DFT and Experimental Study on the Mechanism of Elemental Mercury

Apr 15, 2016 - (21) Smith et al. also discovered in a bench-scale study that HCl ..... of Science and Technology, China (2014CB238904, 2013CB228504)...
1 downloads 0 Views 3MB Size
Download PDF
Article pubs.acs.org/est

DFT and Experimental Study on the Mechanism of Elemental Mercury Capture in the Presence of HCl on α‑Fe2O3(001) Ting Liu,† Lucheng Xue,‡ Xin Guo,*,† Yu Huang,† and Chuguang Zheng† †

State Key Laboratory of Coal Combustion, Huazhong University of Science and Technology, Wuhan 430074, China China United Engineering Corporation, Bin an Road, 1060, Binjiang District, Hangzhou 310052, China

‡

S Supporting Information *

ABSTRACT: To investigate the mechanism of Hg0 adsorption on the α-Fe2O3(001) surface in the presence of HCl, which is considered to be beneficial for Hg0 removal, theoretical calculations based on density functional theory as well as corresponding experiments are carried out. HCl adsorption is first performed on the α-Fe2O3(001) surface, and the Hg0 adsorption on HCl-adsorbed α-Fe2O3(001) surface is subsequently researched, demonstrating that HCl dissociates on the surface of α-Fe2O3, improving the Hg0 adsorption reactivity. With further chlorination of the α-Fe2O3(001) surface, FeCl3 can be achieved and the adsorption energy of Hg0 on the FeCl3 surface reaches −104.2 kJ/mol, representing strong chemisorption. Meanwhile, a group of designed experiments, including Hg0 adsorption on HCl-preadsorbed α-Fe2O3 as well as the coadsorption of both gaseous components, are respectively performed to explore the pathways of Hg0 transformation. Combining computational and experimental results together, the Eley−Rideal mechanism with HCl preadsorption can be determined. In addition, subsequent X-ray photoelectron spectroscopy analysis verifies the appearance of Cl species and oxidized mercury, exhibiting the consistency with experiments.

■

INTRODUCTION Mercury pollution has received increasing worldwide attention in recent years, and the control of mercury emission is presently an area of intense research and difficulty.1 The Minamata Convention on Mercury2 in 2013 has set a goal to develop an international treaty to curb mercury emission, conveying that mercury emission has become a major environmental concern. Elemental mercury, which has been extensively studied, is thermodynamically stable, highly volatile, and slightly soluble in water,3,4 leading to difficult removal. As known, the combustion of fossil fuels is the major anthropogenic source of mercury pollution. Thus, it is necessary to study mercury speciation conversion and explore effective approaches to capture elemental mercury. Chlorine in coal is released as HCl in coal-derived gas, which is an important factor to consider in mercury transformation. In recent years, many research studies5−10 have demonstrated that the presence of HCl could effectively improve the oxidation of mercury in flue gas, and HgCl2 is considered as the oxidized form of mercury.9 However, researchers have not reached a consensus on the specific effect mechanism of HCl,11−13 and with progress from relevant studies, heterogeneous mechanisms, including Eley−Rideal, Langmuir−Hinshelwood, and Mars−Maessen,14,15 are increasingly accepted by researchers and have become the mainstream view in academia. It has also been reported that fly ash16 plays an important role in this heterogeneous reaction; however, iron oxide in fly © XXXX American Chemical Society

ash, which is of scientific and technological importance because it is easily accessed and separated, is nonsecondary pollution, and is less expensive than other sorbents, is proved to be the active component.13,17−20 Wu et al. indicated that Fe2O3 effectively reduced the mercury on iron-based sorbents11 in simulated coal-derived gases, but the presence of HCl inhibited the mercury removal ability of Fe2O3. However, Galbreath et al. demonstrated19 that elemental mercury was mostly oxidized in the presence of 100 ppm of HCl with Fe2O3. Meng et al. reported that with the increase in concentration (0−100 ppm) of HCl, the mercury removal ability of fly ash increased correspondingly.21 Smith et al. also discovered in a bench-scale study that HCl promoted the heterogeneous reaction between HCl and Fe2O3.22 Because of the complexities of the gaseous atmosphere, evaluation of the role of HCl in mercury removal was ambiguous. Meanwhile, our previous computational work and experimental results focused on the reaction mechanism of Hg0 adsorption with H2S15 in the gasification process and clearly demonstrated that Hg0 oxidation followed the Eley−Rideal mechanism. However, a major gas component of HCl, especially in computational research, is relatively defiReceived: January 24, 2016 Revised: April 11, 2016 Accepted: April 15, 2016

A

DOI: 10.1021/acs.est.5b06340 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Article

Environmental Science & Technology

Figure 1. Three stable adsorption configurations of HCl on the α-Fe2O3(001) surface.

Table 1. Different HCl Adsorption Configuration Parameters configuration A B C a

Eadsa(kJ/mol)

RH−Clb(Å)

RFe−Cl (Å)

RO−H (Å)

MFe−Clc

MO−H

−108.1 −159.2 −178.5

3.20 3.27 3.17

2.18 2.19 2.18

0.98 0.98 0.98

0.53 0.53 0.53

0.65 0.64 0.65

Adsorption energy. bBond length. cBond population.

cient,14,15,23 and the role of HCl in Hg0 oxidation needs further exploration. In the present work, density functional theory (DFT) is employed to verify the microcosmic mechanism and characteristics of mercury transformation on α-Fe2O3(001), and corresponding experiments with HCl are designed to explore the reaction pathway of the heterogeneous process. By combining theoretical and experimental research methodologies, this paper offers an intensive understanding of the influence of HCl on the reaction between Hg0 and α-Fe2O3 and provides theoretical reference for the application of iron-based adsorbent. Furthermore, the decomposition behaviors of samples used and XPS tests on the Hg0-adsorbed sample are analyzed to determine the probable chlorine species and mercury products.

considered in our calculations. More details on the optimization of HCl molecule can be found in Supporting Information. Characterization of Catalyst. Pure spherical α-Fe2O3 (Aladdin Chemistry Co. Ltd., particle size of 30 nm, purity 99.5%) and anhydrous iron(III) chloride (≥99.5%), which are also described in Table S1, are used in our designed experiments.23 Experimental Apparatus and Procedures. The experimental system diagram mainly consists of four parts: the gas distribution system, the mercury generator, the reaction equipment, and the online mercury measuring instrument, which is pictured in Figure S5. In tests, 50 mg of α-Fe2O3 sample is used and the inlet Hg0 concentration is around 40 ± 1 μg/m3 with a total inlet gas flow of 1.2 L/min, accompanied by a gas mixture of HCl (50 ppm) and N2 (balance gas). The real-time Hg0 removal efficiency in this paper is defined as η, which is expressed in the following equation:

■

EXPERIMENTAL SECTION Models and Computational Methods. All calculations in this paper are performed using the CASTEP (Cambridge Sequential Total Energy Package),24 which is based on DFT theory.14,15,23 For a detailed description of computational parameter selection and the structure of α-Fe2O3(001) surface, refer to Supporting Information. The adsorption energy in our calculations can be defined as the energy variation between the reactants and products of the adsorption process, and the calculation formula is defined in eq 1. Eads = EAB − (EA + E B)

η = (1 − Cout /C in) × 100%

(2)

where Cin and Cout respectively represent the real-time Hg0 concentration in the inlet and outlet of the reaction tube. Four groups of contrasted Hg0 removal experiments are carried out, as illustrated in Table S2, in which a pretreatment of HCl on α-Fe2O3 is conducted.23 It is prepared through the total gas flow of N2 and HCl over the sample for 2 h at a certain temperature; subsequently, the sample is flushed with nitrogen for 0.5 h and designated as HCl/α-Fe2O3. Then the HCl preadsorbed sample is used to react with Hg0 at the same temperature of pretreatment and represented as HCl/αFe2O3+Hg0. In contrast, a sample that reacts with HCl and Hg0 is marked as HCl+α-Fe2O3+Hg0. X-ray Photoelectron Spectroscopy (XPS) Study. To verify the key elements on the surface of contrasted sorbents before and after adsorbing Hg0 during the experiments, high resolution XPS spectra are measured to characterize the available functional groups. The binding energy (BE) of the C 1s level at 284.80 eV is used as an internal reference to calibrate every spectrum. Additionally, XPS analysis of key elements is based on the NIST X-ray Photoelectron Spectros-

(1)

In this formula, EA and EB respectively represent the energies of the adsorbate and substrate, and EAB is the total energy of the substrate together with adsorbate. The more negative the value, the stronger the absorption. Adsorption Behaviors of HCl on α-Fe2O3(001). To comprehensively research the adsorption behaviors of HCl on the α-Fe2O3(001) surface, different adsorption directions of HCl as well as different adsorption sites including Fe top, O top, and hollow sites of α-Fe2O3(001) are respectively B

DOI: 10.1021/acs.est.5b06340 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Article

Environmental Science & Technology

structures. Based on the calculations of Hg0 on the αFe2O3(001) surface in our previous studies,14,25 the corresponding data are directly quoted for comparison in this paper. All final stable configurations are obtained through structure optimization and illustrated in Figure 3, and the specific parameters of every configuration are listed in Table 2. As pictured in configurations D, E, and F, the Hg atom bonds with surface Fe atoms, respectively, except for Fe1. By observing the parameters of the three configurations in Table 2, we find that the adsorptive behaviors of Hg in Fe2, Fe3, and Fe4 are quite similar, showing the same adsorption energy of −48.3 kJ/mol. In the most stable configuration E, the bond length of Hg−Fe is about 2.74 Å and the bond population is 0.32, which indicates that the extranuclear electron cloud contact ratio is quite large, namely the interaction between Hg and Fe atoms is stronger. In addition, Mulliken’s charge analysis shows that when the structure reaches the stable state, about 0.20 electron between Hg and Fe atoms is transferred. For adsorption of the Hg atom on the pure surface of αFe2O3(001), Hg can adsorb on the surface of the Fe top position, yielding an adsorption energy of −29.9 kJ/mol.14 In contrast, the adsorption energy of the Hg atom on the Fe−Cl surface is enhanced by 61.5%. In addition, it can be observed from Table 2 that the bond population (0.30) of the Hg atom on the pure surface is slightly smaller than on the Cl surface (0.32), while electron transfer amount between Fe and Hg atoms is also relatively smaller. In another case, it is proved that the configuration between the Hg atom and the Fe1−Cl substrate is unstable according to our calculations.14 However, if the Hg atom combines with the surface of the Fe1 atom, and meanwhile the HCl molecule dissolves on the surface of α-Fe2O3(001), the Hg−Cl bond can be theoretically produced by the interaction of Hg and active Cl atoms. Therefore, the Hg−Cl bond combines with the surface of the Fe1 atom to form the stable configuration G, yielding an adsorption energy of −116.8 kJ/mol, and the bond length of Hg−Fe and Hg−Cl is respectively about 2.6 and 2.5 Å. In conclusion, from the perspectives of adsorption energy, bond population, and Mulliken’s charge analysis, HCl that adsorbs on the α-Fe2O3(001) surface improves the surface activation and makes the adsorption of Hg0 on the surface Fe top site become more intensive. To be specific, it reflects the improvement of adsorption energy, the increase in extranuclear electron cloud contact ratio, and the promotion of electron transfer amount. The conclusion inferred from the above results is that HCl dissociates on the Fe2O3 surface and undergoes intense chemical reaction with it, promoting the transformation of Hg0. Furthermore, with continuous adsorption of HCl, it is worth researching the characteristics of elemental mercury oxidation on the Fe2O3 surface, especially the structure of iron oxides. Hg0 Adsorption on the FeCl3(001) Surface. We judge that with the further adsorption of HCl, FeCl3 is probably generated. Related experimental studies also proposed that the existence of FeCl3 had considerable promotion on demercuration26−28 with the adsorbents of activated carbon and zeolite, playing an assignable role on the adsorption of Hg0, while theoretical studies on FeCl3 are comparatively few. Therefore, we establish the adsorption model of FeCl3(001) in line with structure data of FeCl3, which are shown in Figure S6. The calculation results display that Hg0 does not interact with the surface Cl site and prefers to interact with the surface

copy Database. For more details, refer to Supporting Information.

■

RESULTS AND DISCUSSION Computational Research on the Adsorption of HCl on α-Fe2O3(001) Surface. After geometry optimization of all the original configurations, three kinds of similar stable adsorption configurations are derived, and the optimized structure and corresponding parameters are pictured in Figure 1 and Table 1, respectively. In configuration A, HCl dissociately adsorbs on the surface of Fe2O3(001) with an adsorption energy of −108.1 kJ/mol. The Cl atom combines with the surface Fe atom, forming an Fe−Cl bond, and the bond length is 2.18 Å, which is very close to the experimental value (2.17 Å) of the Fe−Cl bond length in FeCl3. On the other hand, the H atom can form a H−O bond of 0.98 Å with OA adjacent to the surface Fe atom that interacts with Cl. During the adsorption process, the H−Cl bond of HCl extends 150% from 1.28 to 3.20 Å. Namely, the H−Cl bond completely breaks after adsorption. From its adsorption energy and changes in bond length, we can judge that HCl exhibits strong dissociated adsorption on the surface of α-Fe2O3(001). In configuration B, the Cl atom also binds with the Fe atom with the bond length of 2.19 Å. Unlike configuration A, the H atom forms a H−O bond of 0.98 Å with OB which is not adjacent to the Fe atom. H−Cl can be stretched to 155% and reaches 3.27 Å, and the adsorption energy is −159.2 kJ/mol. It displays stronger interaction than in configuration A. For configuration C, the Cl atom binds with the surface Fe atom as well, forming an Fe−Cl bond of 2.18 Å, while the H atom binds with OC with a bond length of 0.98 Å. The H−Cl length is dramatically stretched to 3.17 Å and appears to break. From the perspective of energy, configuration C can form the most stable final configuration in three configurations with an absorption energy of −178.5 kJ/mol, indicating strong chemisorption. Hence, under the influence of Fe2O3, HCl dissociates and forms a similar absorptive configuration on the surface of αFe2O3(001), in which the Cl and H atom respectively binds with the surface Fe and O atom. Hg0 Adsorption on the Cl/α-Fe2O3(001) Surface. The adsorption behaviors of Hg0 on the Cl/α-Fe2O3(001) surface are studied in this section, and the most stable configuration C obtained in the previous section is regarded as the adsorptive substrate. The Hg atom is successively placed on each possible absorptive site on the surface of this substrate, and structure optimization is conducted gradually. According to the calculations in the above section, the Cl site is formed on the surface, as illustrated in Figure 2. Thus, the Cl site as well as the Fe top site is considered as the possible adsorption site, and both of them are taken into consideration in the initial

Figure 2. Schematic diagram of adsorption sites of the chlorinated αFe2O3(001) surface. C

DOI: 10.1021/acs.est.5b06340 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Article

Environmental Science & Technology

Figure 3. Stable adsorption configuration of Hg0 on the chlorinated α-Fe2O3(001) surface.

Table 2. Hg0 Adsorption Configuration Parameters on the Cl/α-Fe2O3(001) Surface configuration

Eadsa (kJ/mol)

RHg−Feb (Å)

MHg−Fec

Qd (e)

D E F Hg on pure surface14

−48.3 −48.3 −48.3 −29.9

2.74 2.74 2.73 2.75

0.32 0.32 0.32 0.30

0.19 0.20 0.20 0.17

a

Adsorption energy. bBond length. cBond population. dMulliken’s charge analysis.

Fe atom, ultimately forming stable configuration H as shown in Figure 4.

Figure 5. Schematic diagram of the reaction pathway for Hg0, HCl, and α-Fe2O3(001).

yielding adsorption energy of −48.3 kJ/mol. Second, constant adsorption of HCl might allow complete chlorination on the surface of ferric oxide and form an FeCl3 layer with higher activation, which dramatically promotes the adsorption of Hg0 with an adsorption energy of −104.2 kJ/mol. Hg0 Removal Experiments with HCl on α-Fe2O3 and FeCl3. The above theoretical calculation has clarified that HCl can promote the removal of mercury on the α-Fe2O3(001) surface through chlorination of the substrate. To further explore the mechanism of Hg0 removal with HCl, a group of designed experiments and XPS tests are performed to analyze the chlorine species and probable pathways. Considering that HCl can dissolve on the surface of iron oxides, a series of experiments are carried out to confirm if the intermediate products are responsible for the sorption and removal of gas-phase Hg. Figure 6a shows the Hg0 removal efficiency of the contrasted samples under different experimental conditions. It is found that the removal efficiency of experiment I is nearly zero, that is to say there is basically no adsorption effect of α-Fe2O3 on Hg0 under the condition without HCl. For experiment II, when Hg0 and HCl react simultaneously on the α-Fe2O3 surface, the removal efficiency of Hg0 reaches its peak within 60 min, and the highest efficiency is 51%. Compared with the results of experiments I and II, it is observed that the existence of HCl significantly improves the demercuration ability of α-Fe2O3. Through observing the results of experiment III, we discover that when HCl absorbs to the α-Fe2O3 surface in advance, the subsequent absorption of Hg0 starts earlier compared to experiment II. Meanwhile, the mercury removal efficiency

Figure 4. Stable adsorption configuration of Hg0 on the FeCl3(001) surface.

In configuration H, Hg bonds with two surface Fe atoms, and the bond length is respectively 2.66 and 2.77 Å. Mulliken population of the two Hg−Fe bonds is 0.27 and 0.26, respectively. In this process, the adsorption energy is −104.2 kJ/mol, indicating that the adsorption of Hg0 on the surface of FeCl3 is due to strong chemical adsorption. Compared with the surface of α-Fe2O3(001), adsorption of Hg0 on the surface of FeCl3 is obviously stronger. As a result, existence of HCl will also possibly chloridize ferric oxide and form FeCl3, so as to promote adsorption of Hg0 on the surface of adsorbents dramatically. Reaction Mechanism for Hg0, HCl, and α-Fe2O3(001). Through the theoretical calculations, it is observed that with the constant increase of chlorination degree on the Fe2O3 surface, there might be two stages in the heterogeneous reaction pathway, which is presented in Figure 5. First, HCl dissociately adsorbs on the α-Fe2O3(001) surface, leaving the Cl atom bound strongly with the surface Fe, and Hg0 then reacts with it, D

DOI: 10.1021/acs.est.5b06340 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Article

Environmental Science & Technology

Figure 6. Hg0 removal efficiency on α-Fe2O3 at 80 °C. (a) Reaction time: 1 h; (b) reaction time: 8 h.

increases greatly at the beginning of the reaction, and then it reaches its maximum value of 44%, which is relatively lower than that of simultaneous adsorption of Hg0 and HCl. Combined with the theoretical calculation results, we speculate that the dissociated adsorption of HCl can activate the α-Fe2O3 surface primarily to produce active sites, enhancing the Hg0 oxidation. Additionally, because there is no supplement of HCl absorption sites during the Hg0 absorbing process in experiment III, its removal efficiency is a little lower compared with the simultaneous adsorption of Hg0 and HCl. At the same time, the decomposition behaviors of the Hg0-absorbed sample (HCl +α-Fe2O3+Hg0) are observed. An elemental mercury desorption peak at around 260 °C is shown in Figure S7, indicating that the adsorbed mercury compound was probably formed on the α-Fe2O3 surface. With the continuous adsorption of HCl in experiment II, the Hg0 removal efficiency with a long reaction time is illustrated in Figure 6b. The Hg0 removal efficiency continually decreases from 50% to 35% as the time of reaction is prolonged; however, the efficiency rises again and reaches a level of 70% after about 6 h, which is probably caused by the emergence of some other positive active sites. On the basis of our calculation conclusions, we conjecture that the produced FeCl3 is beneficial to the transformation of Hg0. Then Hg0 oxidation experiments on FeCl3 at 80 °C are performed and shown in Figure S8, in which the Hg0 removal efficiency reaches nearly 55%, indicating that FeCl3 benefits the Hg0 adsorption. Therefore, it is deduced that constant HCl chlorination on α-Fe2O3 plays a significant role in capturing Hg0 with the extended reaction time. After the desired degree of chlorination, α-Fe2O3 is chloridized and FeCl3 can be acquired, working continuously for capturing Hg0. XPS Analysis of the Hg0-Adsorbed Samples. As depicted in Figure 7, the XPS spectra29−31 of Hg 4f and Cl 2p for the contrasted samples at 80 °C are characterized. First, the Cl 2p peak of the HCl/α-Fe2O3 sample mainly centers around 198.7 eV, which is in accordance with the available Xray photoelectron spectrum peak of FeCl2 and FeCl3, corresponding to a binding energy of 198.8 and 199.0 eV. The results represent the probable generation of Cl− and reveal that in the HCl preadsorption on the α-Fe2O3 surface, chlorine species is captured, demonstrating that HCl can dissolve on αFe2O3. From the peak of Hg 4f in the HCl+α-Fe2O3+Hg0 sample, the binding energy at 101.1 eV shows the appearance of Hg2+, which matches with the available peak of HgCl2, corresponding

Figure 7. XPS spectra of Hg 4f and Cl 2p for the contrasted samples with 1 h at 80 °C.

to the binding energy of 101.4 eV. Thus, we confirm that Hg0 was captured over the α-Fe2O3 sorbents with HCl by chemical reaction. However, the binding energy of HgO is 100.6 eV, suggesting that it is possible to form mercury oxide in the reaction process. In the HCl+α-Fe2O3+Hg0 sample at 80 °C, according to the standard spectra peaks of HCl and FeCl3 at respectively 197.5 and 199.0 eV, the peak at 197.7 and 199.1 eV represents the physisorbed HCl as well as the chlorine species in Hg0absorbed samples. As reaction time is prolonged, a large E

DOI: 10.1021/acs.est.5b06340 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Article

Environmental Science & Technology

(11) Wu, S.; Ozaki, M.; Uddin, M. A.; Sasaoka, E. Development of iron-based sorbents for Hg0 removal from coal derived fuel gas Effect of hydrogen chloride. Fuel 2008, 87, 467−474. (12) Laudal, D. L.; Brown, T. D.; Nott, B. R. Effects of flue gas constituents on mercury speciation. Fuel Process. Technol. 2000, 65, 157−165. (13) Presto, A. A.; Granite, E. J. Survey of catalysts for oxidation of mercury in flue gas. Environ. Sci. Technol. 2006, 40 (18), 5601−5609. (14) Guo, P.; Guo, X.; Zheng, C. Computational insights into interactions between Hg species and α-Fe2O3(001). Fuel 2011, 90 (5), 1840−1846. (15) Tao, L.; Guo, X.; Zheng, C. Density functional study of Hg adsorption mechanisms on α-Fe2O3 with H2S. Proc. Combust. Inst. 2013, 34 (2), 2803−2810. (16) Serre, S. D.; Silcox, G. D. Adsorption of elemental mercury on the residual carbon in coal fly ash. Ind. Eng. Chem. Res. 2000, 39 (6), 1723−1730. (17) Kong, F.; Qiu, J.; Liu, H.; Zhao, R.; Ai, Z. Catalytic oxidation of gas-phase elemental mercury by nano-Fe2O3. J. Environ. Sci. 2011, 23 (4), 699−704. (18) Ghorishi, S. B.; Lee, C. W.; Jozewicz, W. S.; Kilgroe, J. D. Effects of fly ash transition metal content and flue gas HCl/SO2 ratio on mercury speciation in waste combustion. Environ. Eng. Sci. 2005, 22 (2), 221−231. (19) Galbreath, K.; Zygarlicke, C.; Tibbetts, J. E.; Schulz, R. L.; Dunham, G. E. Effects of NOx, α-Fe2O3, γ-Fe2O3, and HCl on mercury transformations in a 7-kW coal combustion system. Fuel Process. Technol. 2005, 86 (4), 429−448. (20) Wilcox, J.; Robles, J.; Marsden, D. C.; Blowers, P. Theoretically predicted rate constants for mercury oxidation by hydrogen chloride in coal combustion flue gases. Environ. Sci. Technol. 2003, 37 (18), 4199− 4204. (21) Meng, S.; Duan, Y.; Huang, Z.; Wang, Y.; Yang, L. Effect of flue gas components on mercury adsorption by coal-fired fly ash. Proc. CSEE 2009, 20, 12. (22) Smith, C. A.; Krishnakumar, B.; Helble, J. J. Homo- and heterogeneous mercury oxidation in a bench-scale flame-based flow reactor. Energy Fuels 2011, 25 (10), 4367−4376. (23) Xue, L.; Liu, T.; Guo, X.; Zheng, C. Hg oxidation reaction mechanism on Fe2O3 with H2S: Comparison between theory and experiments. Proc. Combust. Inst. 2015, 35 (3), 2867−2874. (24) Payne, M. C.; Teter, M. P.; Allan, D. C.; Arias, T. A.; Joannopoulos, J. D. Iterative minimization techniques for ab initio total-energy calculations: molecular dynamics and conjugate gradients. Rev. Mod. Phys. 1992, 64 (4), 1045. (25) Liu, T.; Xue, L.; Guo, X.; Zheng, C. DFT study of mercury adsorption on α-Fe2O3 surface: Role of oxygen. Fuel 2014, 115, 179− 185. (26) Yin, Q.; Chen, S.; Zhang, Y.; Wang, X. Removing mercury by modified activated carbon fibers. Chin. J. Chem. Eng. 2012, 6 (8), 2764−2768. (27) Gao, H.; Zhou, J.; Luo, Z.; Cen, K. Experimental study on Hg vapor adsorption of modified activated carbons in simulated flue gas. Proc. CSEE 2007, 27 (8), 26−30. (28) Ren, J.; Zhou, J.; Luo, Z. The application of novel sorbents for mercury vapor removal from simulated flue gases. Proc. CSEE 2007, 27 (2), 48−53. (29) Liu, T.; Xue, L.; Guo, X. Study of Hg0 removal characteristics on Fe2O3 with H2S. Fuel 2015, 160 (15), 189−195. (30) Bukhtiyarova, G. A.; Bukhtiyarov, V. I.; Sakaeva, N. S.; Kaichev, V. V.; Zolotovskii, B. P. XPS study of the silica-supported Fecontaining catalysts for deep or partial H2S oxidation. J. Mol. Catal. A: Chem. 2000, 158 (1), 251−255. (31) Gurgul, J.; Łątka, K.; Hnat, I.; Rynkowski, J.; Dzwigaj, S. Identification of iron species in FeSiBEA by DR UV−vis, XPS and Mö ssbauer spectroscopy: Influence of Fe content. Microporous Mesoporous Mater. 2013, 168, 1−6.

amount of HCl is continuously pumped into the reaction tube, probably leading to detection of residues of excess HCl. The appearance of Cl− is consistent with the above computational and experimental conclusions on the mechanism of Hg0 removal in the presence of HCl. Further research is still needed to study the mercury and chlorine species in more detail.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.est.5b06340. Information regarding models and computational methods, adsorption behaviors of HCl on αFe2O3(001), characterization of catalyst, experimental apparatus and procedures, X-ray photoelectron spectroscopy study, Tables S1 and S2, and Figures S1−S8 (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*Phone: 86-27-87545526. Fax: 86-27-87545526. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Financial support was sponsored by the National Key Basic Research and Development Program, The National Natural Science Foundation of China (NSFC) (Grant Nos. 51176058), and the partial funding from the Ministry of Science and Technology, China (2014CB238904, 2013CB228504).

■

REFERENCES

(1) Agnan, Y.; Le Dantec, T.; Moore, C. W.; Edwards, G.; Obrist, D. New constraints on terrestrial surface-atmosphere fluxes of gaseous elemental mercury using a global database. Environ. Sci. Technol. 2016, 50 (2), 507−524. (2) Kessler, R. The Minamata Convention on Mercury: A first step toward protecting future generations. Environ. Health Persp 2013, 121 (10), A304. (3) Selin, N. E. Global biogeochemical cycling of mercury: A review. Annu. Rev. Env. Resour. 2009, 34 (1), 43. (4) Pavlish, J. H.; Sondreal, E. A.; Mann, M. D.; Olson, E. S.; Galbreath, K. C.; Laudal, D. L.; Benson, S. A. Status review of mercury control options for coal-fired power plants. Fuel Process. Technol. 2003, 82 (2), 89−165. (5) Agarwal, H.; Stenger, H. G. Development of a predictive kinetic model for homogeneous Hg oxidation data. Math. Comput. Model. 2007, 45 (1), 109−125. (6) Miller, S. J.; Dunham, G. E.; Olson, E. S.; Brown, T. D. Flue gas effects on a carbon-based mercury sorbent. Fuel Process. Technol. 2000, 65, 343−363. (7) Hall, B.; Schager, P.; Weesmaa, J. The homogeneous gas phase reaction of mercury with oxygen, and the corresponding heterogeneous reactions in the presence of activated carbon and fly ash. Chemosphere 1995, 30 (4), 611−627. (8) Niksa, S.; Helble, J. J.; Fujiwara, N. Kinetic modeling of homogeneous mercury oxidation: The importance of NO and H2O in predicting oxidation in coal-derived systems. Environ. Sci. Technol. 2001, 35 (18), 3701−3706. (9) Galbreath, K. C.; Zygarlicke, C. J. Mercury transformations in coal combustion flue gas. Fuel Process. Technol. 2000, 65, 289−310. (10) Lu, D. Y.; Granatstein, D. L.; Rose, D. J. Study of mercury speciation from simulated coal gasification. Ind. Eng. Chem. Res. 2004, 43 (17), 5400−5404. F

DOI: 10.1021/acs.est.5b06340 Environ. Sci. Technol. XXXX, XXX, XXX−XXX