The influence of flue gas components and activated carbon injection on mercury capture of municipal solid waste incineration in China

Accepted Manuscript The influence of flue gas components and activated carbon injection on mercury capture of municipal solid waste incineration in China Guoliang Li, Qingru Wu, Shuxiao Wang, Zhijian Li, Hongyan Liang, Yi Tang, Minjiang Zhao, Lei Chen, Kaiyun Liu, Fengyang Wang PII: DOI: Reference:

S1385-8947(17)30851-3 http://dx.doi.org/10.1016/j.cej.2017.05.099 CEJ 16997

To appear in:

Chemical Engineering Journal

Received Date: Revised Date: Accepted Date:

13 March 2017 13 May 2017 15 May 2017

Please cite this article as: G. Li, Q. Wu, S. Wang, Z. Li, H. Liang, Y. Tang, M. Zhao, L. Chen, K. Liu, F. Wang, The influence of flue gas components and activated carbon injection on mercury capture of municipal solid waste incineration in China, Chemical Engineering Journal (2017), doi: http://dx.doi.org/10.1016/j.cej.2017.05.099

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

The influence of flue gas components and activated carbon injection on mercury capture of municipal solid waste incineration in China Guoliang Lia, b, , Qingru Wua, b, &, Shuxiao Wanga, b, Zhijian Lia, b, Hongyan Lianga, b, Yi Tanga, b, Minjiang Zhaoa, b, Lei Chena, b, Kaiyun Liua, b and Fengyang Wanga, b a

State Key Joint Laboratory of Environment Simulation and Pollution Control, School of Environment, Tsinghua

University, Beijing 100084, China b

State Environmental Protection Key Laboratory of Sources and Control of Air Pollution Complex, Beijing

100084, China

Abstract This study investigates the influence of flue gas components and original/modified activated carbon (AC) injection on mercury removal to obtain a suitable mercury removal method for typical municipal solid waste (MSW) incineration plants in China. The field test was conducted to research the mercury emission characteristics of a typical plant. The result shows that the mercury concentration of before and after air pollution control devices (APCDs) is 91.28±47.98 and 25.67±18.13 μg·m-3, respectively, and the average mercury removal efficiency is 71.9%. Based on actual APCDs condition, the flue gas components (O2, SO2, HCl and NO), fly ash, activated carbon and NH4Cl-modfied AC were used to further remove the mercury of flue gas. The O2 and HCl slightly promoted mercury removal while the SO2 and NO addition decreased mercury concentration from 50μg·m-3 to 35.8 and 13.7μg·m-3, respectively. The increase of AC injection amount failed to decrease emission concentration to desired value and imposed heavy financial burden to plants. The method of 2.5 wt.% Cl-modified AC addition reduced mercury concentration to below 5 μg·m-3 in various 

These authors contributed equally to this work and should be considered as co-first authors  Corresponding author: Professor Shuxiao Wang. E-mail: [email protected] Fax: +86 1062773650 1

flue gas condition and did not need to change any existing APCDs or operating parameters, which was considered as the optimal method for mercury removal in MSW incineration. Reaction mechanism of SO2, NO and Cl-modified AC was explored by TPD, XPS and in stitu Diffuse Reflectance Fourier Transform (DRIFT). The SO2 and NO formed effective active groups of SO3-M and NO2-M, and their reaction mechanism follows Langmiur-Hinshelwood mechanism and Eley–Rideal mechanism, respectively. The Cl modification effectiveness was ascribed to C-Cl group formation.

Keyword: mercury removal; municipal solid waste incineration; flue gas, activated carbon

1. Introduction The pollution of mercury, in both inorganic and organic forms, is an important public health and environmental concern due to its persistent, toxic and bioaccummulative properties [1]. The mercury emission reduction from various anthropogenic sources is increasingly attracting public attention. Municipal solid waste (MSW) incineration is one of the five major mercury emission sources in Minamata Convention [2]. In China, annual growth rate of burning capacity is 23.5% in 2008-2013 and 20-30% of MSW was treated by incineration [3]. The capacity increase of MSW incineration possibly enhanced the mercury emission. The incineration temperature can reach 850-1200 oC, at which the mercury all releases into gas phase in the form of Hg0. The gas mercury was captured via subsequent APCDs. In China, the typical MSW incineration plants use the APCDs combination: a dry/semidry scrubbing system for removing acidic gases, an activated carbon injection system for capturing dioxins, and a fabric bag filter. This APCDs combination shows over 60% mercury removal efficiency [4, 5]. According to previous studies, the average mercury concentration of stack outlet was at the range of 9.5-26.4 μg·g-1 in the MSW incineration plants of Guangzhou, Chongqing, and Wuhu 2

[6], which is far higher than 0.05-4.56 μg·m-3 in Japan [7], 1.96-4.71 μg·m-3 in Korea [8] and 3.7 μg·m-3 in US [9]. Moreover, individual monitoring data of some MSW incineration plants was very unstable and varied in the range of 8.7-133.6 μg·g-1 [10]. This result indicates that it is necessary to further investigate more stable pollution control methods for mercury capture. With a view to the fact that the APCDs transformation and upgrading needs high fees, reducing mercury emission via slightly operational adjustment under the condition of existing equipments will be better accepted by MSW incineration plants. The increase of the oxidable components of incineration flue gases, such HCl, SO2 and NO, is a commonly applied method [11]. Nishitani’s study indicated that the proportion of HgCl2 to total Hg increased with the increasing of HCl concentration [12]. The study of Wang also showed that the HCl addition improved the mercury removal performance of coal-fired fly ash [13]. The NO can transform into NO2 and then oxidize Hg0 into bivalent mercury [14-16]. The Hg2+ was easier to remove by the wet flue gas desulfurization devices (WFGDs). Previous studies suggested that the presence of SO2 and O2 in the flue gas led to mercury oxidation and adsorption on the sorbents or fly ash [17-20]. The objects of these studies are coal-fired flue gases and fly ash while few studies focus on the mercury emission reduction of MSW incineration. Compared with coal-fired plant, the APCDs of MSW incineration do not contain WFGDs in China, thus the mercury removal efficiency of APCDs in MSW incineration was possibly not as effective as in coal-fired incineration. Moreover, the flue gas of MSW incineration includes higher HCl content and thus the fly ash usually included higher Cl content (20%) [21, 22], so the HCl addition possibly fails to effectively promote mercury removal as coal-fired flue gas. It is necessary to research whether the elemental mercury can be oxidized by acidic gases and then captured on fly ash or AC. Another effective method of mercury abatement is activated carbon (AC) injection. Though AC was wildly used to clean flue gas of MSW incineration, few plants take mercury emission control into consideration in the processing of AC injection [23, 24]. 3

It is necessary to investigate the relation between AC injection amount and mercury abatement effectiveness. In addition, the modified reagents shows great promotional effect on the removal efficiency of AC in coal-fired flue gas, such as sulfur, halogen, etc. [25, 26]. Ie’ s study showed that the sulfur-impregnated AC had great performance for HgCl2 removal via simulation experiment of MSW incineration flue gas [27]. But this study failed to consider the AC injection amount and the influence of acidic gas and fly ash. In our previous study, the NH4Cl modification greatly promoted the mercury adsorption capacity of sorbents and the acidic gases were also beneficial to the mercury removal [15, 28]. The NO, SO2 and HCl have high content in flue gas of MSW incineration, hence the Cl-impregnated AC is possibly suitable to mercury removal in MSW incineration industry. This study investigated the influence of flue gas components and original/modified AC injection on mercury removal to find a stable mercury controling method in existing APCDs of typical MSW incineration in China. In order to reflect mercury emission characteristics, this study conducted field test to the mercury concentration of a typical MSW incineration plant in China. Meanwhile, the fly ash and AC were sampled for further simulation experiments of mercury removal. According to the actual component of flue gas, the effect of O2, HCl, NO and SO2 was investigated on the mercury removal. Moreover, the effect of AC injection amount and NH4Cl-modified AC is also investigated on the mercury capture. The influence mechanism of acidic gases and AC was illustrated via TPD, in situ DRIFT and XPS analysis.

2. Material and Methods 2.1 Field test Field test was conducted on a 550 t·d-1 incineration line in a Chinese MSW incineration plant, which is located in Fujian Province. The air pollution control devices of flue gases contain: activated carbon injection, dry/semidry scrubbing 4

system and fabric filter. This is a typical device combination and its installing percent is over 90% in Chinese MSW incineration plants. The detailed description of the tested plant, sampling procedures, analysis methods, quality assurance and quality control (QA/QC) were presented in the supporting information. The flue gases before and after pollution control facilities were sampled and analyzed with the Ontario Hydro Method (OH method) [29]. The solid samples (fly ash, activated carbon) were also sampled simultaneously with the flue gas samples. 2.2 Solid sample preparation This study contains four solid samples: original fly ash (FA), heated fly ash (FAH), original activated carbon (AC) and modified activated carbon (ACNCl5). The samples were prepared as follows: In order to remove adsorbed mercury on fly ash, the fly ash sampled in 2.1 was heated to 850 oC and remained for 1 h. The original and heated fly ashes were denoted as FA and FAH, respectively. The original activated carbon sampled in 2.1 was mixed with 1 wt.% NH4Cl solution in a ratio of 1g : 5ml. The mixture was stirred for 12 h and then dried at a water bath at 80 oC until the water vanished. The mixture was further dried at an oven for 12 h at 105 oC. The original and modified activated carbons were denoted as AC and ACNCl5, respectively. 2.3 Characterization techniques The proximate analysis of fly ash and AC was measured according to the method of GB/T212-2008. The ultimate analysis of fly ash and AC was measured by Elementar EA3000 (LEEMAN, China) and X-ray diffraction (XRD). The scanning electron microscope (SEM) was used to measure the surface image of fly ash by ProX (Phenom, Netherlands). The element (Cl, N, O, S and Hg) valence state was analyzed by X-ray photoelectron spectroscopy (XPS) by Axis Ultra DLD (Kratos Analytical 5

Ltd, Britain) with Mg X-ray source. All binding energies were calibrated with the C 1s peak at 284.6 eV. The in stitu DRIFT experiments were performed on Nicolet 6700-FTIR. The samples were finely ground and placed in the sample cell. The feed gas streamed into the cell at 100 mL·min-1. The feed gas contains N2, 6% O2, 500 ppm SO2 or 100 ppm NO. The background spectra were collected after holding at one desired temperature for 30 min. In all cases, the IR spectra were recorded over 32 scans at resolution of 4 cm-1. 2.4 Mercury adsorption testing The scheme of the mercury adsorption testing used to elevate the performance of various sorbents for mercury adsorption was shown in Fig. S2. The testing devices contain an elemental mercury generator, a fixed-bed reactor, a temperature-controlled heating equipment, an online mercury analyzer and a gas flue cleaner. A 200 mL·min-1 N2 flow, serving as carrier gas, passed through the permeation tube to generate the Hg0 vapor. The permeation tube within a U-type glass tube was placed in a temperature-controlled thermostat water bath to remain the Hg0 concentration at 50.0 ± 2 µg·m-3. The temperature-controlled heating equipment was used to heat the quartz fixed bed at 180 oC. According to actual conditions of MSW incineration, the flue gases contain 6% O2, 1000 ppm SO2, 300 ppm HCl, 100-300 NO and balance gas N2. The inlet total gas flow kept at 1000 mL·min-1. The 400 mg sample was placed in the adsorption zone of the fixed bed. The whole gas line was kept warm to avoid the Hg0 condensation. The mercury concentration at the inlet and outlet of the reactor was online measured by a Mercury FreedomTM system (Thermo Fisher Scientific Inc.), which could monitor the concentrations of different gaseous mercury forms, including gaseous elemental mercury (Hg0), gasous oxidized mercury (Hg2+) and gaseous total mercury (Hgt). For the experimental error is inevitable, the Hg0 removal efficiency is the average of three parallel experimental data and the relative error of all parallel experimental data is less than 10 %. 2.5 Analysis of mercury compound in the solid samples 6

The temperature programmed desorption technique (TPD) was used to analyze the existing forms of mercury adsorbed on various sorbents. The TPD schematic and operation details were shown in Fig. S3. The TPD system contains two furnaces and a real-time Lumex Zeeman mercury analyzer (Luemx RA915+, Russsia). The first furnace was heated from room temperature (20 oC) to 700 oC at a constant rate of 2 o

C∙min-1. The second furnace was kept at 800 oC to transform all the mercury

compounds into vaporized elemental mercury. The mercury analyzer was used to measure the mercury concentration. In the TPD experiments, original fly ash was heated with 60 mL∙min-1 of N2 carrier. The different mercury compounds were vaporized at specific temperature in the first furnace, transformed into elemental mercury and then detected by the analyzer. The mercury concentration detected exhibits distinct or partially overlapped peaks which can be used to identify the mercury compound on basis of reference profiles.

3. Results and Discussion 3.1 Experiment section 3.1.1 Field Test

The mercury concentration and existing form ratio of MSW incineration flue gas before and after APCDs was shown in Fig. 1. The mercury concentration of flue gas before and after APCDs is 91.28 ± 47.98 and 25.67±18.13 μg·m-3, respectively. The average mercury removal efficiency of APCDs is 71.9%. Similar result was found in our previous study that the mercury concentration of flue gas is 24.1±6.0 - 26.4±22.7 μg·m-3 in MSW incineration plants of Wuhu and Chongqing. [30, 31]. As shown in Fig. 1, the percentage of Hg0, Hg2+ and HgP is 61.93%, 19.60% and 18.47% respectively before APCDs while that of Hg0, Hg2+ and HgP is 25.96%, 73.88% and 0.17% respectively after APCDs. This result reveals that almost all particulate mercury is removed by fabric filter and thus the key factor of mercury emission abatement is the transformation from Hg0 and Hg2+ to Hgp. According to Fig. 1, the 7

Hg0 is higher percentage and harder to capture than Hg2+ due to its higher volatility [26]. In order to verify the reliability and feasibility of mercury removal effectiveness of various sorbents, the Hg0 is selected as the capture object in the following experiment process. 3.1.2 Simulation experiment of mercury removal

Fig. 2 shows the influence of various components of flue gases on mercury removal capacity of different sorbents. In all simulated experiments, the ratio of sorbent to gas is 400 mg : 1 L, which is the same as in actual flue gas. As shown in Fig. 2A, only FAH is used as sorbent. The pure FAH addition briefly decreased mercury concentration from 50 μg·m-3 to 38.9 μg·m-3 and then immediately returned to initial concentration. Then 5% O 2 addition decreased the Hgt concentration to 40.4μg·m-3 and 1000 ppm SO2 addition further lowered it to 38.7μg·m-3, indicating that the O2 and SO2 promoted the oxidation and adsorption of mercury. On the contrary, the 300 ppm HCl addition did not have obvious stimulative role on mercury removal. This result is not consistent with previous studies that HCl was beneficial to mercury removal [13]. With experiment time reaching 180 min, 100 ppm NO addition decreased the Hgt concentration to 15.7 μg·m-3, suggesting NO addition distinctly facilitated mercury removal. As time went on, the Hgt concentration gradually increased to 24.1 μg·m-3, which is very close to mercury emission concentration (25.67μg·m-3) of actual flue gases in sampled plant. When all flue gas components (O2+SO2+HCl+NO+N2) were added to simulated gas at 220 min, the Hgt concentration did not change obviously whereas the Hg2+ concentration instantaneously went up. The SO2 and HCl introduction possibly competed adsorption sites with adsorbed Hg2+, resulting in much adsorbed Hg2+ release to gas phase again. In Fig. 2B, based on actual AC injection amount, the FAH is mixed with 2.5% AC, which is used to research the mercury removal performance under various flue gas components. The experiment result has no apparent distinction with pure FAH in different 8

acid gases condition. This result indicates that the 2.5% AC addition does not obviously promote the mercury removal. In order to further improve mercury removal efficiency, the AC content of sorbent increased to 25% in Fig. S4. The FAH+25%AC addition decreased the Hgt concentration from 50 μg·m-1 to 33.9 μg·m-3. The O2, SO2 and HCl also further improved mercury removal performance under FAH+25%AC and decreased the Hgt concentration to 35.8, 26.6 and 38.2 μg·m-3, respectively. The NO addition decreased the Hgt concentration to 9.8 μg·m-3. These results indicate that the increase of AC content could promote the mercury removal. However, the cost of 25% AC is so high that most plants cannot afford it. Moreover, the outlet mercury concentration (9.8 μg·m-3) is still relatively higher compared with the emission concentration (0.05-4.71 μg·m-3) in Japan, Korea and US [7-9]. According to the results of influence of various flue gas components, NO plays a more obvious role in promoting mercury removal. Further influence of increasing NO concentration is studied in Fig. 2C. The 100 ppm NO addition decreased the Hgt concentration to 24.8μg·m-3 and 200 ppm NO addition further decreased it to 14.3μg·m-3. However, when NO concentration rose to 300 ppm, the Hgt concentration decreased to 13.7μg·m-3. This implied that positive influence of NO addition concentration on mercury removal has a limitation and fails to further improve beyond the limitation. In order to further explore appropriate method of mercury removal in MSW incineration industry, the AC was impregnated with 5 wt.% NH4Cl to generate ACNCl5 using as mercury sorbent. As shown in Fig. 2D, the FAH mixed with 2.5% ACNCl5 is applied to mercury removal at various flue gases. Although the NH4Cl impregnation decreased the pore property, it increased the mercury removal performance [32]. When the flue gas was switched into sorbent, the total mercury concentration rapidly dropped to 5.0μg·m-3. The positive effect of O2 and SO2 addition on mercury removal in Fig. 2D is not as obvious as in Fig. 2A and 2B, indicating that mercury removal mechanism of ACNCl5 is different from that of fly ash. On the contrary, the HCl addition decreased the mercury concentration to zero. Moreover, the NO or all flue gases addition could also remove all mercury. This result implies that the 2.5% ACNCl5 addition can meet the 9

requirement of mercury removal in various conditions of actual flue gas. Furthermore, this method does not need extra pollution control devices or even to improve AC injection amount, because the existing APCDs in China, including activated carbon injection, dry/semidry scrubbing system and fabric filter, can absolutely meet the equipment requirement of this method. 3.1.3 Analysis of mercury compound in the solid samples

According to the result of field test, the captured mercury from flue gas main adsorbed on the fly ash, hence the analysis of mercury existing form on fly ash is beneficial to understanding adsorption mechanism in flue gas. As show in Fig. 3, the desorption curve of fly ash contains three characteristic peaks. The peak at 284 oC was ascribed to the Hg(NO3)2 [33]. The peaks at 314 oC and 530 oC were attributed to HgO and HgSO4, respectively [13, 34]. The percentage of Hg(NO3)2, HgO and HgSO4 is 43.69%, 47.85% and 8.46%, respectively. Though the flue gas has high HCl in MSW incineration, there is no HgCl2 or Hg2Cl2 peaks appearing at 130-240 oC [13, 33, 34]. 3.2 Sample Characterization 3.2.1 Proximate and ultimal analysis

As shown in Table 1, on proximated analysis, the major component of FA and AC is ash and fixed carbon based, respectively. According to ultimate analysis, the major elements of AC are C and O, whereas the C, H, O content of FA is relative low. N and S content of FA is relatively high, which is possibly attributed to nitrate and sulfate. This result is consistent with TPD experiment. 3.2.2 Elemental analysis of fly ash

10

The major chemical component of fly ash was analyzed by XRD and shown in Table 2. The original fly ash (FA) and heated fly ash (FAH) did not have obvious difference in chemical components. The CaO, SiO2, Al2O3 and Fe2O3 are the main metal oxides in fly ash. Al2O3, Fe2O3 and TiO2 are usually considered as effective active components for mercury oxidation [13, 35-37]. After high temperature treatment, the SO3 and Cl content of FAH slightly decreased compared with the FA, which is likely resulted from volatization of sulfate and volatile organic compounds under high temperature. 3.2.3 SEM analysis Fig. S5 shows the SEM image of fly ash. In Fig. S5A, most of the fly ash particles of various sizes are irregular in shape. This is very different from the fly ash spherical particles from coal combustion [36]. The image of FA is further enlarged into 25000 times in Fig. S5B. The surface of FA has a layer of salt and volatile. When fly ash was treated with high temperature, the salt and volatile disappeared on FA, which is consistent with XRD result. On the contrary, after mercury adsorption experiment, used FAH appears salt layer on the surface again, indicating that new compound produced during adsorption process. 3.2.4 In stitu DRIFT Result

Based on the result of simulation experiment, the NO and SO2 have good ability to promote mercury removal performance, hence the reaction mechanism of NO/SO2 with mercury is significative for mercury removal. The in situ DRIFT spectra were recorded to explore the NO and SO2 adsorption species on the fly ash. As show in Fig. 4A, the bands at 1603 and 1626 cm-1 appeared first at initial stage of 100 ppm NO and Hg0 addition. The peak at 1603 cm-1 was assigned to adsorbed NO2 on Lewis acid sites [38], and the peak at 1626 cm-1 was attributed to bridging nitrate. The chemical construction of 11

adsorbed NO2 and bridging nitrate is shown in Fig. 4A. As time went by, two new peaks at 1340 and 1457cm-1 appeared, which were ascribed to monodentate nitrate and ionic nitrate [39-41]. The two peaks possibly contributed to the formation of nitrate mercury on fly ash. According to the appearance order of various characteristic peaks, the NO first reacted with active oxygen of metal oxide to generate NO2 adsorbed on Lewis acid sites. Based on previous study result, the NO2 could oxidize the Hg0 to HgO and Hg(NO3)2. With the Hg(NO3)2 formation increasing, the new nitrate characteristic peaks at 1340 and 1457 cm-1. Fig. 4B shows IR spectra of SO2 adsorbed on fly ash. The peaks appearing at 1376 and 1344 cm-1 represents the surface sulfate and SO2 adsorbed on Lewis acid sites, respectively [42]. When experiment time reached 10 min, the bulk-like sulfation species are characterized by a broad band at approximately 1200 cm-1, indicating that much sulfate salt formed inside fly ash [43]. According to these results, the SO2 possibly first adsorbed on Lewis acid sites and then was further reacted with metal oxide to form sulfate. In this process, the Hg0 was also oxidized into HgSO4 in the presence of O2. The sulfate mercury possibly has high mobility from the surface to the bulk [44]. Therefore, as time went on, the bulk-like sulfate appeared at 1376 cm-1 since sulfate mercury spilled over from the surface sulfates to the bulk. 3.2.5 XPS Analysis Result

As shown in Fig. 5, the XPS was used to analyze the valence state of various elements before and after mercury experiments. In Fig. 5A, only one peak representing Si appears in the spectra of Hg 4f [45], which indicates the absence of mercury on fresh FAH. On the contrary, after the mercury adsorption testing in Fig. 5B, a new peak appears at 104.3 eV contributing ing to mercury in oxidized state (Hg2+) [46]. Both Fig. 5C and 5D shows sulfate peak at 169.2 eV in the S 2p spectra of fresh and used FAH [47]. But in Fig. 5D, a new sulfate formation generated on the 12

peak at 170.3 eV on used FAH, which is considered to be HgSO4. In Fig. 5E, the FAH did not have any peaks of N-containing groups after high temperature treatment. After mercury adsorption experiment, both nitrite and nitrate characteristic peaks appeared at 403.8 and 407.4 cm-1 [10, 15], and the nitrite peak was possibly ascribed to suface storage of adsorbed NO2 on Lewis acid sites [48]. Fig. 5G and 5H shows the Cl 2p spectra of the FHA before and after mercury experiment. Based on Table 2, the main ingredients of Fly ash are metallic compound, such as Ca, Al, Fe, Mg, Na and K. The peak at 198.4 eV is possibly the Cl- ion of CaCl2, FeCl3, NaCl and KCl [49]. The peak at 199. 8 eV is possibly the Cl- of AlCl3 and MgCl2 [50, 51]. The area ratio of low and high binding energy peaks was the same in Cl 2p spectra of fresh and used FHA. This result indicates that the species of chloride-containing compound did not change on the FAH in mercury adsorption experiments. In Fig. 5I and 5J, the chemically adsorbed and lattice oxygen was observed at 532.3, 531.4 and 530.4 cm-1, respectively [36]. The ratio of chemically adsorbed and lattice oxygen decreased from 0.49 to 0.39, indicating that some chemically adsorbed oxygen transformed into lattice oxygen. Fig. 5K and 5L shows the spectra change of Cl 2p on the surface of ACNCl5 before and after mercury adsorption. The Cl- and C-Cl peaks appeared in 198.6 and 200.2 cm-1 and the area ratio of Cl- and C-Cl peaks increased from 0.31 to 1.20, indicating that some C-Cl groups transformed into Cl- forms. This result is very different from the result of FAH. 3.3 Mercury Adsorption Mechanism 3.3.1 Various acid gases influence on mercury adsorption HCl Influence The HCl was usually considered as a promotional factor on mercury removal in previous studies [35, 52]. But the positive influence of HCl addition is not obvious on FAH based on Fig. 3A. According to previous studies, the HCl commonly promotes 13

the mercury removal in two ways. First, the HCl can be reduced to Cl2 or Cl via Deacon reaction under the catalysts at more than 300 oC [35, 53, 54]. Second, the Cl reacts with carbon atom to form C-Cl groups, which can serve as mercury adsorption sites [32]. In this study, the reaction temperature was kept at 180 oC, which is far lower than 300 oC, hence the Deacon reaction could not happen at this temperature. According to Table 1, the C content of FA is only 1.22% and that of FAH could be lower after high temperature treatment. Therefore, the HCl is difficult to form C-Cl on the surface of FAH. Moreover, the Cl content of FAH is very high (5.4%) on the basis of Table 2, which is far highrer than the Cl content of coal-fired fly ash [55]. This implies that the FAH contains so much Cl-containing compound that few adsorption sites are available to HCl adsorption. The two activation routes were both cut off in this study. Thereby, the HCl addition did not form many adsorption sites for mercury removal. SO2 Influence The SO2 addition slightly promoted the mercury removal in Fig. 2. The reaction mechanism is described as follows. First, based on in stitu DRIFT result, the SO2 could first Lewis acid sites and react with chemically adsorbed oxygen (O*) to form SO3. The Lewis acid sites are likely provided by metal oxide, such as Al2O3 and Fe2O3,on the basis of Table 2. Second, elemental mercury was oxidized into HgO by chemically adsorbed oxygen (O*), so the lattice oxygen ratio increased in XPS analysis. Third, the HgO was sulfated with adsorbed state SO3 to form HgSO4, hence the HgSO4 appeared on the TPD analysis result [13, 56]. As the reaction progressed, the formed mercury sulfate possibly diverted from the surface to the bulk since the mercury sulfate has higher mobility. This result is consistent with the XPS analysis result that a new sulfate salt was produced during mercury adsorption reaction. Moreover, the TPD experiment further proved the existence of sulfate mercury on the fly ash. This process belongs to Langmuir-Hinshelwood (L-H) mechanism. This reaction mechanism could be described via following reactions. 14

Hg0 → Hg0(ad)

(1)

Hg0(ad) + O* → HgO(ad)

(2)

SO2 + M → SO2-M

(3)

SO2-M + O* →SO3-M

(4)

SO3-M + HgO(ad) → HgSO4 + M

(5)

NO Influence The NO is the most effective acidic gas on the mercury removal amongst HCl, SO2 and NO. According to in stitu DRIFT, the NO likely reacted with chemically adsorbed oxygen or metal oxide (Al2O3, Fe2O3, etc.) to form NO2 adsorbed on Lewis acid sites or bridging nitrate [57]. The NO2 has higher oxidizability than NO, hence it has stronger affinity for mercury and can recouple with Hg to form HgO or nitrate mercury in the presence of O2 [58, 59]. As time went on, the new peaks appearance of nitrate and nitrite in Fig. 4A contributes to the accumulation of reaction product between active site NO2 and mercury. Furthermore, the characteristic peaks of nitrite and nitrate appeared in XPS analysis after mercury adsorption, proving the existence of active sites and nitrate mercury. The Hg(NO3)2 and HgO appearance also in TPD analysis also verifies the above inference. In this reaction process, the chemically oxygen are transformed into lattice oxygen (HgO). This inference is supported by the XPS result that the ratio of chemically adsorbed oxygen and lattice oxygen decreased after mercury adsorption experiment. According to the above analysis, the reaction mechanism of NO and Hg0 follows the Eley–Rideal (E-R) mechanism. The NO first adsorbs on the Lewis acid site to form NO2 and then oxidize the adsorbed state Hg0 to Hg(NO3)2 or HgO. Because the amount of Lewis acid site is limited, the NO increase from 200 ppm to 300 ppm slightly improve the mercury removal efficiency. The NO reaction mechanism could be described via following reactions. Hg0 → Hg0(ad)

(1) 15

NO + M2O3 → O-N-O-M2O2

(6)

O-N-O-M2O2 + Hg0(ad) → Hg-O-N-O-M2O2

(7)

2Hg-O-N-O-M2O2 + O2 → Hg(NO3)2 + M2O2

(8)

2M2O2 + O2 → 2 M2O3

(9)

3.3.2 ACNCl5 adsorption mechanism The 2.5 wt.% ACNCl5 addition shows excellent mercury removal performance. On the basis of result of Fig. 5K and 5L, the ratio of Cl- and C-Cl peaks increased from 0.31 to 1.20, indicating that the some C-Cl groups transformed into Cl- forms. This result indicates that the C-Cl can serve as adsorption sites and transform Hg0 into HgCl2. This inference has been proven via our previous study [32]. According to the above analysis, the mechanism of halide modified chars could be described as follows: (1) the chloride ions reacted with carbon atom to form C-Cl groups; (2) the elemental mercury transferred from gas phase to the surface of chars; (3) the C-Cl groups oxidized elemental mercury into mercury halides. The adsorption mechanism was described via following reactions. Hg0 → Hg0(ad)

(1)

Hg0(ad) + C-Cl → C-Hg-Cl

(10)

2Hg0(ad) + 2C-Cl → Hg2Cl2

(11)

Hg0(ad) + 2C-Cl → HgCl2

(12)

Hg2Cl2 + 2C-Cl → 2HgCl2

(13)

3.4 Comparison of various methods

As shown in Table 3, the promotional methods for mercury removal are compared 16

from four aspects: effect, cost, influence on devices and operating parameters. The acidic gas addition possibly corrodes the incineration boiler and pollution controlling devices. Moreover, excessive acidic gas concentration will increase the burden on subsequent dry/semidry scrubbing system for removing acidic gases. The increase of AC injection leads to a certain degree of mercury removal improvement. Yet the high cost for AC injection exceeds the affordability of MSW incineration plants in China. In addition, a large amount of AC injection enhances the pressure on fly ash capture of fabric bag filter and fly ash treatment. Above all, the acidic gas addition and 25% AC injection fail to decrease mercury emission concentration to an optimal level. As shown in Table 3, compared with other methods, the excellent effect and low cost make Cl-modified sorbent to be a good choice for mercury removal of MSW incineration. Furthermore, only 5 wt.% NH4Cl impregnation has little influence on devices and the injection invariability does not change any parameters for pollution controlling devices. To sum up, the ACNCl5 is recommended to be the optimal method for further mercury removal.

4. Conclusion In this study, the influence of flue gas components and original/modified AC injection on mercury removal was investigated on basis of the field test of a typical MSW incineration plant. The field test result shows that the mercury emission concentration of sampled plant is 25.67±18.13 μg·m-3, the mercury removal efficiency of APCDs is 71.9% and the mercury is captured in form of Hgp. The influence study of flue gas components indicates that the HCl fails to promote mercury removal while the SO2 and NO play positive role in the mercury removal. The invalidation of Deacon reaction at low temperature and the failure of C-Cl group formation on low carbon fly ash result in losing promotive efficacy of HCl addition on mercury removal. The SO2 adsorbs on Lewis acid sites and forms SO3-M, which reacts with HgO to form HgSO4. The NO addition forms NO2-M on Lewis acid sites of fly ash, which oxidizes Hg0 into HgO and Hg(NO3)2. The reaction process of SO2 and NO with Hg0 17

confirms to Langmuir-Hinshelwood (L-H) and Eley–Rideal (E-R) mechanism, respectively. Though the increase of AC injection amount slightly enhances mercury removal efficiency, the high cost exceeds the affordability of MSW incineration plants. The Cl-modified AC performs excellent mercury capture ability in various flue gases, which does not need to change any operating parameters. Based on the comparison in four aspects (effect, cost, influence on devices and operating parameters) the Cl-modified AC is recommended to be the optimal method for further mercury removal.

Acknowledgements This work was sponsored by the Major State Basic Research Development Program of China (973 Program) (No. 2013CB430000) and National Science Foundation of China (21521064).

18

Reference [1] P.B. Tchounwou, W.K. Ayensu, N. Ninashvili, D. Sutton, Environmental exposure to mercury and its toxicopathologic implications for public health, Environmental toxicology, 18 (2003) 149-175. [2] Minamata Convention on Mercury, (2013) www.mercuryconvention.org. [3] H. Cheng, Y. Hu, Municipal solid waste (MSW) as a renewable source of energy: Current and future practices in China, Bioresource Technology, 101 (2010) 3816-3824. [4] H. Cheng, Y. Hu, Mercury in municipal solid waste in China and its control: a review, Environmental science & technology, 46 (2012) 593-605. [5] H. Zhang, P.J. He, L.M. Shao, Fate of heavy metals during municipal solid waste incineration in Shanghai, Journal of hazardous materials, 156 (2008) 365-373. [6] D. Zhen-ya, S. Hai-tao, W. Feng-yang, L. Zhi-jian, L. Shao-pu, W. Shu-xiao, Mercury Emission Characteristics and Mercury Concentrations of Municipal Solid Waste in Waste Incineration Plants, ENVIＲONMENTAL SCIENCE, 37 (2016) 3766-3773. [7] F. Takahashi, T. Shimaoka, A. Kida, Atmospheric mercury emissions from waste combustions measured by continuous monitoring devices, Journal of the Air & Waste Management Association, 62 (2012) 686-695. [8] J.H. Kim, J.M. Park, S.B. Lee, D. Pudasainee, Y.C. Seo, Anthropogenic mercury emission inventory with emission factors and total emission in Korea, Atmospheric Environment, 44 (2010) 2714-2721. [9] Standards of Performance for New Stationary Sources and Emission Guidelines for Existing Sources: Commercial and Industrial Solid Waste Incineration Units; Final Rule, FEDERAL REGISTER, 81 (2016). [10] S. Haitao, STUDY ON ATMOSPHERIC MERCURY EMISSION CHARACTERISSTICS FROM MUNICIPAL SOLID WASTE INCINERATION INDUSTRY IN CHINA, (2016). [11] J. Wilcox, E. Rupp, S.C. Ying, D.-H. Lim, A.S. Negreira, A. Kirchofer, F. Feng, K. Lee, Mercury adsorption and oxidation in coal combustion and gasification processes, International Journal of Coal Geology, 90-91 (2012) 4-20. [12] T. Nishitani, I. Fukunaga, H. Itoh, T. Nomura, The relationship between HCL and mercury speciation in flue gas from municipal solid waste incinerators, Chemosphere, 39 (1999) 1-9. [13] F. Wang, S. Wang, Y. Meng, L. Zhang, Q. Wu, J. Hao, Mechanisms and roles of fly ash compositions on the adsorption and oxidation of mercury in flue gas from coal combustion, Fuel, 163 (2016) 232-239. [14] C. Richardson, T. Machalek, S. Miller, Effect of NOx Control Processes on Mercury Speciation in Utility Flue Gas, Journal of the Air & Waste Management Association (Air & Waste Management Association), 52 (2002) 941-947. [15] G. Li, B. Shen, Y. Li, B. Zhao, F. Wang, C. He, Y. Wang, M. Zhang, Removal of element mercury by medicine residue derived biochars in presence of various gas compositions, J Hazard Mater, 298 (2015) 162-169. [16] J.C. Hower, C.L. Senior, E.M. Suuberg, R.H. Hurt, J.L. Wilcox, E.S. Olson, Mercury capture by native fly ash carbons in coal-fired power plants, Progress in Energy and Combustion Science, 36 (2010) 510-529. [17] C. Gómez-Giménez, D. Ballestero, R. Juan, B. Rubio, M.T. Izquierdo, Mercury capture by a regenerable sorbent under oxycoal combustion conditions: Effect of SO2 and O2 on capture efficiency, Chemical Engineering Science, 122 (2015) 232-239. 19

[18] C. Zhu, Y. Duan, C.-Y. Wu, Q. Zhou, M. She, T. Yao, J. Zhang, Mercury removal and synergistic capture of SO2/NO by ammonium halides modified rice husk char, Fuel, 172 (2016) 160-169. [19] M.A. Uddin, T. Yamada, R. Ochiai, E. Sasaoka, S. Wu, Role of SO2 for Elemental Mercury Removal from Coal Combustion Flue Gas by Activated Carbon, Energy & Fuels, 22 (2008) 2284-2289. [20] L.P. Belo, L.K. Elliott, R.J. Stanger, T.F. Wall, Impacts of Sulfur Oxides on Mercury Speciation and Capture by Fly Ash during Oxy-fuel Pulverized Coal Combustion, Energy & Fuels, 30 (2016) 8658-8664. [21] Y. Mu, A. Saffarzadeh, T. Shimaoka, Influence of ignition process on mineral phase transformation in municipal solid waste incineration (MSWI) fly ash: Implications for estimating loss-on-ignition (LOI), Waste management, 59 (2017) 222-228. [22] C.K. Goh, S.E. Valavan, T.K. Low, L.H. Tang, Effects of different surface modification and contents on municipal solid waste incineration fly ash/epoxy composites, Waste management, 58 (2016) 309-315. [23] K. Felsvang, R. Gleiser, G. Juip, K.K. Nielsen, Activated carbon injection in spray dryer/ESP/FF for mercury and toxics control, Fuel Processing Technology, 39 (1994) 417-430. [24] H. Tejima, I. Nakagawa, T.-a. Shinoda, I. Maeda, PCDDs/PCDFs reduction by good combustion technology and fabric filter with/without activated carbon injection, Chemosphere, 32 (1996) 169-175. [25] Z. Tan, L. Sun, J. Xiang, H. Zeng, Z. Liu, S. Hu, J. Qiu, Gas-phase elemental mercury removal by novel carbon-based sorbents, Carbon, 50 (2012) 362-371. [26] I. Ie, C. Hung, Y. Jen, C. Yuan, W. Chen, Adsorption of vapor-phase elemental mercury (Hg0) and mercury chloride (HgCl2) with innovative composite activated carbons impregnated with Na2S and S0 in different sequences, Chemical Engineering Journal, 229 (2013) 469-476. [27] I.R. Ie, W.C. Chen, C.S. Yuan, C.H. Hung, Y.C. Lin, H.H. Tsai, Y.S. Jen, Enhancing the adsorption of vapor-phase mercury chloride with an innovative composite sulfur-impregnated activated carbon, Journal of hazardous materials, 217-218 (2012) 43-50. [28] G. Li, B. Shen, Y. Wang, S. Yue, Y. Xi, M. An, K. Ren, Comparative study of element mercury removal by three bio-chars from various solid wastes, Fuel, 145 (2015) 189-195. [29] American Society for Testing and Materials (ASTM), ASTM D6784-02 Standard Test Method for Elemental, Oxidized, Particle-Bound and Total Mercury in Flue Gas Generated from Coal-Fired Stationary Sources (Ontario Hydro Method). in: A.S.f.T.a. Materials (Ed.), West Conshohocken, PA, 2002. [30] Z. Duan, S. Haitao, W. Fengyang, L. Zhijian, L. Shaopu, W. Shuxiao, Mercury Emission Characteristics and Mercury Concentrations of Municipal Solid Waste in Waste Incineration Plants, ENVIRONMENTAL SCIENCE, 37 (2016) 459-465. [31] Z. Duan, S. Haitao, W. Fengyang, Z. Lei, W. Shuxiao, Y. Bin, Mercury Distribution Characteristics and Atmospheric Mercury Emission Factors of Typical Waste Incineration Plants in Chongqing, ENVIRONMENTAL SCIENCE, 37 (2016) 459-465. [32] G. Li, S. Wang, Q. Wu, F. Wang, B. Shen, Mercury sorption study of halides modified bio-chars derived from cotton straw, Chemical Engineering Journal, 302 (2016) 305-313. [33] M. Rumayor, M. Diaz-Somoano, M.A. Lopez-Anton, M.R. Martinez-Tarazona, Mercury compounds characterization by thermal desorption, Talanta, 114 (2013) 318-322. [34] M. Rallo, M.A. Lopez-Anton, R. Meij, R. Perry, M.M. Maroto-Valer, Study of mercury in by-products from a Dutch co-combustion power station, Journal of hazardous materials, 174 (2010) 28-33. 20

[35] Y. Gao, Z. Zhang, J. Wu, L. Duan, A. Umar, L. Sun, Z. Guo, Q. Wang, A Critical Review on the Heterogeneous Catalytic Oxidation of Elemental Mercury in Flue Gases, Environmental Science & Technology, 47 (2013) 10813-10823. [36] L. Xing, Y. Xu, Q. Zhong, Mn and Fe Modified Fly Ash As a Superior Catalyst for Elemental Mercury Capture under Air Conditions, Energy & Fuels, 26 (2012) 4903-4909. [37] A.D. Jew, E.C. Rupp, D.L. Geatches, J.-E. Jung, G. Farfan, L. Bahet, J.C. Hower, G.E. Brown, J. Wilcox, Mercury Interaction with the Fine Fraction of Coal-Combustion Fly Ash in a Simulated Coal Power Plant Flue Gas Stream, Energy & Fuels, 29 (2015) 6025-6038. [38] Z. Wang, G. Liu, H. Zheng, F. Li, H.H. Ngo, W. Guo, C. Liu, L. Chen, B. Xing, Investigating the mechanisms of biochar’s removal of lead from solution, Bioresource Technology, 177 (2015) 308-317. [39] Y. Su, M.D. Amiridis, In situ FTIR studies of the mechanism of NOx storage and reduction on Pt/Ba/Al2O3 catalysts, Catalysis Today, 96 (2004) 31-41. [40] C. Sedlmair, Elementary steps of NOx adsorption and surface reaction on a commercial storage–reduction catalyst, Journal of Catalysis, 214 (2003) 308-316. [41] Q. Li, M. Meng, Z.Q. Zou, X.G. Li, Y.Q. Zha, Simultaneous soot combustion and nitrogen oxides storage on potassium-promoted hydrotalcite-based CoMgAlO catalysts, Journal of hazardous materials, 161 (2009) 366-372. [42] R. Jin, Y. Liu, Y. Wang, W. Cen, Z. Wu, H. Wang, X. Weng, The role of cerium in the improved SO2 tolerance for NO reduction with NH3 over Mn-Ce/TiO2 catalyst at low temperature, Applied Catalysis B: Environmental, 148-149 (2014) 582-588. [43] M. Waqif, P. Bazin, O. Saur, J.C. Lavalley, G. Blanchard, O. Touret, Study of ceria sulfation, Applied Catalysis B: Environmental, 11 (1997) 193-205. [44] T. Hamzehlouyan, C.S. Sampara, J. Li, A. Kumar, W.S. Epling, Kinetic study of adsorption and desorption of SO2 over γ-Al2O3 and Pt/γ-Al2O3, Applied Catalysis B: Environmental, 181 (2016) 587-598. [45] A. Saha, D.N. Abram, K.P. Kuhl, J. Paradis, J.L. Crawford, E. Sasmaz, R. Chang, T.F. Jaramillo, J. Wilcox, An X-ray Photoelectron Spectroscopy Study of Surface Changes on Brominated and Sulfur-Treated Activated Carbon Sorbents during Mercury Capture: Performance of Pellet versus Fiber Sorbents, Environmental Science & Technology, 47 (2013) 13695-13701. [46] N.D. Hutson, B.C. Attwood, K.G. Scheckel, XAS and XPS Characterization of Mercury Binding on Brominated Activated Carbon, Environmental Science & Technology, 41 (2007) 1747-1752. [47] Y. Yao, V. Velpari, J. Economy, Design of sulfur treated activated carbon fibers for gas phase elemental mercury removal, Fuel, 116 (2014) 560-565. [48] K. Tikhomirov, O. Kröcher, M. Elsener, A. Wokaun, MnOx-CeO2 mixed oxides for the low-temperature oxidation of diesel soot, Applied Catalysis B: Environmental, 64 (2006) 72-78. [49] N. V.I., S. Y.V., M. A.G., N. L.A., B. I.B., Zh. Neorg. Khimii 19, 1353, (1974). [50] M. Miyake, H. Motonami, S. Shiomi, T. Hirato, Electrodeposition of purified aluminum coatings from dimethylsulfone–AlCl3 electrolytes with trimethylamine hydrochloride, Surface and Coatings Technology, 206 (2012) 4225-4229. [51] E. Magni, G.A. Somorjai, Preparation of a model Ziegler-Natta catalyst: electron irradiation induced titanium chloride deposition on magnesium chloride thin films grown on gold, Surface Science, 345 (1996) 1-16. [52] H. Xu, Z. Qu, C. Zong, W. Huang, F. Quan, N. Yan, MnOx/Graphene for the Catalytic Oxidation and Adsorption of Elemental Mercury, Environmental science & technology, 49 (2015) 6823-6830. 21

[53] W. Hou, J. Zhou, C. Yu, S. You, X. Gao, Z. Luo, Pd/Al2O3Sorbents for Elemental Mercury Capture at High Temperatures in Syngas, Industrial & Engineering Chemistry Research, 53 (2014) 9909-9914. [54] J. Wilcox, T. Okano, Ab initio-based Mercury Oxidation Kinetics via Bromine at Postcombustion Flue Gas Conditions, Energy & Fuels, 25 (2011) 1348-1356. [55] K.C. Galbreath, C.J. Zygarlicke, J.E. Tibbetts, R.L. Schulz, G.E. Dunham, Effects of NOx, α-Fe2O3, γ-Fe2O3, and HCl on mercury transformations in a 7-kW coal combustion system, Fuel Processing Technology, 86 (2005) 429-448. [56] H. Li, C.-Y. Wu, Y. Li, L. Li, Y. Zhao, J. Zhang, Impact of SO2 on elemental mercury oxidation over CeO2–TiO2 catalyst, Chemical Engineering Journal, 219 (2013) 319-326. [57] Y. Li, P.D. Murphy, C.-Y. Wu, K.W. Powers, J.-C.J. Bonzongo, Development of Silica/Vanadia/Titania Catalysts for Removal of Elemental Mercury from Coal-Combustion Flue Gas, Environmental science & technology, 42 (2008) 5304-5309. [58] A. Fuente-Cuesta, M.A. Lopez-Anton, M. Diaz-Somoano, M.R. Martínez-Tarazona, Retention of mercury by low-cost sorbents: Influence of flue gas composition and fly ash occurrence, Chemical Engineering Journal, 213 (2012) 16-21. [59] A. Zhang, W. Zheng, J. Song, S. Hu, Z. Liu, J. Xiang, Cobalt manganese oxides modified titania catalysts for oxidation of elemental mercury at low flue gas temperature, Chemical Engineering Journal, 236 (2014) 29-38.

22

Fig. 1 Mercury emission characteristics of MSW incineration flue gas before and after APCDs

23

100 ppm NO

Mercury concentration of flue gas (gm-3)

Mercury concentration of flue gas (gm-3)

turn off SO2

A

60

Hg0 Hg2+ Hgt

50 40 all flue gas

30

turn off HCl

5%O2

20 fly ash

1000 ppm SO2

300 ppm HCl

10 0 0

50

100

150

200

250

300

turn off SO2

B

60

all flue gas

40 30

turn off HCl

5%O2 fly ash

1000 ppm SO2

20

100 ppm NO add

0 0

50

100

150

200 ppm NO add

Hg0 Hg2+ Hgt

300 ppm NO add

40 30 20 10 0 200

400

200

250

300

t (min)

50

0

300 ppm HCl

10

350

Mercury concentration of flue gas (gm-3)

Mercury concentration of flue gas (gm-3)

C

Hg0 Hg2+ Hgt

50

t (min)

60

300 ppm NO

600

800

1000

60

D

Hg0 Hg2+ Hgt

fly ash

50 40

300 1000 ppm SO2 ppm HCl

100 ppm NO

30 20 5%O2

turn off SO2

turn off HCl

all flue gas

10 0 0

50

100

150

200

250

300

350

t (min)

t (min)

Fig. 2 The influence of various components of flue gas on mercury removal capacity of different sorbents, A: FAH; B: FAH + 2.5% AC; C: FAH + 2.5% AC; D: FAH+2.5%ACNCl5

24

Fig. 3 The analysis of mercury existing form in fly ash

25

A

B

60 min 60 min 30 min

Absorbance (a.u.)

Absorbance (a.u.)

30 min 20 min 10 min 5 min 3 min

20 min 10 min 5 min 3 min

1 min

1 min

0 min

1250

1500

1750

0 min

3000

3500

4000

1250

Wavenumbers (cm-1)

1500

3000

3250

3500

3750

4000

Wavenumbers (cm-1)

Fig. 4 In situ DRIFT spectra of FAH obtained at 180 ◦C in the presence of 50 μg·m-3 Hg0 + 6% O2 with 100 ppm NO (A) and 500 ppm SO 2 (B)

26

Hg 4f

A fresh FAH

96

98

100

102

104

106

108

110

96

98

Binding energy (eV)

100

102

104

106

108

110 160

162

F

164

S 2p

D used FAH

fresh FAH

Binding energy (eV)

N 1s

E

166

168

170

172

174

176 160

162

164

Binding energy (eV)

G

N 1s

used FAH

fresh FAH

S 2p

C

Hg 4f

B used FAH

166

168

170

172

174

176

Binding energy (eV)

Cl 2p

Cl 2p

H

fresh FAH

used FAH

394 396 398 400 402 404 406 408 410 412 414 394 396 398 400 402 404 406 408 410 412 414 416 190 192 194 196 198 200 202 204 206 208 190 192 194 196 198 200 202 204 206 208

O 1s

I fresh FAH

526

528

Binding energy (eV)

Binding energy (eV)

Binding energy (eV)

O 1s

J used FAH

530

532

534

Binding energy (eV)

536

538 526

528

K

Binding energy (eV)

530

532

534

536

538 192

Binding energy (eV)

194

196

198

Cl 2p

L

Cl 2p

used ACNCl5

fresh ACNCl5

200

202

204

Binding energy (eV)

206

208

192

194

196

198

200

202

204

206

208

Binding energy (eV)

Fig. 5 XPS analysis of fresh and used sorbents: (A), (B): Hg 4f; (C), (D): S 2p; (E), (F): N 1s; (G), (H): Cl 2p; (I), (J): O 1s; (K), (L): Cl 2p

27

Table 1 Proximate and ultimate analysis of FA and AC

Sample

Proximate analysis (%)

Ultimate analysis (%)

Moisture FC

Ash

Volatile

C

H

O

N

S

FA

1.32

1.78

89.72

7.18

1.22

0.16

5.29

0.71

1.14

AC

3.94

59.47

17.46

19.13

61.95

2.33

15.44

0.66

0.00

28

Table 2 Major chemical component of fly ash by XRD Component (%)

Sample CaO

SiO2

Al2O3

Fe2O3

MgO

Na2O

K2O

TiO 2

Cl

SO3

P2O5

FA

27.9

24.2

16.4

4.5

3.1

3.2

2.2

1.5

6.1

5.9

3.5

FAH

30.9

25.1

15.7

4.6

3.7

2.2

1.8

1.4

5.4

3.6

4.3

29

Table 3 Comparison of various promotional methods for mercury removal Method

Effect

Cost

Influence on devices

Operating parameters

HCl

Bad

Modest

Yes

Changed

SO2

Good

Modest

Yes

Changed

NO

Good

Modest

Yes

Changed

AC Increase

Good

High

No

Changed

ACNCl5

Excellent

Low

No

Unchanged

30

Graphical abstract

31

Highlights 1. The SO2 can form effective group SO3-M on Lewis acid site. 2. The NO addition forms NO2-M group oxidizing Hg0 into HgO and Hg(NO3)2. 3. The HCl addition fails to promote mercury removal. 4. The Cl-modified AC performs excellent mercury capture ability in various flue gases. 5. Cl-modified AC is recommended to be the optimal method for mercury removal.

32