Recent developments on gas–solid heterogeneous oxidation removal of elemental mercury from flue gas

Abstract

Mercury is a toxic and persistent environmental pollutant which has been recognized as a global threat to human health and our ecosystem because mercury bio-accumulates in the food chain and can be transformed into the more neurotoxic methylmercury. Among current and emerging abatement technologies for elemental mercury in flue gas, gas–solid heterogeneous oxidation is nowadays gaining increasing attention due to several inherent advantages. The catalysts and adsorbents are key materials that control the heterogeneous catalytic oxidation and adsorption of Hg⁰ from flue gas. Here we present a review of the recent developments on several catalysts and adsorbents, including noble metal-based catalysts, non-noble metal-based catalysts (transition metal oxides and selective catalytic reduction catalysts), activated carbon/coke-based sorbents, biochar-based sorbents, fly ash-based sorbents, mineral material-based sorbents and other novel catalysts. The key process parameters and kinetic reaction mechanisms and advantages and disadvantages of various emerging catalysts/adsorbents and technologies of Hg⁰ removal are described in detail.

Graphical abstract

Introduction

Mercury is an environmental persistent pollutant of great public concern because of its well-known high neurological toxicity, and well-documented food chain transport and bioaccumulation in its different forms, such as methylmercury with its concomitant adverse effects on our ecosystem and human health (Li et al. 2009). Human exposure to mercury occurs primarily by consumption of contaminated fish, resulting in such detrimental effects on human health, including neurological disorders, kidney damage, and birth defects. Therefore, United States Environmental Protection Agency (US EPA) identified mercury as a toxic and hazardous air pollutant under Title III of the 1990 Clean Air Act Amendments (CAAA) (Qiao et al. 2009). The total amount of anthropogenic mercury emission is about 1000–6000 tons per year (Yang et al. 2007). Combustion activities such as the burning of fossil fuels, municipal solid wastes, and medical wastes are the largest source of mercury emissions, which accounts for more than 90% of all anthropogenic mercury emissions (Reddy et al. 2012). To abate mercury emissions, some countries and regions have developed very stringent laws. In 2011, the United States Environmental Protection Agency (US EPA) promulgated the first national standard for mercury emissions, namely the Mercury and Air Toxics Standards (MATS), which aims to limit emissions of mercury and other toxic substances in power plants (Gao et al. 2013b). Also, in 2013, the United States Environmental Protection Agency updated the national emission standard (MATS), stipulating mercury emission limit below 0.003 lb GWh⁻¹ (Zhao et al. 2015b). In July 2011, the State Environmental Protection Administration of China (SEPA) released a new national standard (GB 13223-2011) of air pollutants for power plants, which requires new coal-burning power plants’ atmospheric mercury emissions should be less than 30 μg/m³ (Ancora et al. 2016). Therefore, the need and knowhow to curb mercury emissions are nowadays gaining significant global attention.

To reduce the emission of air pollutants, most coal-fired power plants have been installed some air pollution control devices (APCDs). Fabric filters (FF) and electrostatic precipitators (ESP), wet flue gas desulfurization (WFGD) system, and selective catalytic reduction (SCR) devices can effectively control particulate matter, SO₂ and NO_x in flue gas, respectively. During the combustion process, the elemental mercury in fuel is released into the flue gas in the form of vapor. This gaseous elemental mercury is subsequently oxidized partially to Hg²⁺ by heterogeneous (gas–solid) and homogeneous (gas–gas) reactions (Lee et al. 2002). Therefore, mercury in typical flue gas consists of three forms: elemental mercury (Hg⁰), oxidized mercury (Hg²⁺) and particulate-bound mercury (Hg^p) (Chi et al. 2009). Some studies have reported that the existing conventional air pollution control devices (APCDs) for reducing emissions of SO₂, NO_x and particulate matter can achieve a certain degree of mercury removal (Zheng et al. 2011; Wang et al. 2010b). For example, the Hg²⁺ can be efficiently removed by the existing wet flue gas desulfurization (WFGD) equipment due to its high water solubility (Li et al. 2011a). Fabric filters (FF) and/or electrostatic precipitators (ESP) can easily capture Hg^p from flue gas (Cao et al. 2008). In contrast, Hg⁰ with high volatility and low solubility in water is very difficult to be effectively removed by existing APCDs (Gutiérrez et al. 2007; Galbreath and Zygarlicke 1996). To reduce operating costs, the use of existing conventional APCDs to remove elemental mercury from flue gas is considered as an effective option for mercury abatement. Therefore, one of the core issues of mercury emission control is the efficient oxidation of elemental mercury (Hg⁰) into the oxidized form (Hg²⁺).

To effectively control mercury emissions from coal-fired boilers, many Hg⁰ control technologies have been developed over the past few decades, including adsorptive removal (Vidic and Siler 2001; Tan et al. 2012a; Chung et al. 2009), catalytic oxidation (Wang et al. 2010a; He et al. 2013), advanced oxidation (Wang et al. 2010c; An et al. 2014; Xu et al. 2008), and traditional chemical oxidation technologies (Wang et al. 2007; Hutson et al. 2008; Stergarsek et al. 2010). Adsorption processes utilizing modified and supported sorbents can effectively remove Hg⁰ in flue gas by converting it to Hg^p and Hg²⁺ (Pavlish et al. 2004; Wu et al. 2012). In addition, catalytic oxidation processes such as selective catalytic reduction (SCR) and using catalysts composed of noble metals, metal oxides, and multi-metal oxides can simply, efficiently, and cost-effectively oxidize Hg⁰ to Hg²⁺ (Kamata et al. 2009; Yang et al. 2010). Among technologies for Hg⁰ removal from flue gas, the gas–solid heterogeneous adsorption and catalytic oxidation are recognized as the most promising (Pavlish et al. 2004; Wu et al. 2012; Kamata et al. 2009; Yang et al. 2010). While some reviews on mercury control have been published in the past few decades, these reviews appear to be limited in scope and/or outdated due to the prolific research productivity in this field, and hence, there is a need for a more comprehensive review of the recent developments and emerging technologies. (Gao et al. 2013b; Zheng et al. 2012; Pavlish et al. 2003; Fu et al. 2010). This exhaustive review discusses the emerging catalysts and adsorbents, including noble metal-based catalysts, non-noble metal-based catalysts (transition metal oxides and SCR catalysts), activated carbon- and coke-based sorbents, biochar-based sorbents, fly ash-based sorbents, mineral material-based sorbents, and other novel catalysts, in detail. Some challenges, problems, and future research directions of Hg⁰ removal using these catalysts and adsorbents are also discussed. The key process parameters, advantages, and disadvantages of current and emerging technologies are summarized, and the reaction kinetics and mechanistic aspects of gas–solid heterogeneous catalytic oxidation and adsorption of Hg⁰ from flue gas are described in detail.

Gas–solid heterogeneous oxidation of mercury

It is well known that Hg⁰ in flue gas is very difficult to be captured due to its low solubility in water and high volatility. However, Hg^p can be removed in particle controllers, and the oxidized mercury (Hg²⁺) can be easily captured by the wet flue gas desulfurization (WFGD) system due to its high water solubility. Therefore, a combination of wet flue gas desulfurization (WFGD) system and elemental mercury heterogeneous oxidation is considered as a promising method for Hg⁰ control. To date, a number of heterogeneous catalyst and adsorption systems have been developed for Hg⁰ oxidation or removal and categorized into seven groups, namely noble metal-based catalysts, non-noble metal-based catalysts, activated carbon-/coke-based sorbents, biomass char-based sorbents, fly ash-based sorbents, mineral material-based sorbents, and other novel catalysts.

Noble metal-based catalysts

Noble metals such as Au, Pd, Ag, Ru, and Ir have been considered as potential Hg⁰ oxidation catalysts due to their regeneration performance and excellent mercury adsorption capacity. To obtain a high mercury removal capacity, the noble metals are usually supported on materials with well-developed pore structures and large Brunauer–Emmet–Teller (BET) surface areas, such as alumina, silica, zirconia, titania, carbons, and zeolite. The modification conditions and Hg⁰ removal capacities of the investigated noble metal catalysts are summarized in Table 1.

Table 1 Reaction conditions and Hg⁰ removal performance of noble metal catalysts

Pd has been recognized as a promising catalyst for mercury removal (Granite et al. 2006). In the study by Hou et al. (2014a), the Pd catalyst exhibited high mercury removal efficiency in the operating temperature range of 200–270 °C, reporting that, up to 270 °C, a catalyst containing 8% Pd provided 90% Hg⁰ removal efficiency and retained good stability at mid-temperatures. Li et al. (2014a) also tested the effects of three operating temperatures, 250, 300, 350 °C, on mercury adsorption capacity and found the mercury adsorption efficiency at 250 °C was higher than those at 300 and 350 °C, confirming the positive effects of operating at mid-temperatures. Hou et al. (2014a) and Han et al. (2012, 2016) tested the effects of H₂ and CO on mercury removal and observed that H₂ and CO could enhance the removal efficiency of elemental mercury as a result of the reduction of PdO to Pd metal. Hou et al. (2014a) also studied the effect of HCl on Hg⁰ removal over Pd based catalyst and found that HCl promoted mercury removal. Yue et al. (2015) examined the effects of H₂S on mercury removal over Pd/AC catalyst and showed that H₂S significantly inhibited Hg⁰ adsorption and removal efficiency, possibly due to the reaction of H₂S with the PdO to form PdS. Li et al. (2014a) suggested two Hg⁰ removal pathways over the Pd/AC catalyst in N₂-Hg-H₂S atmosphere, reaction of Hg⁰ with elemental palladium (Pd⁰) to produce Hg-Pd amalgam or the reaction of some oxygen-containing functional groups on the surface of activated carbon (AC) with Pd⁰ to form PdO. However, as shown in Fig. 1, the possible product in the N₂-Hg-H₂S atmosphere, PdS, is difficult to reduce to Pd⁰, suggesting that this product could be inhibitory to the mercury removal process.

Fig. 1

(reproduced with permission from Li et al. 2014a)

Schematic diagram of the pathways of mercury removal over the Pd/activated carbon samples in N₂–Hg–H₂S atmosphere. The possible product in the N₂–Hg–H₂S atmosphere, PdS, is difficult to reduce to Pd⁰, suggesting that this product could be inhibitory to the mercury removal process

Gold-based catalysts are also considered as promising alternatives for Hg⁰ removal because gold has the ability to adsorb and react with Hg⁰ on its surface to form amalgam (Presto and Granite 2009; Zhao et al. 2006). Song and Lee (2016) synthesized a gold (Au)-based catalyst via an impregnation method and found the catalyst to achieve a 97% elemental mercury oxidation. Gómez-Giménez et al. (2015) studied the effect of SO₂ and O₂ on mercury removal and showed that these flue gas components promoted mercury removal in the presence of gold nanoparticles, attributable to the catalytic activity of Au. Ballestero et al. (2013) examined the regenerability of the Au-based catalyst through several cycles of Hg⁰ capture regeneration and found that when the regeneration temperature was 220 °C, the Au-based catalyst maintained a high mercury removal efficiency in several regeneration cycles. In the process of elemental mercury oxidation, some reactants such as chlorine atoms have been shown to play an important role since gold could dissociate the adsorbed Cl₂ molecule into Cl atoms, which subsequently could react with Hg⁰ to form HgCl₂, enhancing Hg⁰ removal (Dranga and Koeser 2015). Lim and Wilcox (2013) examined the Hg⁰ oxidation via a Langmuri–Hinshelwood (L–H) mechanism and suggested that the adsorbed Cl₂ (or HCl) could react with Hg⁰ to produce HgCl and HgCl₂, as shown in Fig. 2, illustrating that the Hg⁰ oxidation on the surface of Au is a step-by-step Hg⁰ oxidation (Hg → HgCl → HgCl₂) rather than a direct oxidation of Hg⁰ to HgCl₂.

Fig. 2

(reproduced with permission from Lim and Wilcox 2013)

Reaction pathways of mercury oxidation on the surface of Au. The Hg⁰ oxidation on the surface of Au is a step-by-step Hg⁰ oxidation (Hg → HgCl → HgCl₂) rather than a direct oxidation of Hg⁰ to HgCl₂

Other noble metals such as Ag, Ru, and Ir also have been reported to be effective catalysts for mercury removal from flue gases (Karatza et al. 2011; Yan et al. 2011). Zhao et al. (2015d) prepared a Ag-based catalyst by an impregnation method and demonstrated its excellent performance for mercury removal in a simulated flue gas. Rungnim et al. (2015) synthesized Ag/TiO₂ catalyst samples by loading 5% Ag on TiO₂ powder and investigated possible synergistic effects between Ag and TiO₂ toward Hg⁰ removal using periodic density functional theory (DFT) calculations. They showed an improved Hg⁰ removal, suggesting the synergy resulted from the promotion of electron transfer from adsorbed elemental mercury to Ag/TiO₂ catalyst, with the concomitant effect of greatly enhancing the mercury removal.

It has been reported that RuO₂ is an excellent mercury oxidation catalyst and that halogen gases play an important role in the mercury oxidation process (Chen et al. 2014; Liu et al. 2016b, 2017). Liu et al. (2016b, 2017) studied the effect of halogen gas on mercury removal using RuO₂/TiO₂ catalyst in the presence of HCl or HBr, and the results showed 85 and 90% mercury removal in the presence of 10 ppm HCl and 1 ppm HBr, respectively, and that HgCl₂ and HgBr₂ were the main respective oxidation products. Liu et al. (2017) also found that the RuO₂/TiO₂ catalyst exhibited a good resistance to SO₂ poisoning under bituminous coal flue gas (SO₂ > 2000 ppm in flue gas). It was suggested that the oxidation reaction mechanism of elemental mercury follows the Deacon process as shown in Fig. 3. Chen et al. (2016) prepared the IrO₂-based catalyst via a sol–gel method and also found that the novel IrO₂-modified catalyst displayed a higher catalytic activity for mercury oxidation in a flue gas system, and the mechanism also followed the Deacon reacting scheme illustrated in Fig. 3.

Fig. 3

(reproduced with permission from Liu et al. 2016b)

Schematic diagram of Hg⁰ oxidation reaction over RuO₂ catalyst in the presence of HCl or HBr. In the presence of HCl or HBr gas, the RuO₂ catalysts follow the Deacon process

Non-noble metal-based catalysts

Transition metal oxides-based catalysts

Transition metal oxides, including mainly Fe₂O₃, CuO, MnO_2, and CeO₂, commonly supported on carriers such as alumina, silica, titania, have been tested as potential elemental mercury oxidation catalysts. The advantages of these oxides compared with the noble metal catalysts, include the lower cost, widely available sources, and the relatively high catalytic oxidation activity. These supporters not only could increase the dispersion degree of metal oxides, but in some cases, also participate in the mercury removal process. Typical modification conditions and the resulting Hg⁰ removal capacities are summarized in Table 2.

Table 2 Reaction conditions and Hg⁰ removal performance of non-noble metal-based catalysts

Copper-based catalysts are considered as promising mercury removal catalysts due to their abilities to store/release oxygen via the redox reaction between Cu²⁺ and Cu⁺ (Tsai et al. 2013; Li et al. 2013c; Du et al. 2015). Liu et al. (2015b) synthesized Cu/Al₂O₃ catalyst via a wetness incipient method and reported that with optimal loading of 10 wt% Cu, more than 95% Hg⁰ oxidation efficiency was attained during the first 20 h at 140 °C. It was also observed that the loading value of CuCl₂ has a significant effect on the activity of the catalyst. At low CuCl₂ loadings, it was speculated that CuCl₂ could react with Al₂O₃ to form copper aluminate (CuAlO₂) which was inactive for mercury oxidation, while high loadings of CuCl₂ would be expected to be present in a highly dispersed amorphous state on the surface of the CuAlO₂, which contributed to mercury removal. It was also observed that high loading of Cu into the Al₂O₃ support exhibited excellent SO₂ poisoning resistance under 10 ppm HCl (Yamaguchi et al. 2008). Zhou et al. (2014) tested the effect of HCl on Hg⁰ removal using CuCl₂/TiO₂ catalyst, and they also found that the Cl atoms in HCl had a positive effect on Hg⁰ removal. Xu et al. (2014a) suggested that CuO had showed a good performance for Hg⁰ removal in the presence of low level HCl, and with a CuO/TiO₂ catalyst prepared by a wetness impregnation method, they reported Hg⁰ removal efficiency of nearly 100% obtained with HCl concentration of 5 ppm. The positive effect of HCl was attributed mainly to the production of active atomic chlorine species.

Manganese-based catalysts are attractive potential alternatives for Hg⁰ capture from flue gas due to their low cost and expected excellent oxidation performance, stemming from their inherent multiple oxidation states (Xu et al. 2015a; Li et al. 2010). Yu et al. (2015) investigated the performance of Hg⁰ removal using M/Al catalysts (M = Mg²⁺, Zn²⁺, Cu²⁺, and Mn²⁺), and they found that compared with Mg/Al, Zn/Al, and Cu/Al catalysts, Mn/Al catalysts exhibited the highest Hg⁰ removal performance at 300 °C. They concluded that Mn⁴⁺ species, which was the main active sites, played a very important role in the removal process of Hg⁰. Xu et al. (2015b) reported that the improved removal of Hg⁰ from flue gas, achieved with heterogeneous reaction between Hg⁰ and Mn⁴⁺, resulted from the transition of high valence (Mn⁴⁺) to low valence Mn (Mn³⁺ and Mn²⁺). Xie et al. (2013) also obtained similar results in the investigation of Hg⁰ removal using Mn-based catalysts. Zhang et al. (2015a) examined the influence of calcination temperature in the 200–800 °C range on Hg⁰ capture using MnO_x/TiO₂ sorbents. It was observed that the calcination temperature had an important effect on the activity and structure of the MnO_x/TiO₂ catalysts. The catalyst exhibited excellent performance for Hg⁰ removal at high temperature of 400 °C; however, BET surface area, pore volume, and the content of Mn⁴⁺ of the catalyst decreased at calcination temperatures greater than 400 °C. Scala and Cimino (2015) studied the effect of flue gas composition on Hg⁰ capture using manganese-based catalysts, and their results showed that both CO and CO₂ reduced the Hg⁰ capture performance, while NO had no detectable effect, and 50 ppm HCl significantly improved the Hg⁰ removal. Zhang et al. (2014a, 2015c) proposed that the Hg⁰ oxidation by HCl over manganese-based catalyst followed the Hg → HgCl → HgCl₂ pathway, rather than the direct production of HgCl₂.

Scala and Cimino (2015) and Xie et al. (2012) examined the effect of SO₂ on Hg⁰ removal both MnO_x-based and Mn-TiO₂ catalysts, and the results showed that SO₂ had a negative effects on the performance of both catalysts, mainly due to the competitive adsorption between Hg⁰ and SO₂. Zhang et al. (2017b) also reported that the presence of SO₂ weakened the Hg⁰ removal capacity of the MnO_x-based catalyst. To further enhance the Hg⁰ removal effectiveness of Mn-based catalysts in the presence of SO₂, some metal elementals (Cu, Fe, Ce, Mo) have been utilized to modifying agents. Wang et al. (2013) prepared CuO–MnO_2–Fe₂O₃/Al₂O₃ catalyst by an improved impregnation method and studied the effect of SO₂ concentration on Hg⁰ removal, reporting that SO₂ has little effect on Hg⁰ removal due to the larger affinity between Cu and sulfur. Zhao et al. (2016a) examined the effect of SO₂ on Hg⁰ capture using Mo-doped Mn/CNT catalyst, showing that the presence of SO₂ improved Hg⁰ removal, and attributing this to Mo promoting of the conversion of SO₂ to SO₃, with concomitant improvement in Hg⁰ removal efficiency. Wang et al. (2014) also investigated the effect of SO₂ on Hg⁰ removal using MnO_x–CeO₂/Al₂O₃ catalyst, and found that the addition of Ce effectively resisted the poisoning effect of SO₂ on the catalysts.

Cerium oxide and Ce-based catalysts have gained widespread attention due to the unique redox cycle between Ce⁴⁺ and Ce³⁺, excellent oxygen storage capacity and high oxidation capacity, and resistant to SO₂ poisoning (Li et al. 2011b). In the process of redox reaction of Ce⁴⁺ and Ce³⁺, bulk oxygen species and surface oxygen vacancies with high mobility are easily produced, which facilitate the effectiveness of Hg⁰ removal. Li et al. (2013a) synthesized Ce-based catalysts using TiO₂ nanoparticles by an ultrasonic-assisted impregnation method, and reported that the addition of 1200 ppm SO₂ into a flue gas system enhanced the performance of Hg⁰ capture. In addition, the results of Ma et al. (2017) showed that the addition of CeO₂ improved the water vapor resistance of the catalyst, and even with 5% water vapor in the flue gas, the high-level removal efficiency of Hg⁰ was only slightly reduced. Considering the superior activity and the unique redox cycles of Ce⁴⁺/Ce³⁺couple, the incorporation of CeO₂ into other metal oxide catalysts is generally believed to improve their Hg⁰ removal performances.

Zhou et al. (2013) and Hou et al. (2014b) investigated the Hg⁰ removal over CeO₂–TiO₂ catalysts and reported that when HCl or H₂S was present alone in the flue gas, more than 97% of Hg⁰ was captured, while the simultaneous presence of HCl and H₂S resulted a prohibitive effect on the effectiveness of Hg⁰ capture. Zhou et al. (2013) also found that the presence of H₂ and CO have a negligible effect on the capture of Hg⁰ at 150 °C. Zhang et al. (2015e) synthesized a series of Ce-based V₂O₅/TiO₂ catalysts by an ultrasound-assisted impregnation method and found that the V(1)Ce(10)Ti catalyst had the best Hg⁰ oxidation performance. Li et al. (2017d) examined the synergistic effect of CeO₂ and CuO using CuTi, CeTi and CuCeTi catalysts prepared by a sol–gel method. They found that, unlike the CuTi and CeTi catalysts, the Hg⁰ removal efficiency of CuCeTi catalyst at 200 °C was about 99.0%, the high value ascribed to the combined effect of the presence of both CeO₂ and CuO.

Selective catalytic reduction catalysts

Recently, selective catalytic reduction (SCR) systems have been applied in many coal-fired power plants for NO_x removal due to its higher economy of scale, efficiency, and selectivity. Typical selective catalytic reduction (SCR) catalysts usually apply TiO₂ and some catalytically active components (such as WO₃, V₂O₅ and/or MoO₃) as precursors and activators, respectively. The modification conditions and Hg⁰ removal capacities of SCR type catalysts are summarized in Table 2. V₂O₅ is the major active ingredient of the selective catalytic reduction (SCR) catalyst, which can be employed not only to control the emission of NO_x but also to remove Hg⁰ from flue gas. Zhao et al. (2015c) reported that the V₂O₅-rich SCR catalyst exhibited a superior Hg⁰ removal performance in the range of 250–350 °C. For WO₃–V₂O₅/TiO₂ catalysts, the Hg⁰ oxidation in the presence of both O₂ and HCl was found to follow the Eley–Rideal mechanism (Gao et al. 2013a). Wang et al. (2015a) also investigated the Hg⁰ removal in CO₂-enriched flue gas using WO₃–V₂O₅/TiO₂ catalysts, and they found that high concentration of CO₂ (80 vol%) promoted the capture efficiency of Hg⁰, but inhibited the removal of NO. MoO₃ is often introduced into the catalyst’s formulation to improve its resistance to SO₂ poisoning. Zhao et al. (2014) found that the V₂O₅–MoO₃/TiO₂ catalyst was excellent Hg⁰ oxidation, and the Hg⁰ removal process could be explained by the Mars–Maessen mechanism. To further study the Hg⁰ capture performance of this catalyst system in actual flue gas, Zhao et al. (2015a) performed a test in a coal-fired power plant, and  reported higher than 90% Hg⁰ removal efficiency.

Selective catalytic reduction (SCR) system is widely applied in coal-fired power plant to simultaneously control the emissions of NO_x and Hg⁰. However, the conventional selective catalytic reduction catalysts are not effective enough for the removal of Hg⁰ in the presence of low HCl concentrations and are often suppressed by the presence of SO₂ and NH₃ in the flue gas (Kamata et al. 2008). Therefore, some metal oxides are usually used to modify the selective catalytic reduction (SCR) catalysts. Huang et al. (2016) prepared the Fe₂O₃/SCR catalyst by an impregnation method and found that the introduction of Fe₂O₃ could significantly improve the Hg⁰ removal ability of the SCR catalyst. The active temperature window of Fe₂O₃/SCR catalyst was found to range from 150 to 450 °C, which is wider than that of conventional SCR catalysts. They suggested that the Fe³⁺ could react with HCl to release active Cl species by the Mars–Maessen mechanism and then the generated active Cl species could participate in the Hg⁰ removal by the L–H mechanism. The proposed plausible Hg⁰ oxidation mechanism is shown in Fig. 4.

Fig. 4

(reproduced with permission from Huang et al. 2016)

Schematic of the possible Hg⁰ oxidation mechanism in HCl-O₂ on over the Fe₂O₃/SCR catalyst. The active chlorine species generated by the reaction of Fe³⁺ and HCl can react with adsorbed Hg⁰ to form HgCl₂. The gas-phase O₂ in flue gas regenerated the chemisorbed oxygen and lattice oxygen

Chi et al. (2017) prepared a series of Ce-Cu-modified selective catalytic reduction (SCR) catalysts by ultrasonic-assisted impregnation method for simultaneous removal of Hg⁰ and NO_x and found that a 7%Ce–1%Cu/SCR catalyst showed a superior performance at 200–400 °C. The catalyst also exhibited higher resistance to water vapor and SO₂. The Hg⁰ removal performance of MnO_x-treated commercial SCR catalysts was also evaluated (Chiu et al. 2015), and the results showed that both 5 and 10% MnO_x-impregnated SCR catalysts had higher Hg⁰ oxidation efficiency.

Activated carbon/cokes based sorbents

Activated carbon/cokes have been proven to be effective sorbents for Hg⁰ removal, and sulfur, halogens, and metal oxides are the most common additives/modified reagents, which have been widely studied for the modifications of these sorbents to improve their removal efficiencies for Hg⁰. The modification conditions and Hg⁰ removal capacities of activated carbon-/coke-based sorbents are summarized in Table 3.

Table 3 Reaction conditions and Hg⁰ removal performance of carbon sorbents

Sulfured carbon sorbents

Sano et al. (2017) performed a laboratory-scale test of Hg⁰ removal over sulfur-impregnated activated carbon and raw activated carbon and reported that S (sulfur) impregnation resulted in 50 times higher Hg⁰ removal than the performance of the raw activated carbon. Hsi and Chen (2012) studied the effects of acidic/oxidizing gases, O₂, HCl, SO₂, and NO which are commonly found in the flue gas, on Hg⁰ removal using simulated flue gas over sulfur-impregnated activated carbon. They observed the flue gas components had strong positive effect on the catalyst’s performance, with the largest Hg⁰ removal capacity of 2310 μg/g obtained in the presence of O₂, HCl and NO.

Ie et al. (2013) synthesized a series of innovative composite powdered activated carbon (PACs) by an impregnation method using aqueous-phase sodium sulfide (Na₂S) and vapor-phase elemental sulfur (S⁰) in different sequences and investigated their performances in the removal of Hg⁰ or HgCl₂. They found that the Hg⁰ and HgCl₂ removal capacities of powdered activated carbon (PACs) impregnated with aqueous Na₂S solution followed by gaseous sulfur (S⁰), respectively, were 1.98 and 1.42 times higher than those of the samples impregnated in the opposite sequence. Yao et al. (2014) also studied the performance of activated carbon fibers functionalized with sulfur-containing groups and reported that sulfur impregnation decreased the pore volume and surface area of activated carbon fibers. However, compared with the raw activated carbon fiber samples, the Hg⁰ removal capacity of sulfur-treated activated carbon fibers increased due to the incorporation of the sulfur groups.

Halogenated carbon sorbents

Zhou et al. (2015) prepared Br-based activated carbon by an impregnation method and evaluated the in-flight Hg⁰ capture performance in an entrained flow reactor. They found that the Hg⁰ removal efficiency of raw activated carbon was significantly enhanced by the NH₄Br impregnation. Yao et al. (2013) also prepared Br-based activated carbon fibers using KBr solution, and by KBr impregnation, bromine vapor, and electrochemical modification methods, respectively. For the brominated activated carbon fibers, the introduction of Br atoms promoted the Hg⁰ oxidation process. They also found that the brominated activated carbon fibers modified by bromine vapor and electrochemical methods using KBr solution exhibited stable Hg⁰ removal capacity (30–33% capture), which was retained up to 3 months. Rupp and Wilcox (2014) examined the effects of flue gas components (NO_x, SO₂) on Hg⁰ removal using brominated activated carbon fibers and reported that while NO_x promoted the oxidation of Hg⁰, SO₂ prevented the Hg⁰ adsorption, and the interaction of NO_x and SO₂ with Br decreased sorbent’s performance.

Tsai et al. (2017) investigated CuCl₂-impregnated activated carbon for Hg⁰ removal using a fixed-bed reactor system. Results from the tests showed that the Cl-impregnated samples achieved better Hg⁰ removal capacity than non-impregnated samples, with the Hg⁰ removal capacity of the 8% CuCl₂-impregnated sample reported to be 631.1 μg/g. Li et al. (2017a) prepared NH₄Cl-modified activated carbons for Hg⁰ removal by an impregnation method, and they found that Cl-doped activated carbons exhibited a good performance for Hg⁰ capture. De et al. (2013) impregnated activated carbons using various halogens such as ammonium and potassium halides. They observed that the introduction of halide ions greatly enhanced the capacity of the activated carbons for Hg⁰ removal. For the same loading values of halide (I, Cl and Br) ions, the Hg⁰ capture performance of ammonium halide-modified activated carbon was higher than those of potassium halide-modified activated carbon. Also, the I-impregnated sample exhibited the highest Hg⁰ removal capacity compared to Cl- and Br-impregnated samples.

Tong et al. (2017) synthesized the I-impregnated activated carbons using an impregnation method and investigated the Hg⁰ capture, and the adsorption mechanism and the effects of simulated flue gas components. They found that the formation of I₂ molecules on the surface of I-impregnated activated carbons significantly promoted Hg⁰ removal and proposed the plausible adsorption mechanism shown in Fig. 5. They also observed that low concentrations of SO₂ had a promotional effect on Hg⁰ oxidation, but high concentrations of SO₂ had a negative impact on Hg⁰ capture. They also found that the Hg⁰ removal efficiency significantly increased with increasing NO concentration from 0 to 100 ppm, while high NO concentration of 300 ppm showed antagonistic effects.

Fig. 5

(reproduced with permission from Tong et al. 2017)

Possible adsorption mechanism for Hg⁰ under simulated flue gas. The formation of I₂ molecules, SO₃²⁻/SO₄²⁻ active species and NO₂ active species on the surface of I-impregnated activated carbons significantly promoted Hg⁰ removal

Metal oxides-modified carbon sorbents

Zhao et al. (2016b) studied the use of activated coke impregnated with CuO (a CuO/AC-H sample), focusing on the effects of the copper loading, reaction temperature, calcination temperature, and flue gas components (NO, O₂) on Hg⁰ capture, and found the optimal reaction temperature, copper loading value, and calcinations temperature to be 160 °C, 8%, and 300 °C, respectively, and that NO and O₂ showed positive effects on Hg⁰ capture. CeO₂ has been widely investigated as one of the catalysts for selective catalytic reduction (SCR) of NO_x, and Hg⁰ removal due to its large oxygen storage capacity and unique redox couple Ce³⁺/Ce⁴⁺, and excellent ability to shift between CeO₂ and Ce₂O₃ under oxidizing and reducing conditions, respectively (Zhang et al. 2017a; Zhao et al. 2017a). Zhang et al. (2017a) prepared CeO₂-supported semi-coke (SC) sorbents by an impregnation method and observed much better Hg⁰ removal capacity than that of unmodified semi-coke (SC) but high concentration of H₂O vapor showed inhibitory effects. It was demonstrated that the Ce–OH groups formed by the reaction of CeO₂ and H₂O vapor consumed lattice oxygen on the surface of samples, with the concomitant effect of decreasing the Hg⁰ removal efficiency. Zhao et al. (2017a) also obtained similar results in studying the effect of water vapor on Hg⁰ removal performance over CeO₂-supported semi-coke (SC) sorbents. Wu et al. (2017), Xie et al. (2015), and Zhang et al. (2016d) prepared Ce–Mn-co-modified activated carbons (AC), Mn–Ce-mixed oxides-modified activated coke (MnCe/AC), and Mn/Ce-modified semi-coke (Mn/Ce-SC) by an impregnation method, respectively, and found the modified sorbents to exhibit excellent Hg⁰ capture capability. Wu et al. (2015b) investigated the performance of CoCe/AC sorbents prepared by an impregnation method for Hg⁰ capture from flue gas at 110–230 °C and reported superior performance compared to Ce/AC, Co/AC, and virgin AC, with a 92.5% Hg⁰ removal achieved at 170 °C. Based on the results obtained from XPS and TGA analyses, the valence transitions of Co³⁺/Co²⁺ and Ce⁴⁺/Ce³⁺ produced lattice oxygen, promoting Hg⁰ oxidation and removal. Wang et al. (2016c) also reported that activated coke (AC) impregnated with CeO₂ and Fe₂O₃ (denoted Fe₂O₃-CeO₂/AC), significantly improved Hg⁰ removal capacity.

Plasma-treated carbon sorbents

In recent years, plasma modification has received widespread attention in research for functionalizing catalyst and sorbents. For example, non-thermal plasma produces energetic electrons, ions, and active radicals, which could improve the pore structure of sorbents and increase the active functional groups on the surface of the sorbents. Some investigators (Zhang et al. 2015b, 2016c, e) have shown that plasma modification could form multiple functional groups on the surface of sorbents, ameliorating the Hg⁰ removal process.

Zhang et al. (2015b) used a non-thermal plasma technology to modify activated carbon (AC) in air environment and found the modified sample to have a higher Hg⁰ removal efficiency than the corresponding raw sample. The results of XPS showed that the modification by non-thermal plasma increased the content of ester groups (C(O)–O–C) and carbonyl groups (C=O) on the activated carbon, which played a key role in Hg⁰ removal. Zhang et al. (2016e) also obtained the similar results in studying the effect of oxygen non-thermal plasma modification, reporting that the modified activated carbon (AC) exhibited a high removal performance for Hg⁰ from flue gas. Zhang et al. (2016c) modified activated carbon (AC) by Cl₂ non-thermal plasma method and found the sample to greatly enhance Hg⁰ removal by increasing the chlorinated (Cl) active sites on the surface of the activated carbon (AC). The corresponding XPS analysis indicated that a large number of C–Cl groups resulted from the treatment, which could have oxidized the Hg⁰ to HgCl₂, as illustrated in Fig. 6.

Fig. 6

(reproduced with permission from Zhang et al. 2016c)

Mechanism of modified activated carbon (AC) for Hg⁰ removal. A large number of C–Cl groups generated by the Cl₂ non-thermal plasma treatment can oxidize Hg⁰ to HgCl₂, thus promoted the removal of Hg⁰

Biomass char-based sorbents

The results of some studies (Hua et al. 2010; Lee et al. 2006; Diamantopoulou et al. 2010) have indicated that injection of activated carbon into the flue gas system is a promising method for Hg⁰ removal from flue gas. However, large activated carbon (AC)/Hg⁰ ratio and high operation costs have limited large-scale applications (Hsi et al. 2002; Scala et al. 2011). Biomass char is the by-product of biomass pyrolysis under oxygen-free conditions. With the low costs and the simplicity of preparation, it could be considered as an attractive alternative to activated carbon (AC) (Liu et al. 2011). Therefore, pyrolysis chars, which are made from cheap and renewable resources, have been extensively studied recently in the field of Hg⁰ removal (Hsi et al. 2011; Klasson et al. 2010; Fuente-Cuesta et al. 2012). However, they require physical techniques to modify pore structure such as specific surface area, pore volume, and pore size and/or chemical modification to increase active functional groups on the surface. For example, the use of active ingredients such as halogens, metal oxides, and acid to modify biochars has been reported to improve their performance for Hg⁰ removal. The modification conditions and Hg⁰ removal capacities of modified biochars are summarized in Table 4.

Table 4 Reaction conditions and Hg⁰ removal performance of biochars

The results of a number of studies (Johari et al. 2016a, b; Klasson et al. 2014) have suggested that the pyrolysis conditions could also substantially influence the yield and physicochemical properties of chars. Johari et al. (2016b) prepared a series of coconut pith (CP) chars at different pyrolysis temperatures and found the Hg⁰ removal capacity to increase with increasing pyrolysis temperature, with the highest removal capacity of 6067.49 μg/g obtained at 900 °C. Klasson et al. (2014) prepared four different biochars (almond shells, cottonseed hulls, lignin, and chicken manure) at different pyrolysis temperature and reported that chicken manure exhibited the best Hg⁰ removal performance of 95% from flue gas at 650 and 800 °C.

Halogens-modified biochar

Many studies have reported that chemical modifications of sorbents using halogens could significantly enhance the Hg⁰ removal from flue gas (Li et al. 2015a, b; Shen et al. 2015a). The halogens (chloride, bromide, and iodide) have been demonstrated as effective reagents for modification of sorbents to improve their performance in Hg⁰ removal from flue gas. Li et al. (2015a) and Shen et al. (2015a) prepared low-cost sorbents using municipal solid waste and medicinal residues by a chloride impregnation method and reported that NH₄Cl-modified sorbents showed improved performance for Hg⁰ removal. Li et al. (2015b) investigated the effects of flue gas composition on Hg⁰ removal using NH₄Cl-impregnated medicine residue biochars and reported that the presence of O₂ and NO increased Hg⁰ removal, but water vapor suppressed the removal process. A dual effect of SO₂ concentration was observed on Hg⁰ capture, that is, low SO₂ concentration enhanced Hg⁰ removal while high SO₂ concentration was antagonistic.

Li et al. (2015c) carried out a comparative study of NH₄Cl modified biochars from three solid wastes (medicinal residues, municipal solid wastes, and cotton straw), showing that the chemically modified biochars, especially the modified cotton straw char exhibited higher Hg⁰ removal capacity than modified activated carbon (AC). In addition, the biochar derived from waste tire also demonstrated an excellent Hg⁰ removal performance, resulting from generated mercury sulfide chemisorption sites on the surface of the biochar (Li et al. 2017b). Shen et al. (2017) studied NH₄Cl-modified biochar sorbents derived from waste tea and found that the generated C–Cl and C=O groups on the surface of the biochar promoted the oxidation of Hg⁰, resulting in an excellent Hg⁰ removal. Xu et al. (2016c) synthesized a novel Cl-Char composite by the co-pyrolysis of biomass (wood and paper) and polyvinyl chloride (PVC) and reported a 90% Hg⁰ removal capacity at 140 °C, which was more than 2.5–5 times than that of a raw char. In addition, biochars modified by metal chlorides have been evaluated for their Hg⁰ capture performances from flue gas (Shu et al. 2013; Tan et al. 2015). Shu et al. (2013) studied mulberry twig chars (MT) modified by ZnCl₂, H₂O₂, and NaCl, respectively, and reported that the ZnCl₂-impregnated char was better than the other chemically treated samples for Hg⁰ removal. Tan et al. (2015) also compared Hg⁰ capture performance of bamboo charcoal (BC) impregnated by ZnCl₂ and FeCl₃ and found that the impregnated BCs was better than raw bamboo charcoal (BC), with the FeCl₃-impregnated BCs showing the highest Hg⁰ removal efficiency (> 99.9%) at 140 °C.

It is well known that chemical modification of sorbents with bromine plays a key role in the adsorption and oxidation of Hg⁰ (Yang et al. 2018a, b; Tang et al. 2017; Zhu et al. 2016). Yang et al. (2018a, b) reported that sargassum chars’ effectiveness for Hg⁰ removal was greatly improved after NH₄Br and NH₄Cl impregnation, with the NH₄Br-modified samples showing improved performance attributable to the generation of C–Br and C=O groups on the surface of the sorbents. Tang et al. (2017) developed a low-cost sorbent based on rice husk char (RHC) using HBr impregnation method and demonstrated that the modified rice husk char (RHC-HBr) had higher Hg⁰ removal capacity (57.84 μg/g) than those of activated carbon (AC). Zhu et al. (2016) evaluated the performances of chemically treated samples of rice husk char (RHC) and commercial coal-based activated carbon (CAC) and found that modification of NH₄Cl and NH₄Br significantly increased the Hg⁰ removal efficiency of rice husk char compared to CAC and that the NH₄Br-modified rice husk char exhibited the highest Hg⁰ removal performance.

Tan et al. (2012b) reported that KI modification of bamboo charcoal (BC) by an impregnation method, while it resulted in the decrease in the total volume and BET surface area, the modified BC (BC-I) exhibited superior capacity for Hg⁰. The results of XPS analysis of the used samples appear to support the generations of C-I_x compounds and I₂ and subsequent reactions with Hg⁰ to form iodated mercuric compounds, thus contributing to a higher Hg⁰ removal efficiency. Liu et al. (2018) also obtained similar results in their study of the removal of Hg⁰ using the KI-modified sargassum and enteromorpha chars. Li et al. (2016a, 2017c) synthesized cotton straw char sorbents using three different ammonium halides to capture Hg⁰ from flue gas and found that the Hg⁰ removal efficiency was in the order of NH₄I > NH₄Br > NH₄Cl. It was also noted that high reaction temperature improved the Hg⁰ removal performance of the NH₄I-modified sorbents.

Metal oxides-modified biochar

In recent years, metal oxides have been widely studied as effective sorbent modifiers for Hg⁰ capture due to their low costs and high activities. Among metal oxides used to modify biochars-based sorbents for Hg⁰ capture are FeO_x, CeO_x, CuO_x, MnO_x, and ZrO₂ (Yang et al. 2016a, 2017b; Zhou et al. 2017; Xu et al. 2018; Zeng et al. 2017). Yang et al. (2016a) prepared a novel magnetic sorbents (MBC) based on sawdust char by one-step pyrolysis of FeCl₃-laden method and showed that the modified sample has improved Hg⁰ removal capacity compared with those of raw biochar. XPS analysis indicated that the generated Fe₃O₄ and C=O groups were the major active oxidation/adsorption sites for Hg⁰ removal. The plausible mechanism of Hg⁰ removal proposed is depicted in Fig. 7.

Fig. 7

(reproduced with permission from Yang et al. 2016a)

Mechanism of magnetic sorbents for Hg⁰ removal. The generated Fe₃O₄ and C=O groups on the surface of novel magnetic sorbents (MBC) by one-step pyrolysis of FeCl₃-laden method significantly promoted Hg⁰ removal

Zhou et al. (2017) studied Hg⁰ removal by wheat straw char impregnated with K₂FeO₄ reagent, and the results appeared to show that K₂FeO₄ impregnation effectively improved pore structure of the wheat straw char, leading to enhancement in Hg⁰ removal. Yang et al. (2017b) further studied the Hg⁰ removal performance of wheat straw char modified by Mn–Ce-mixed oxides and found that the Mn/Ce redox cycle played an important role in Hg⁰ removal. Xu et al. (2018) modified rice straw char (RS) by impregnation with Cu-Ce-mixed oxides to remove Hg⁰ from flue gas and reported significant enhancement up to 95.26% efficiency. Zeng et al. (2017) prepared metal oxides (MnO_x and ZrO₂) and halide ions (I⁻) modified peanut shells char (6Mn-6Zr/PSC-I3) and demonstrated that the sample exhibited superior Hg⁰ removal capacity (15028.4 μg/g). Based on XPS analysis, two reaction stages could be detected in the Hg⁰ removal process. As shown in Fig. 8, at the initial reaction stage, Hg⁰ was first removed by the chemical adsorption sites of C-I groups. The Hg⁰ oxidation caused by the hydroxyl (OH) oxygen and lattice oxygen played a key role at the final reaction stage.

Fig. 8

(reproduced with permission from Zeng et al. 2017)

Reaction mechanism of 6Mn-6Zr/PSC-I3 for Hg⁰ removal at different reaction stages. At the initial reaction stage, Hg⁰ is first removed by the chemical adsorption sites of C-I groups. And the Hg⁰ oxidation caused by the hydroxyl (OH) oxygen and lattice oxygen played a key role at the final reaction stage

Other modification

In addition to the modification of biochar sorbents with halogens and metal oxides, other chemical modification methods involving the use of acid and alkali to increase surface activity, and physical modification such as plasma mainly to change pore structure, are also employed to improve the Hg⁰ removal performance of sorbents derived from biomass char (Xu et al. 2016a; Li et al. 2015d; Lopez-Anton et al. 2015; Niu et al. 2017; Wang et al. 2018). Xu et al. (2016a) modified bamboo char (BC) using an oxidizing agent (HNO₃) and showed that the modification by HNO₃ increased the Hg⁰ capture efficiency from the flue gas. The improvement was ascribed to the oxygen functional groups (such as carboxylate, carboxyl, and carbonyl groups) on the modified bamboo. In addition, the presence of water vapor improved the Hg⁰ removal performance. Li et al. (2015d) modified pyrolyzed char from waste tire by H₂SO₄ and HNO₃, respectively. The results showed that the raw pyrolyzed char (T6) exhibited superior Hg⁰ removal performance compared with those of acid-modified char (T6 N and T6S), attributable to the loss of sulfide functional groups on the modified samples.

Lopez-Anton et al. (2015) developed a low-cost sorbent based on leather industry waste by KOH activation and showed that the modified samples achieved the highest Hg⁰ removal capacity under the N₂/O₂ atmosphere. Niu et al. (2017) treated corn stalk samples by the dielectric barrier discharge (DBD) plasma method under N₂/O₂/H₂O atmosphere and found the DBD plasma-treated corn stalk sorbents to have a higher Hg⁰ removal capacity compared with that of a raw corn stalk. The XPS analysis indicated that oxygen-containing functional groups increased significantly on the surface of the samples after the dielectric barrier discharge (DBD) plasma treatment, which played an important role in the removal of Hg⁰. Wang et al. (2018) treated six straw chars by Cl₂ non-thermal plasma method and found that the Hg⁰ removal efficiency increased from 10% to over 80% after the treatment. For example, as shown in Table 4, the Hg⁰ removal capacity of T6Cl was more than 36 times than that of T6 (tobacco straw). The improved results could be ascribed to the generated C–Cl groups on the samples, which served as activated sites for Hg⁰ removal.

Fly ash-based sorbents

Many investigators (Wang et al. 2016b; Hower et al. 2010) have identified fly ash (FA), a by-product of coal combustion as a promising alternative to activated carbon (AC) due to its very low cost and abundance. Related studies indicated that the fly ash has the ability to oxidize and adsorb Hg⁰ in flue gas because of the presence of some oxides such as CaO, TiO₂, Fe₂O₃, CuO, Al₂O₃, and unburned carbon as part of its composition (Borderieux et al. 2004; Guo et al. 2010; Dunham et al. 2003). However, compared with activated carbon (AC), the Hg⁰ removal performance of the fly ash is relatively poor (Cao et al. 2009) and requires some physical and chemical modification methods including the use of halogens and metal oxides to improve its capacity for Hg⁰ (Zhao et al. 2010; Bisson et al. 2013). The modification conditions and Hg⁰ removal capacities of raw fly ash and modified sorbents are summarized in Table 5.

Table 5 Reaction conditions and Hg⁰ removal performance of raw fly ash and modified sorbents

It is well known that the compositions of fly ash played an important role in Hg⁰ removal (Wang et al. 2016a; Yang et al. 2016e, 2017a, c). Wang et al. (2016a) investigated the Hg⁰ removal mechanism and performance of fly ash, and they found that the fly ash had a 60% Hg⁰ removal efficiency in simulated flue gas and that the presence of TiO₂, Fe₂O₃, and Al₂O₃ provided better improvement in performance compared to CaO and MgO and Al₂O₃. Furthermore, it was demonstrated that the reaction process of heterogeneous oxidation on fly ash followed an Eley–Rideal mechanism, with Fe₂O₃ considered as one of the active components on fly ash for Hg⁰ removal (Yang et al. 2017a). Yang et al. (2016e, 2017c) reported Hg⁰ removal at 100 °C of 89.5% for Fe₂O₃ and investigated the Hg⁰ reaction mechanism on its surface in the presence of HCl, suggesting that the main reaction process was in accordance with: Hg⁰ → FeHgCl(s) → HgCl₂.

Halogens modification is considered to be an effective method to enhance the adsorption and oxidation of Hg⁰ (Zhang et al. 2017c). Zhang et al. (2014b) compared three different halogenated fly ashes in an entrained flow reactor and found that the fly ash modified by HBr exhibited better Hg⁰ removal ability as compared with metal halogens such as CaCl₂ and CaBr₂. Song et al. (2014) and Zhang et al. (2015f) also studied the Hg⁰ removal performance of HBr-modified fly ash in a fixed-bed reactor and an entrained flow reactor, respectively, and found significant improvement over unmodified fly ash. Zhang et al. (2017d) further investigated the effect of NO on Hg⁰ removal of HBr-modified fly ash in an entrained flow reactor and demonstrated that the introduction of NO improved the Hg⁰ removal performance of the fly ash, as a result of the reaction of NO and HBr in the presence of O₂. Li et al. (2013b) developed some halogen-modified fly ash by an impregnation method and found that compared to bromine and chlorine, the iodine-modified fly ash exhibited better Hg⁰ removal performance. It has been shown that both the metal ions and halogen ions contained in metal halogens acted as active sites and improved the Hg⁰ removal performance (Xu et al. 2013; Yang et al. 2016b, c, d). Xu et al. (2013) suggested that metal halogens, such as CuBr₂, CuCl₂, and FeCl₃ loaded on fly ash, promoted the removal of Hg⁰ from flue gas due to the positive role played by Cu²⁺ and Fe³⁺ cations. Yang et al. (2016b, c, d) developed a novel magnetic catalyst (CuCl₂-MF) based on CuCl₂ modified fly ash and found that the fly ash modified by 6% CuCl₂ achieved 90.6% Hg⁰ removal from flue gas at 150 °C. In addition, when HCl was introduced into the flue gas, the CuCl₂-MF catalyst exhibited an excellent resistance to SO₂ poisoning. XPS and EPR analyses suggested that Cu and Cl adsorption sites were involved in the Hg⁰ removal process. As shown in Fig. 9, the reaction between CuCl₂ and Hg⁰ appears cyclical in the presence of HCl and O₂. In addition, the regeneration performance of CuCl₂-MF catalyst was also studied. The results of this study indicated that the regenerated catalyst showed a relatively higher Hg⁰ removal capacity after thermal desorption and restoration of HCl and O₂.

Fig. 9

(reproduced with permission from Yang et al. 2016b, c, d)

Reaction process for Hg⁰ removal over CuCl₂-MF sample in the presence of HCl and/or O₂. The Hg⁰ removal over CuCl₂-MF samples is attributed to the synergistic role of both Cu and Cl atoms in CuCl₂, and the reaction between CuCl₂ and Hg⁰ appears cyclical in the presence of HCl and O₂

In recent years, some metal oxides (e.g., manganese oxides, cobalt oxides, and iron oxides) have been used to modify fly ash (FA) before its use to remove Hg⁰ from flue gas. Xing et al. (2012) modified fly ash by manganese oxides and iron oxides and found that modification with Mn and Fe increased the Hg⁰ removal efficiency. In particular, the Mn(2)-Fe(3)-FA samples exhibited the highest Hg⁰ removal efficiency compared with raw fly ash in the presence of O₂. The XPS analysis indicated that the Mn⁴⁺ and Fe³⁺, which served as active sites, could react with absorbed Hg⁰ to form HgO, thereby promoting the Hg⁰ removal. Xu et al. (2014b) synthesized Co-modified fly ash by a wet impregnation method and found that the sample impregnated with 9 wt% Co was very effective in Hg⁰ capture, attributable to the presence of Co₃O₄ and its reaction with Hg⁰ to form mercury oxides as shown in Fig. 10.

Fig. 10

(reproduced with permission from Xu et al. 2014b)

Reaction mechanism of Hg⁰ removal. The generated Co³⁺ on the surface of fly ash sorbents significantly promoted Hg⁰ removal and O₂ played a crucial role in oxidation reactions

Mineral material-based sorbents

Mineral material-based sorbents have been widely studied for the treatment of Hg⁰ removal in flue gas due to its low prices, abundance, and environmentally benign nature. However, various nature mineral sorbents such as zeolites, clays, and bentonites have a relatively poor capacity for Hg⁰ removal, prompting the use of some agents such as halogens, metal halogens, and metal oxides under suitable modification conditions summarized in Table 6, to improve their effectiveness.

Table 6 Reaction conditions and Hg⁰ removal performance of mineral material-based sorbents

Zeolites are regarded as promising sorbents and good alternatives to activated carbon, due to their unique framework structures, favorable cation exchange properties and low cost (Wang et al. 2015b; Du et al. 2014; Chiu et al. 2014; Qi et al. 2015; Fan et al. 2012a, b). Wang et al. (2015b) investigated some zeolite sorbents for Hg⁰ removal performance and demonstrated an efficiency of over 75% within 480 min at 100 °C. Du et al. (2014) developed CuCl₂-impregnated zeolites, and in general, found their over 80% Hg⁰ removal performances were comparable to those of activated carbons. Chiu et al. (2014) further studied the effect of CuCl₂ modification on the physicochemical properties zeolites and their resulting effectiveness in the simultaneous removal of Hg⁰, NO, and SO₂. The results of this study showed that the introduction of CuCl₂ decreased the pore volume and total surface area, and the CuCl₂-modified samples exhibited higher Hg⁰ removal performances compared with their unmodified equivalents under both simulated flue gas and N₂ atmospheres. Qi et al. (2015) investigated the performance of FeCl₃-modified zeolites (FeCl₃-HZSM-5) and demonstrated that the improved Hg⁰ capture efficiency obtained was due to higher surface areas and the surface-generated active Cl species. Metal oxides, with strong active and thermal stabilities, have been used as modification additives to improve the Hg⁰ removal capacity of sorbents. Fan et al. (2012a, b) studied the Hg⁰ removal from flue gas using both CeO₂- and CuO-modified zeolites in a laboratory-scale fixed-bed system. They found that not only did they improve Hg⁰ removal compared to raw zeolite (HZSM-5), but the CeO₂/HZSM-5 and Cu/HZSM-5 also exhibited higher activities for NO removal.

Clay has been used as sorbent for Hg⁰ removal due to its high abundance, good thermal stability, and layered structure. Cai et al. (2014) studied Hg⁰ removal using KI- and KBr-modified clays in simulated flue gas conditions. The results indicated that the modification of KI and KBr significantly enhanced the Hg⁰ removal, and the KI-modified clays had better Hg⁰ removal capacity compared with KBr-modified clays. Based on these results, Cai et al. (2014) and Shen et al. (2015b) further synthesized KI-impregnated titanium-pillared clay (KI-Ti-PILC) for use to capture Hg⁰ in flue gas and found that the much better performance over the raw clay was due to its more developed mesopores and higher specific surface area. Also, some metal oxides such as CeO₂ and MnO_x have been employed as modification additives due to their higher oxidation activities for Hg⁰ capture (He et al. 2016a, b). He et al. (2016a) developed a CeO₂-modified pillared clay sorbent via an impregnation method, and they found that the sorbent (15CeTPC) showed a high oxidation activity of 88.2% for Hg⁰ in flue gas at 300 °C in the presence of 5% O₂. He et al. further synthesized Ce-MnO_x-modified pillared clay catalysts (Ce-MnO_x/Ti-PILC), which also showed excellent Hg⁰ removal performance (He et al. 2016b).

Bentonite, a type of clay mineral composed of montmorillonite, has been also used for treatment of Hg⁰ in flue gas (Li et al. 2014b; Ding et al. 2012; Shao et al. 2016). Li et al. (2014b) synthesized the ammonium bromide-modified bentonite via an impregnation method and found that the modification enhanced the Hg⁰ removal efficiency. Ding et al. (2012) also synthesized a number of bentonite-based sorbents modified by CuCl₂, NaClO₃, KBr, or KI, and reported that the KI-modified and CuCl₂-modified samples achieved better performance of about 90% Hg⁰ removal at 120 °C. Furthermore, Shao et al. (2016) synthesized a novel KI-modified bentonite-starch sorbent (B-S-I) and found it to be more effective for Hg⁰ removal than that of KI-modified bentonite sorbent (B-I). It was suggested that the starch–iodine complex formed by the reaction of iodine and starch promoted Hg⁰ removal via the ability to release I₂, which could react with Hg⁰ to form iodated mercuric compounds, thus promoting Hg⁰ removal, as shown in the reaction mechanism depicted in Fig. 11.

Fig. 11

(reproduced with permission from Shao et al. 2016)

Reaction mechanism of Hg⁰ removal by B-S-I. The starch-iodine complex formed by the reaction of iodine and starch promoted Hg⁰ removal via the ability to release I₂, which could react with Hg⁰ to form iodated mercuric compounds, thus promoted Hg⁰ removal

Other novel Hg⁰ removal technologies

In addition to the catalysts and sorbents extensively discussed above, other novel capture processes for Hg⁰ in flue gas systems involving photocatalysis, plasma catalytic oxidation, and microwave catalytic oxidation under various modification conditions have been developed as attractive alternatives to conventional technologies, and are summarized in Table 7 (Wu et al. 2015a; Zhuang et al. 2014; Zhang et al. 2016a; An et al. 2016; Yang et al. 2012a, b; Liu et al. 2015a; Wei et al. 2015a, b).

Table 7 Reaction conditions and Hg⁰ removal performance of novel removal methods

Photocatalytic oxidation has been considered as a promising technology to remove Hg⁰ in flue gas because it is a green process with superior oxidation ability (Wu et al. 2015a; Zhuang et al. 2014; Zhang et al. 2016a). Wu et al. (2015a) synthesized TiO₂ hollow sphere via a hydrothermal method and evaluated its performance for Hg⁰ in flue gas under the irradiation of ultraviolet lamp. The results indicated that the sample showed an excellent photocatalytic oxidation for Hg⁰ oxidation with a conversion of 82.75%. Zhuang et al. (2014) developed carbon-modified TiO₂ nanotubes by a hydrothermal method, which achieved an effective Hg⁰ removal performance under the white light LED lamp irradiation. Zhang et al. (2016a) synthesized some BiOX (X denotes Cl, Br, and I) photocatalysts via a sample coprecipitation method for use to capture Hg⁰ in flue under fluorescent light. The results of the study indicated that compared with BiOCl, BiOBr, BiOI exhibited much better Hg⁰ removal capacity. It was suggested that the presence of a hole (h⁺) and ion (·O²⁻) played key roles in BiOBr reaction system, while for BiOI reaction system, the generated I₂ might be the main species for Hg⁰ oxidation.

The plasma catalytic oxidation technology has obtained much attention due to its ability to oxidize Hg⁰ via the generation of active species such as O₃, OH, HO₂, and O (An et al. 2016). Yang et al. (2012a, b) studied the oxidation of Hg⁰ using TiO₂ power via non-thermal plasma coupled with photocatalysis and found that the combined plasma-photocatalysis system resulted in a synergistic effect, promoting improved Hg⁰ oxidation performance. Liu et al. (2015a) investigated the Hg⁰ removal performance of SiO₂, TiO₂, and SiO₂ or TiO₂ supported transition metal oxide catalysts at low temperatures using a plasma-catalyst reactor. The results showed that while the non-thermal plasma could effectively enhance the Hg⁰ oxidation, the presence of Mn/TiO₂ catalysts resulted in the highest Hg⁰ removal efficiency of about 99% under a SED of 2.3 ± 0.3 J/L.

Wei et al. (2015a, b) synthesized Mn/γ-Al₂O₃ and Mn/zeolite catalysts via an incipient wetness impregnation method for microwave catalytic oxidation of Hg⁰ in flue gas under the integrated ozone atmosphere. They reported more than 90% Hg⁰ removal efficiency in the integrated microwave and ozone system, and also attributed the higher efficiency to the presence of ozone and large amounts of free radicals (O, HO₂, and OH) and their strong ability to oxidize Hg⁰.

Proposed mechanism for the heterogeneous oxidation of elemental mercury

Typically, the Hg⁰ can be oxidized to Hg²⁺ by the heterogeneous reactions or/and homogeneous reactions. The mechanistic aspects of Hg⁰ removal using sorbents and catalysts have been extensively studied by numerous investigators (Zhao et al. 2015d; Chen et al. 2016; Zhang et al. 2017d; Xu et al. 2014b). It is well known that sorbents and catalysts promote heterogeneous reactions, which are faster reaction rate than homogeneous reactions (Presto and Granite 2006). The Deacon process, Eley–Rideal, Langmuir–Hinshelwood, and Mars–Maessen are among some of the mechanistic approaches, which have been employed to explain and quantify the heterogeneous oxidation of Hg⁰ in flue gas.

The Deacon reaction

The mechanism assumes that the process by which Cl₂ (or Cl atom) can be generated by the reaction of HCl and O₂ or air at high temperature (e.g., 300–400 °C) as in Eq. (1) is the main pathway for Hg⁰ catalytic oxidation in flue gas.

$$4{\text{HCl}}_{{({\text{g}})}} + {\text{O}}_{{2({\text{g}})}} \to 2{\text{Cl}}_{{2({\text{g}})}} + 2{\text{H}}_{2} {\text{O}}$$

(1)

The Deacon reaction could produce a large amount of Cl₂ in the presence of some sorbents and catalysts, thereby promoting Hg⁰ removal (He et al. 2016a). Based on the results of Xu et al. (2014a) and Du et al. (2014), the Deacon reaction may be the main pathway for Hg⁰ removal over the Cu-based sorbents and catalysts in HCl and O₂ atmosphere. Zhao et al. (2017b) suggested that the different reaction temperature ranges have significant effects on the Deacon reaction. As shown in Fig. 12, Hg⁰ could be adsorbed by Mo or Ag on the surface of catalyst to form the Mo-Hg or silver amalgam at low reaction temperature and then combine with the active Cl species produced by a reaction of HCl and Ag-Mo/V-Ti to form soluble and adsorbable HgCl₂, namely the Semi-Deacon reaction. When the reaction temperature is in the range of 350–450 °C, the generated Cl₂ could begin to react with the gaseous Hg⁰ to form HgCl₂, namely the Full Deacon reaction. Chen et al. (2014) by employing the Deacon mechanism explained that the gaseous O₂ was firstly adsorbed and activated by Ru_cus to generate the active O species, and then the produced active O species reacted with HCl to generate Cl₂. The reaction pathways can be described as follows:

Fig. 12

(reproduced with permission from Zhao et al. 2017b)

Reaction mechanism for Hg⁰ removal over Ag-Mo/V-Ti at different reaction temperatures. The Mo-Hg or silver amalgam formed by the reaction of adsorbed Hg⁰ and Mo or Ag can react with active Cl to produce HgCl₂ at low reaction temperature, namely, the Semi-Deacon reaction. When the reaction temperature is in the range of 350–450 °C, the generated Cl₂ could begin to react with the gaseous Hg⁰ to form HgCl₂, namely the Full Deacon reaction

$$2{\text{Ru}}_{\text{cus}} + {\text{O}}_{2} \to 2{\text{Ru}}_{\text{cus}} - {\text{O}}^{*}$$

(2)

$${\text{Ru}}_{\text{cus}} - {\text{O}}^{*} + {\text{HCl}} \to {\text{Ru}}_{\text{cus}} - {\text{OH}} - {\text{Cl}}^{*}$$

(3)

$$4{\text{Ru}}_{\text{cus}} - {\text{OH}} - {\text{Cl}}^{*} \to 4{\text{Ru}}_{\text{cus}} - {\text{Cl}}^{*} + 2{\text{H}}_{2} {\text{O}} + {\text{O}}_{2}$$

(4)

$$2{\text{Ru}}_{\text{cus}} - {\text{Cl}}^{*} \to 2{\text{Ru}}_{\text{cus}} + {\text{Cl}}_{2}$$

(5)

$$2{\text{Ru}}_{\text{cus}} - {\text{Cl}}^{*} + {\text{Hg}}^{0} \to 2{\text{Ru}}_{\text{cus}} + {\text{HgCl}}_{2}$$

(6)

The Eley–Rideal mechanism

In general, this mechanism assumes surface reaction involving physically adsorbed reactant (A) and chemisorbed reactant (B) or reactant (A) in gas phase and chemisorbed reactant (B), and vice versa. That is, the adsorbed active species, such as HCl, could react with the gas-phase or weakly adsorbed Hg⁰ to form Hg²⁺ (Zhao et al. 2014; Wang et al. 2016a). The Eley–Rideal reaction mechanism has been employed in the study of Hg⁰ oxidation over selective catalytic reduction (SCR) catalysts in the presence of HCl (Yang et al. 2017d; Zhang et al. 2015d). It has been suggested that in the Hg⁰ oxidation over SCR catalysts, the HCl was firstly adsorbed on the surface of the catalyst to generate active Cl sites, which could react with the gas-phase or weakly adsorbed Hg⁰ to produce HgCl₂ (Wang et al. 2013). The specific reaction mechanism can be described as follows:

$$4{\text{HCl}}_{{ ( {\text{g)}}}} + {\text{O}}_{2} \to 2{\text{H}}_{2} {\text{O}} + 4{\text{Cl}}_{{ ( {\text{ad)}}}}^{*}$$

(7)

$${\text{Hg}}_{{ ( {\text{g)}}}}^{0} + {\text{Cl}}_{{ ( {\text{ad)}}}}^{*} \to {\text{HgCl}}_{{ ( {\text{ad)}}}}$$

(8)

$${\text{HgCl}}_{{ ( {\text{ad)}}}} + {\text{Cl}}_{{ ( {\text{ad)}}}}^{*} \to {\text{HgCl}}_{{ 2 ( {\text{ad)}}}}$$

(9)

$${\text{HgCl}}_{{ 2 ( {\text{ad)}}}} \to {\text{HgCl}}_{{2({\text{g}})}}$$

(10)

Similarly, the reaction of H₂S and Hg⁰ also followed the Eley–Rideal reaction mechanism (Zhou et al. 2013). Related results (Hou et al. 2014a; Li et al. 2014a; Han et al. 2016; Yue et al. 2015) suggested that H₂S could be oxidized by some active species to form adsorbed active sulfur species (S_ad), which could further reacts with the gas-phase Hg⁰ to generate HgS. The reaction mechanism can be described as follows:

$${\text{H}}_{2} {\text{S}}_{{({\text{g}})}} + {\text{O}}^{*} \to {\text{S}}_{{ ( {\text{ad)}}}} + {\text{H}}_{2} {\text{O}}$$

(11)

$${\text{S}}_{{ ( {\text{ad)}}}} + {\text{Hg}}^{0} \to {\text{HgS}}$$

(12)

The Langmuir–Hinshelwood mechanism

Langmuir–Hinshelwood (L–H) mechanism (also known as Langmuir–Hinshelwood–Hougen–Watson (LHHW) in chemical reaction engineering) generally employs Langmuir’s adsorption isotherm for chemisorption and assumes equilibrium adsorption and that the surface reaction is controlling. It has been used extensively to describe the bimolecular reaction between two species adsorbed on the surface of sorbents and catalysts (Zhao et al. 2016a; Liu et al. 2016a). Based on this reaction mechanism, the adsorbed Hg⁰ could react with some adsorbed oxidant species, such as HBr and HCl (Lim and Wilcox 2013; Song et al. 2014). The results of prior studies indicate that the Hg⁰ oxidation on the surface of some metal oxides-based sorbents and catalysts followed the Langmuir–Hinshelwood mechanism (Zhang et al. 2017b; Hou et al. 2014b; Huang et al. 2016). Jampaiah et al. (2015) and Wang et al. (2014) suggested that the Hg⁰ removal on the Mn/Ce catalysts could be described by the Langmuir–Hinshelwood mechanism, whereby the adsorbed Hg⁰ could react with adsorbed active species to form Hg²⁺ as in reactions in Eqs. (13–17). Negreira and Wilcox (2013) also obtained similar results in the oxidation of Hg⁰ over vanadia–titania selective catalytic reduction (SCR) catalyst. In addition, some investigators have indicated that the Hg⁰ oxidation on the surface of catalyst in the presence of SO₂ could also be explained by the Langmuir–Hinshelwood mechanism and suggested that the active species derived from SO₂ could react with adsorbed Hg⁰ to form HgSO₄ (Li et al. 2013a; Chiu et al. 2015; Zhang et al. 2016b).

$$2{\text{HCl}}_{{ ( {\text{g)}}}} + {\text{O}}^{*} \to 2{\text{Cl}}_{{ ( {\text{ad)}}}}^{*} + {\text{H}}_{2} {\text{O}}$$

(13)

$${\text{Hg}}_{{({\text{g}})}}^{0} \to {\text{Hg}}_{{({\text{ad}})}}^{0}$$

(14)

$${\text{Cl}}_{{({\text{ad}})}}^{*} + {\text{Hg}}_{{({\text{ad}})}}^{0} \to {\text{HgCl}}_{{({\text{ad}})}}^{*}$$

(15)

$${\text{HgCl}}_{{({\text{ad}})}}^{*} + {\text{Cl}}_{{({\text{ad}})}}^{*} \to {\text{HgCl}}_{{2({\text{ad}})}}$$

(16)

$${\text{HgCl}}_{{2({\text{ad)}}}} \to {\text{HgCl}}_{{2({\text{g}})}}$$

(17)

The Mars–Maessen mechanism

The Mars–Maessen mechanism had been considered by numerous investigators as the most plausible mechanism for Hg⁰ oxidation on the surface of metal oxide-based sorbents and catalysts (Wu et al. 2015b; Xu et al. 2016b; Qu et al. 2015). In this mechanism, the adsorbed Hg⁰ could react with the lattice oxygen to form a binary mercury oxide. The Hg⁰ oxidation mechanism on the Fe₂O₃–SiO₂ catalyst could be described by the reactions in Eqs. (18–22) (where M denotes Fe) (Tan et al. 2012c). Firstly, the gas-phase Hg⁰ is assumed to adsorb on the surface of catalysts to form adsorbed Hg⁰. Then the adsorbed Hg⁰ is oxidized by the lattice oxygen from metal oxides to form HgO. Finally, the consumed lattice oxygen could be regenerated and replenished by the gas-phase oxygen from flue gas. The oxidation of Hg⁰ by other metal oxides such as CuO_x, MnO_x, and CeO₂ could also be explained by the Mars–Maessen mechanism (Zeng et al. 2017; Chiu et al. 2017; Li et al. 2015e). Also, other investigators have used the Mars–Maessen mechanism to explain the oxidation of Hg⁰ over some multi-metal oxide-based catalysts (Zhang et al. 2015e, 2016d; Li et al. 2016b; Zhao et al. 2016c). Zhao et al. (2015c) suggested that the Hg⁰ oxidation on CeO₂–V₂O₅ catalyst surface followed the Mars–Maessen mechanism, where the synergistic effect between CeO₂ and V₂O₅ played an important role on the oxidation of Hg⁰. They proposed the plausible reaction pathways as in the Eqs. (18–22), and the reaction mechanism is illustrated in Fig. 13.

Fig. 13

(reproduced with permission from Zhao et al. 2015c

Reaction mechanism of Hg⁰ oxidation on CeO₂–V₂O₅ catalyst. The Hg⁰ oxidation on CeO₂–V₂O₅ catalyst surface followed the Mars–Maessen mechanism, where the synergistic effect between CeO₂ and V₂O₅ played an important role on the oxidation of Hg⁰)

$${\text{Hg}}_{{({\text{g}})}} \to {\text{Hg}}_{{({\text{ad}})}}$$

(18)

$${\text{Hg}}_{{({\text{ad)}}}} + {\text{M}}_{x} {\text{O}}_{y} \to {\text{HgO}}_{{({\text{ad}})}} + {\text{M}}_{x} {\text{O}}_{y - 1}$$

(19)

$${\text{M}}_{x} {\text{O}}_{y - 1} + 1/2{\text{O}}_{2} \to {\text{M}}_{x} {\text{O}}_{y}$$

(20)

$${\text{HgO}}_{{ ( {\text{ad)}}}} \to {\text{HgO}}_{{({\text{g}})}}$$

(21)

$${\text{HgO}}_{{({\text{ad}})}} + {\text{M}}_{x} {\text{O}}_{y} \to {\text{HgM}}_{x} {\text{O}}_{y + 1}$$

(22)

Summary, challenges, future research suggestions, and prospects

In this review, recent development on several catalysts and adsorbents for Hg⁰ heterogeneous oxidation removal, including mainly noble metal-based catalysts, non-noble metal-based catalysts (transition metal oxides and selective catalytic reduction catalysts), activated carbon-/coke-based sorbents, biochar-based sorbents, fly ash-based sorbents, mineral material-based sorbents and other novel catalysts, are extensively discussed. Some future research suggestions and potential directions for the development of green and cost-effective technologies are summarized here.

The noble metal-based catalysts have excellent Hg⁰ removal capacity and are generally regenerable and reusable to a large extent of use, but the very high costs and scarce sources greatly limited their developments and applications. Compared with noble metals, transition metal oxides and selective catalytic reduction (SCR) catalysts have several advantages such as much lower costs and more extensive sources. However, the catalytic activity for Hg⁰ of transition metal oxides and selective catalytic reduction catalysts is often relatively low. Besides, their Hg⁰ removal performance is greatly affected by the components of flue gas, such as halides, sulfides, vapors and alkali metal salts, and other heavy metals. The activity and stability of transition metal oxides and SCR catalysts for removing Hg⁰ still need to be improved significantly using doping and other modification methods that utilize precious metals, transition metals, and nonmetals (including mixed doping and modification of multiple components). In addition, possible poisoning or deactivation of transition metal oxides and SCR catalysts by mercury itself, also needs further future investigations.

Activated carbon injection (ACI) method has been proven as one of the effective ways for Hg⁰ removal from flue gas. However, the large activated carbon (AC)/Hg⁰ ratio and high operation costs have limited its development. The modification with various chemical reagents (e.g., halides, sulfurs, acids, alkaline and metal oxides) can significantly improve the Hg⁰ removal capacity of activated carbon, but also further increase the costs. Biochars, fly ash, and mineral materials are considered as the potential alternatives to activated carbon due to their much lower costs and more extensive sources. However, they have low Hg⁰ adsorption capacity due to the poor adsorption sites on their surface. To improve the effectiveness of these adsorbents, chemical reagents are also used to modify them by increasing active sites on the surface. Unfortunately, the leaking and secondary pollution of the modified chemical reagents used over these adsorbents have greatly hindered the development and practical applications. In recent years, various advanced oxidation processes have been widely applied in the field of flue gas purification. Therefore, exploring more green and clean modification methods, such as free radical-based advanced oxidation methods, should be considered an important future priority. However, there could a limitation of terrestrial biomass (e.g., the reduction of cultivated land area and the dispersity of biomass straw). But, the ocean contains a huge biomass resource, which could be utilized. Therefore, actively exploring the utilization of marine biomass resources such as all kinds of large algae and microalgae (e.g., using marine biomass to prepare biochars and activated carbon) could provide significant resources for human development.

At present, a large number of adsorbents have been developed, but most of these sorbents lack adequate recycling and regeneration capabilities, which greatly increased the costs of application, operation, and post-processing costs and related environmental issues due to solid waste treatment problems. Developing magnetically separable and renewable sorbents should be considered as an important research direction in the future. In addition, it is reported that most of the magnetic adsorbents are still difficult to be completely separated from magnetic impurities, for example, in coal fly ash due to similar magnetic properties. Therefore, in order to completely separate the magnetic adsorbents from the magnetic impurities successfully, significant improvements in multistage magnetic field separation processes are desirable and should be pursue vigorously in future research. The separation of sorbents from fly ash can be solved by the magnetic property of sorbent materials. Therefore, magnetic properties of magnetic adsorbents could also be effectively regulated through various preparation and modification methods, and based on magnetic differences, the separation problem of adsorbents could be more effectively addressed.

In addition, other technologies such as photocatalytic oxidation, plasma catalytic oxidation, microwave catalytic oxidation, and covalent organic frameworks (COFs) adsorption oxidation developed to remove Hg⁰ in flue gas have demonstrated good Hg⁰ oxidation performance. However, some problems limiting process development such as high investment and operating costs, low reliability, and stability of systems/devices, low activity and anti-poisoning ability of catalysts/adsorbents and others, need to be addressed before large-scale deployment. Also, technologies utilizing catalytic or photocatalytic membrane systems should be exploited as they could remarkably reduce the demand of oxidant (by improving its retainability) and have better efficiencies for Hg⁰ removal from flue gas.

Among the aforementioned catalysts, the selective catalytic reduction (SCR) catalyst is considered the most promising, with the greatest benefit of providing simultaneous removal of NO_x and Hg⁰ from flue gas, and reducing investment and operating costs of existing SCR denitrification device as it could be retrofitted into its current configuration. Furthermore, research initiatives into the development of sustainable adsorbents, such biochars-based adsorbents, as potential alternatives to conventional activated carbon, should be intensified because of their very low costs and readily available renewable sources.

Conclusion

Regulatory requirements and increased public concerns regarding mercury elevation levels and persistence in the atmosphere have stimulated worldwide research efforts to develop technologies for mercury emission control. In particular, the heterogeneous catalytic oxidation and adsorption of Hg⁰ from flue gas has recently been an area of major focus because of its important scientific and practical significance. The catalysts and/or adsorbents are the key to the success of the heterogeneous oxidation removal technologies for Hg⁰ from flue gas. This review provides the state-of-the-art knowledge of the chemistry and the fundamental mechanistic aspects of gas–solid heterogeneous oxidation and adsorption processes for the removal of Hg⁰ from the flue gas systems. It evaluates the performance and economic viability of various catalysts/sorbents for Hg⁰ removal. However, this review also reveals a number of areas in which additional research are needed. These include the development of more resistant, regenerable, effective, and versatile catalysts and adsorbents; and engineering-based research such as cost–benefit analysis, techno-economic modeling and optimization of the heterogeneous catalytic and adsorptive processes for mercury removal from flue gas systems. It is hoped that this review has stimulated thinking beyond the cases presented and should spur further research needed to further the development of greener, sustainable, and more cost-effective technologies to remove Hg⁰ from flue gas.