 mL min⁻¹ balanced with N₂. All of the pipelines that Hg⁰ passed through were heated to approximately 90 °C to prevent mercury deposition on the internal surface of the pipelines. The amount of Hg⁰ adsorbed on the pipeline was measured prior to each test. The results indicated that the Hg⁰ adsorption concentrations on the pipelines were less than 0.4 μg m⁻³. Prior to each Hg⁰ removal test, the sample was purged with N₂ for half an hour to identical conditions. The Hg⁰ removal capacity of the breakthrough curves can be expressed as follows:where C(Hg⁰_inlet) and C(Hg⁰_outlet) are the Hg⁰ concentrations at the inlet and outlet of the fixed-bed quartz reactor, respectively.

Results and discussion

Effect of the different metal oxides on Hg⁰ removal

The various outer electron distributions of metal oxides would result in diversified chemical properties, which would further influence the removal capacity. To determine the optimal metal oxide additives loaded on activated carbon for Hg⁰ removal, a series of experiments were conducted. The concentrations of all of the metal contents used in the study were 5%. As shown in Fig. 1, a significant difference in the removal performance was observed among the sorbents. The Hg⁰ removal performance of the MnO_x-, CoO_x-, CeO_x-, FeO_x- and CuO_x-loaded ACs were ordered Mn > Co > Ce > Fe > Cu, indicating that the 5% Mn/AC exhibited the highest Hg⁰ removal performance. The high removal performance for Hg⁰ over the 5% Mn/AC may be due to the outer shell of the Mn element being unfilled, which results in a more effective nuclear charge. This type of orbital structure facilitates the generation of coordination compounds between AC and the Mn ion, which may favour coordination and other surface reactions that enhance the Hg⁰ removal performance.⁴¹ In addition, the outer electron distribution of MnO_x favours reactions with Hg⁰ to form polymer species, which may be another important pathway for Hg⁰ removal. Based on these results, MnO_x was selected as the optimal active component for use in the metal oxide-loaded activated carbon.

Fig. 1 Effect of different active components loaded on AC on Hg⁰ removal at 120 °C. Reaction conditions: 8% O₂ balanced with N₂.

Effect of the Mn content on Hg⁰ removal

As an important parameter of material, the active component loading has a great effect on the dispersion and aggregation of the metal oxide that is loaded on the carrier, further affecting its removal capacity. Therefore, the effect of the Mn loadings on Hg⁰ removal was investigated to determine the optimal Mn contents. As shown in Fig. 2, no Hg⁰ removal was observed over the virgin AC. The addition of 1% Mn significantly enhanced the removal performance, indicating that AC and MnO_x played a synergetic role in Hg⁰ removal. For the Mn/AC samples with different loadings, the mercury removal performance continued to improve as the Mn loading increased from 5% to 10% but the increasement grew dramatically slow compared to that observed for a Mn content increase from 1% and 3%. For comparison, an additional loading of 20% was adopted to determine the best Mn content for Hg⁰ removal. As shown in Fig. 2, further increasing of the Mn loading to 20% weakened the Hg⁰ removal performance over the first 18 h. Based on both the cost and removal capacity, 10% was determined to be the optimal Mn content for Hg⁰ removal over AC under the experimental conditions, and subsequent studies using Mn/AC employed a Mn loading of 10%.

Fig. 2 Hg⁰ breakthrough curves over virgin AC and Mn/AC samples with different Mn contents at 120 °C. Reaction conditions: 8% O₂ balanced with N₂.

Characterization of the samples

X-ray diffraction analysis. XRD and BET techniques were used to reveal the reasons for the effect of the Mn content on Hg⁰ removal. To identify the structure of the manganese species on the samples, XRD measurements of virgin AC and Mn/AC were conducted, and the results are shown in Fig. 3. For virgin AC, two strong diffraction peaks were detected at 2θ = 26.66° and 44.58°. However, the intensity of the two peaks decreased as the Mn oxide loading increased. Moreover, when the loading reached 20%, the peak due to AC disappeared, indicating that strong interactions existed between MnO_x and AC in the samples. For the Mn/AC samples with a low Mn content (<20%), the typical Mn₃O₄ diffraction peaks were observed in the XRD patterns, and the intensity of the Mn₃O₄ diffraction peaks changed slightly as the Mn content increased from 5% to 10%, which is consistent with the tendency of Hg⁰ removal performance. Because the intensity of the diffraction peaks is positively correlated with the crystal contents, it can be concluded that Mn₃O₄ crystal particles is beneficial to Hg⁰ removal. In contrast, when the loading was 20%, the Mn₃O₄ diffraction peak intensity was the largest, and peaks corresponding to Mn₂O₃ appeared. In combination with the results from the breakthrough curves for Hg⁰ removal that are shown in Fig. 2, these results indicate that the Hg⁰ removal capacity is not exclusively determined by the Mn₃O₄ crystal particles.

Fig. 3 X-ray diffraction patterns of Mn/AC samples with different Mn contents.

BET analysis. The samples with different loadings were further analysed using N₂ physisorption to clarify the difference in the Hg⁰ removal performance. The physical properties of the Mn/AC samples including the BET surface area, total pore volume and micropore volume are summarized in Table 1. It could be observed that the virgin AC showed the highest BET surface area of 918.58 m² g⁻¹ and micropore volume of 0.380 cm³ g⁻¹, respectively. However, these values decreased with the increase of the Mn loading. This change may be due to the Mn/AC pores being blocked as excessive Mn accumulates on the AC surfaces, which is in agreement with previous results.^4,42 It is important to note that when the Mn loading reached 20%, the BET surface area and total pore volume decreased sharply, which is consistent with the Hg⁰ removal performance. This result indicates that the surface area of the activated carbons might be another factor affecting the Hg⁰ removal performance, which is accordance with the previous results^43–45 that the elemental mercury removal efficiency is related to both the manganese loading and BET surface area. However, in their experiments, they found that amorphous or poorly crystalline states of manganese oxides are responsible for the elemental mercury, which is counter to the result reported here. The role of the support is to provide high surface area to maximize the number of collisions between elemental mercury and the sorbent.⁴³ In addition, the crystal nucleation and growth habit of manganese oxide during the process of calcination is related to elemental mercury removal. Therefore, the above mentioned factors probably results in the differences.

Table 1 Physical characteristics of Mn/AC samples

Samples	BET surface areas (m^2 g^−1)	Total volume (cm^3 g^−1)	Micropore volume (cm^3 g^−1)
AC	919	0.412	0.380
1% Mn/AC	916	0.415	0.362
3% Mn/AC	911	0.417	0.354
5% Mn/AC	880	0.397	0.345
7% Mn/AC	850	0.391	0.332
10% Mn/AC	792	0.362	0.310
20% Mn/AC	632	0.295	0.250

---

Effect of the gaseous components of flue gas on Hg⁰ removal

Effect of SO₂. The effect of SO₂ on Hg⁰ removal was studied by varying the concentration over the 10% Mn/AC samples at 120 °C. As shown in Fig. 4(a), adding 200 ppm of SO₂ dramatically decreased the Hg⁰ removal performance compared to that in a N₂ atmosphere. An increase in the SO₂ concentration from 200 ppm to 1000 ppm further decreased the Hg⁰ removal performance. The results indicate that SO₂ has a detrimental effect on Hg⁰ removal in a pure N₂ atmosphere. As shown in Fig. 4(b), a similar phenomenon was observed for the Hg⁰ removal performance under an O₂ atmosphere, suggesting that SO₂ also inhibits Hg⁰ removal under an O₂ atmosphere. However, the scope of the deterioration effect was much less significant compared to that in a N₂ atmosphere. The following three reasons may explain the inhibition phenomena: (1) SO₂ consumed the reactive oxygen, (2) SO₂ reacted with the manganese oxide to form manganese sulfate and (3) competitive adsorption between Hg⁰ and SO₂. If the reaction between Hg⁰ and SO₂ over the 10% Mn/AC sorbents followed the Mars–Maessen mechanism, SO₂ would compete with Hg⁰ for the same lattice oxygen or surface oxygen active sites due to the preferential adsorption of SO₂ on the samples. However, the O₂ volume concentration under the experimental conditions was more than tens of times higher than that of SO₂ and ten thousand times higher than that of Hg⁰. Therefore, the first reason was unlikely to be responsible for the inhibitive effect of SO₂ on Hg⁰.

Fig. 4 Effect of SO₂ concentration on Hg⁰ removal (a) in the absence and (b) presence of O₂ over the 10% Mn/AC at 120 °C. Reaction conditions: 200 and 1000 ppm of SO₂ and 8% O₂ balanced with N₂.

To further understand the mechanisms responsible for the prohibitive effect of SO₂ on Hg⁰ removal, the results from breakthrough experiments of Hg⁰ over 10% Mn/AC in the absence or presence of 1000 ppm SO₂ under different conditions were compared. As shown in Fig. 5, under the four different flue gases, the Hg⁰ removal performance follows the order SO₂ < N₂ < SO₂ + O₂ < O₂. The Hg⁰ removal performance under SO₂ and O₂ was superior to that in a SO₂ atmosphere, indicating that the inhibitive effect of SO₂ is weaken due to the presence of O₂. In addition, the Hg⁰ removal performance under SO₂ and O₂ was much higher than that observed under pure N₂, which when combined with the results shown in Fig. 4(b), indicates that the newly formed species were not responsible for Hg⁰ removal, but the enhancement of O₂ is superior to the inhibition of SO₂, promoting Hg⁰ removal in the presence of O₂. The presence of O₂ is unlikely to enhance the physisorption capacity of Hg⁰. Previous studies^41,46 have demonstrated that when exposed to manganese oxide, SO₂ can react with surface OH groups to generate manganese sulfate that results in damage to the cation vacancies over the Mn/AC sorbent. It is important to note that the O₂ gas replenishes the lattice oxygen over the Mn/AC sorbent surface, which weakens the inhibition effect of SO₂ on Hg⁰. Nevertheless, due to the higher affinity of SO₂ with the Mn/AC sorbent, even at an O₂ concentration that is much higher than that of SO₂, the damage to the cation vacancies that is caused by SO₂ cannot be completely offset over the Mn/AC sorbent. For the third reason, if only adsorbed Hg⁰ can react with the active surface, it is very likely that SO₂ competes with Hg⁰ for adsorptive sites, limiting the Hg⁰ removal performance.

Fig. 5 Breakthrough curves of Hg⁰ over 10% Mn/AC in the absence/presence of SO₂ under different conditions at 120 °C. Reaction conditions: 1000 ppm of SO₂ and 8% O₂ balanced with N₂.

Effect of NO. To study the effect of NO concentration on the Hg⁰ removal, 10% Mn/AC was exposed to N₂ with the varied concentrations of NO. As shown in Fig. 6, the Hg⁰ removal performance of 10% Mn/AC highly depends on the NO concentration. The addition of a low NO concentration of 100 ppm inhibited the uptake of mercury. When the NO concentration was increased to 200 ppm, the initial Hg⁰ breakthrough ratio decreased almost to zero, and subsequently increased to a stable value below 0.3. Later, the Hg⁰ breakthrough ratio decreased gradually as the reaction progressed. However, a further increase in the NO concentration to 300 ppm enhanced the removal of elemental mercury. In general, a low NO concentration on Hg⁰ removal over the Mn/AC sorbent exhibited an inhibitory effect in a pure N₂ atmosphere, while that of high NO concentration enhanced the removal. According to the in situ FTIR characterization results, in the absence of O₂, NO is weakly adsorbed on the metal oxide surface, and as the gas oxygen content increases, the adsorbed NO can be oxidized to NO₂, NO⁺ and NO₃⁻ on the surface of the metal oxide,^47,48 and these species are oxidizing species and may be responsible for Hg⁰ removal. Therefore, NO may initially cover the active sites for Hg⁰ adsorption over the Mn/AC sorbent due to the much higher concentration compare to that of Hg⁰. Afterward, NO reacted with active oxygen to generate limited oxidizing species, which are insufficient to oxidize all of the Hg⁰, thus resulting in a decrease in Hg⁰ removal. As the added NO concentration increases, the amount of active species that accumulated over the Mn/AC sorbent surface gradually increased, hence, promoting the Hg⁰ removal. The results also suggest that the NO concentration boundary for effective Hg⁰ removal over the 10% Mn/AC sorbent should be located between 200 ppm and 300 ppm under current experimental conditions.

Fig. 6 Effect of NO concentration in the absence of O₂ on the Hg⁰ removal over 10% Mn/AC at 120 °C. Reaction conditions: 100, 200 and 300 ppm of NO balanced with N₂.

As demonstrated above, adequate active oxidizing species are favourable for Hg⁰ removal. Therefore, the effect of the addition of O₂ to the 100 ppm of NO balanced with N₂ on the removal of Hg⁰ was investigated to further identify the mechanism. As shown in Fig. 7, the coexistence of O₂ and NO resulted in a rapid decrease in the Hg⁰ breakthrough ratio to nearly zero when the Hg⁰-laden feed gas flow was switched to the fixed bed reactor, and almost no change in the Hg⁰ breakthrough ratio was observed for the subsequent 20 h. The removal of 100% of the Hg⁰ indicated that in comparison to the individual NO atmosphere, the great enhancement in the Hg⁰ removal performance due to the coexistence of O₂ and NO. In addition, the Hg⁰ removal performance under O₂ and NO is superior to that under an O₂ atmosphere. Therefore, it can be concluded that it is the products formed by the presence of O₂ and NO over the manganese oxide facilitate the Hg⁰ removal, which is consistent with our hypothesis.

Fig. 7 Breakthrough curves of Hg⁰ over 10% Mn/AC in the absence/presence of NO under different conditions at 120 °C. Reaction conditions: 100 ppm of NO and 8% O₂ balanced with N₂.

Effect of HCl. HCl is considered to be the most important flue gas component that exerts the largest effect on Hg⁰ removal. Therefore, the effect of HCl on Hg⁰ removal was investigated. Fig. 8 shows the effect of HCl on the Hg⁰ removal over the Mn/AC sorbent at 120 °C. Regardless of the presence or absence of O₂, when 2 ppm of HCl was added, the outlet Hg⁰ concentration quickly decreased to zero and remained at that level beyond 20 h. This result indicates that HCl dramatically enhances the Hg⁰ removal. Additionally, the Hg⁰ removal performance under HCl is better than that with O₂ over the Mn/AC sorbent even though its concentration is much lower than that of O₂. Previous studies have demonstrated that HCl can generate a sufficient amount of active chlorine species over metal oxides, therefore greatly enhancing the Hg⁰ removal.^4,40,49

Fig. 8 Breakthrough curves of Hg⁰ over 10% Mn/AC in the absence/presence of HCl under different conditions at 120 °C. Reaction conditions: 2 ppm of HCl and 8% O₂ balanced with N₂.

Discussion on the mechanism for Hg⁰ removal

The flue gas components affect the type of mercury products that form on the Mn/AC sorbent, thus and provide insight into the potential application of the sorbent. Based on the previously mentioned experiment results, the Hg⁰ removal mechanism over the Mn/AC sorbent in the presence of various flue gases can be explained as follows (shown in Fig. 9):

Fig. 9 Proposed removal mechanisms toward Hg⁰ with the simulated multicomponent gas over Mn/AC.

In the first step, the gaseous Hg⁰ collided with the samples and was adsorbed on the sample surface via physical adsorption to form an adsorbed Hg⁰ state (ads).

Hg⁰ + surface → Hg⁰ (ads) (2)

Subsequently, a more stable Hg state was formed by the reaction between the adsorbed Hg⁰ and lattice oxygen or surface oxygen following the Mars–Maessen mechanism, which was called as HgO (ads).

Hg⁰ (ads) + O* → HgO (ads) (3)

The addition of SO₂ to the inlet gas causes a quick spike in the mercury concentration in the effluent gas. The physically adsorbed SO₂ competes with Hg⁰ (ads) for the same sites, and the formation of the MnSO₄,⁴¹ intermediate product leads to deactivation during Hg⁰ removal. The presence of O₂ compensates for the damage to the cation vacancies that was caused by SO₂, thus promoting Hg⁰ removal to some extent.

HgO (ads) + SO₂ (g) → Hg⁰ (g) + SO₃ (ads)/SO₄²⁻ (4)

In the presence of NO, the Hg⁰ removal was limited by the low NO concentration due to the consumption of active oxygen species and promoted by a high NO concentration due to the NO₃⁻ or other active species (NO_x) that formed. In contrast, with excess O₂, the Hg⁰ removal performance increased even at low NO concentrations. The mechanism for the formation of Hg(NO₃)₂ over the Mn/AC sorbent involves a multistep reaction process.

NO + O* → NO₂ (ads) (5)

2NO₂ (ads) + O* + HgO (ads) → Hg(NO₃)₂ (ads) (6)

In the presence of HCl, the mechanism for the formation of HgCl_x over the Mn/AC sorbent involves active chlorine oxidizing Hg⁰.

Cl + * → Cl* (ads) (7)

2HgO (ads) + 2Cl* (ads) + 4H → Hg₂Cl₂ (ads) + 2H₂O (8)

Hg₂Cl₂ (ads) + 2Cl* → 2HgCl₂ (ads) (9)

Conclusions

The Hg⁰ removal performance of the Mn/AC sorbent was studied in a lab-scale fixed-bed system. The results indicated that the Hg⁰ removal performance was higher than that of AC under the simulated flue gas conditions. The 10% Mn/AC sample exhibited the highest Hg⁰ removal ability. Based on the XRD and BET results, the Mn₃O₄ crystal particles and surface area were responsible for the Hg⁰ removal. The inhibitory effect of SO₂ on Hg⁰ removal was attributed to the formation of Mn(SO₄)_x as well as competitive adsorption between SO₂ and Hg⁰. NO was found to have contrary effect on the Hg⁰ removal in a pure N₂ atmosphere. The Hg⁰ removal was limited by a low concentration of NO and enhanced by a high concentration of NO. The formation of N-containing active species that were generated on the sample surface facilitated the removal of Hg⁰ in the presence of NO and O₂ to produce Hg(NO₃)₂. In addition, the presence of HCl can effectively oxidize Hg⁰ to produce HgCl₂.

This research provides insight into optimizing the preparation and operating conditions for Hg⁰ removal, which is crucial for potential industrial applications of the sample. Additionally, this knowledge regarding the impact of flue gas on Hg⁰ removal will provide a theoretical foundation for choosing layout sites, so as to improve the overall performance of the Mn/AC sorbent.

Acknowledgements

This work was supported by the National Basic Research Program (973) of China (No. 2013CB430005), the National Hi-Tech Research and Development Program (863) of China (No. 2013AA065501) and the Special Research Funding for Public Benefit Industries from National Ministry of Environmental Protection (No. 201309018).

References

1. X. Fan, C. Li, G. Zeng, X. Zhang, S. Tao, P. Lu, Y. Tan and D. Luo, Energy Fuels, 2012, 26, 2082–2089.
2. L. H. Tian, C. T. Li, Q. Li, G. M. Zeng, Z. Gao, S. H. Li and X. P. Fan, Fuel, 2009, 88, 1687–1691.
3. H. Shen, I. R. Ie, C. S. Yuan, C. H. Hung, W. H. Chen, J. Luo and Y. H. Jen, J. Hazard. Mater., 2015, 289, 235–243.
4. W. Q. Xu, H. R. Wang, X. Zhou and T. Y. Zhu, Chem. Eng. J., 2013, 243, 380–385.
5. H. Yang, Z. Xu, M. Fan, A. E. Bland and O. R. Judkins, J. Hazard. Mater., 2007, 146, 1–11.
6. S. S. Tao, C. T. Li, X. P. Fan, G. M. Zeng, P. Lu, X. Zhang, Q. B. Wen, W. W. Zhao, D. Q. Luo and C. Z. Fan, Chem. Eng. J., 2012, 210, 547–556.
7. Z. Q. S. J. Zhao, N. Q. Yan, Z. Li, W. F. Zhu, J. Pan, J. F. Xu and M. D. Li, RSC Adv., 2015, 5, 30841–30850.
8. K. C. Galbreath and C. J. Zygarlicke, Environ. Sci. Technol., 1996, 30, 2421–2426.
9. Y. Liu, C. Tian, B. Yan, Q. Lu, Y. Xie, J. Chen, R. Gupta, Z. Xu, S. M. Kuznicki, Q. Liu and H. Zeng, RSC Adv., 2015, 5, 15634–15640.
10. Y. Shi, S. Deng, H. Wang, J. Huang, Y. Li, F. Zhang and X. Shu, RSC Adv., 2016, 6, 15999–16009.
11. X. Q. Wang, P. Wang, P. Ning, Y. X. Ma, F. Wang, X. L. Guo and Y. Lan, RSC Adv., 2015, 5, 24899–24907.
12. H. Xu, Y. Ma, S. Zhao, W. Huang, Z. Qu and N. Yan, RSC Adv., 2016, 6, 63559–63567.
13. D. Jampaiah, K. M. Tur, P. Venkataswamy, S. J. Ippolito, Y. M. Sabri, J. Tardio, S. K. Bhargava and B. M. Reddy, RSC Adv., 2015, 5, 30331–30341.
14. C. X. Hu, J. S. Zhou, S. He, Z. Y. Luo and K. F. Cen, Fuel Process. Technol., 2009, 90, 812–817.
15. S. Niksa, C. V. Naik, M. S. Berry and L. Monroe, Fuel Process. Technol., 2009, 90, 1372–1377.
16. Z. Q. Tan, J. Xiang, S. Su, H. C. Zeng, C. S. Zhou, L. S. Sun, S. Hu and J. R. Qiu, J. Hazard. Mater., 2012, 239, 160–166.
17. Y. Xie, C. Li, L. Zhao, J. Zhang, G. Zeng, X. Zhang, W. Zhang and S. Tao, Appl. Surf. Sci., 2015, 333, 59–67.
18. Z. Mei, Z. Shen, Q. Zhao, W. Wang and Y. Zhang, J. Hazard. Mater., 2008, 152, 721–729.
19. B. M. Reddy, N. Durgasri, T. V. Kumar and S. K. Bhargava, Catal. Rev., 2012, 54, 344–398.
20. B. A. Dranga, L. Lazar and H. Koeser, Catalysts, 2012, 2, 139–170.
21. D. Li, J. Han, L. Han, J. Wang and L. Chang, J. Environ. Sci., 2014, 26, 1497–1504.
22. J. A. Hrdlicka, W. S. Seames, M. D. Mann, D. S. Muggli and C. A. Horabik, Environ. Sci. Technol., 2008, 42, 6677–6682.
23. J. Wang, J. Yang and Z. Liu, Fuel Process. Technol., 2010, 91, 676–680.
24. H. Li, C.-Y. Wu, Y. Li, L. Li, Y. Zhao and J. Zhang, Chem. Eng. J., 2013, 219, 319–326.
25. S. J. Yang, Y. F. Guo, N. Q. Yan, D. Q. Wu, H. P. He, Z. Qu and J. P. Jia, Ind. Eng. Chem. Res., 2011, 50, 9650–9656.
26. K. Li, X. Tang, H. Yi, P. Ning, Y. Xiang, C. Wang and X. Xu, Res. Chem. Intermed., 2012, 40, 169–177.
27. Y. X. Zhao, M. D. Mann, J. H. Pavlish, B. A. F. Mibeck, G. E. Dunham and E. S. Olson, Environ. Sci. Technol., 2006, 40, 1603–1608.
28. Z. Tan, L. Sun, J. Xiang, H. Zeng, Z. Liu, S. Hu and J. Qiu, Carbon, 2012, 50, 362–371.
29. H. Li, C.-Y. Wu, Y. Li and J. Zhang, Appl. Catal., B, 2012, 111–112, 381–388.
30. S. J. Yang, Y. F. Guo, N. Q. Yan, D. Q. Wu, H. P. He, J. K. Xie, Z. Qu and J. P. Jia, Appl. Catal., B, 2011, 101, 698–708.
31. C. He, B. Shen, J. Chen and J. Cai, Environ. Sci. Technol., 2014, 48, 7891–7898.
32. H. Xu, Z. Qu, C. Zong, W. Huang, F. Quan and N. Yan, Environ. Sci. Technol., 2015, 49, 6823–6830.
33. J. F. Li, N. Q. Yan, Z. Qu, S. H. Qiao, S. J. Yang, Y. F. Guo, P. Liu and J. P. Jia, Environ. Sci. Technol., 2010, 44, 426–431.
34. S. J. Yang, Y. F. Guo, N. Q. Yan, Z. Qu, J. K. Xie, C. Yang and J. P. Jia, J. Hazard. Mater., 2011, 186, 508–515.
35. S. H. Qiao, J. Chen, J. F. Li, Z. Qu, P. Liu, N. Q. Yan and J. P. Jia, Ind. Eng. Chem. Res., 2009, 48, 3317–3322.
36. H. Li, C. Y. Wu, Y. Li, L. Li, Y. Zhao and J. Zhang, J. Hazard. Mater., 2012, 243, 117–123.
37. P. Wang, S. Su, J. Xiang, F. Cao, L. Sun, S. Hu and S. Lei, Chem. Eng. J., 2013, 225, 68–75.
38. P. Y. Wang, S. Su, J. Xiang, H. W. You, F. Cao, L. S. Sun, S. Hu and Y. Zhang, Chemosphere, 2014, 101, 49–54.
39. C. H. Chiu, H. C. Hsi, H. P. Lin and T. C. Chang, J. Hazard. Mater., 2015, 291, 1–8.
40. L. Tong, W. Q. Xu, X. Zhou, R. H. Liu and T. Y. Zhu, Energy Fuels, 2015, 29, 5231–5236.
41. L. Yang, X. Jiang, Z.-S. Yang and W.-J. Jiang, Ind. Eng. Chem. Res., 2015, 54, 1689–1696.
42. W. Q. Xu, L. Tong, H. Qi, X. Zhou, J. Wang and T. Y. Zhu, Ind. Eng. Chem. Res., 2015, 54, 146–152.
43. L. Ji, P. M. Sreekanth, P. G. Smirniotis, S. W. Thiel and N. G. Pinto, Energy Fuels, 2008, 22, 2299–2306.
44. X. Zhang, C. Li, L. Zhao, J. Zhang, G. Zeng, Y. E. Xie and M. E. Yu, Appl. Surf. Sci., 2015, 347, 392–400.
45. J. He, G. K. Reddy, S. W. Thiel, P. G. Smirniotis and N. G. Pinto, J. Phys. Chem. C, 2011, 115, 24300–24309.
46. N. Tang, Y. Liu, H. Q. Wang and Z. B. Wu, J. Phys. Chem. C, 2011, 115, 8214–8220.
47. G. Busca, L. Lietti, G. Ramis and F. Berti, Appl. Catal., B, 1998, 18, 1–36.
48. K. Krishna, G. B. F. Seijger, C. M. van den Bleek, M. Makkee, G. Mul and H. P. A. Calis, Catal. Lett., 2003, 86, 121–132.
49. H. Qi, W. Xu, J. Wang, L. Tong and T. Zhu, J. Environ. Sci., 2015, 28, 110–117.

---