NAC

Fuel 202 (2017) 328–337

Contents lists available at ScienceDirect

Fuel journal homepage: www.elsevier.com/locate/fuel

Full Length Article

Influence of textures, oxygen-containing functional groups and metal species on SO2 and NO removal over Ce-Mn/NAC Fang Ning-Jie a, Guo Jia-Xiu a,b,⇑, Shu Song a, Li Jian-Jun a,b, Chu Ying-Hao a,b a b

College of Architecture and Environment, Sichuan University, Chengdu 610065, China National Engineering Research Center for Flue Gas Desulfurization, Chengdu 610065, China

a r t i c l e

i n f o

Article history: Received 13 February 2017 Received in revised form 6 April 2017 Accepted 8 April 2017

Keywords: Activated carbon Manganese Cerium Desulfurization Denitration

a b s t r a c t A series of Ce and Mn bimetal modified activated carbon were prepared by excessive impregnation method. The removal of SO2 or NO was carried out at a fixed bed reactor under a simulated flue gas. The results showed that Ce(1)-Mn/NAC exhibits excellent SO2 removal capacity at 80 °C with breakthrough sulfur capacity of 113 mg/g, and Ce(7)-Mn/NAC displays a good NO removal capacity and maintains 100% removal efficiency in the range of 130–220 °C. The specific surface area and pore volume decline first and then increase with the increase of Ce loadings, and the pore sizes of samples center at 0.4–4 nm. C@O and O@CAO groups are favorable for SO2 oxidation, and CAO can promote the oxidation of SO2 and NO. The AOH groups can promote the adsorption and oxidation of SO2 and NO, but the ways that they acted are different. Mn2O3, Mn3O4, Ce2O3 and CeO2 are observed in fresh samples, but MnO2 and Ce2(SO4)3 are detected and Ce2O3 disappears after desulfurization. MnO2 appears after denitration when Ce loading amount is small, but it disappears gradually with the increase of Ce loadings. CeO2 gradually decreases while Ce2O3 is increasing after denitration, indicating that CeO2 can convert into Ce2O3 in NO removal process. Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction The emission of SO2 and NOx from burning of fossil fuels has brought many environmental problems, such as haze, acid rain and photo–chemical smog. Separate removal of SO2 or NOx has been widely studied. For example, activated carbon/coke can be used to remove SO2 from flue gas [1]. Selective catalytic reduction (SCR) with NH3 has been mainly applied for NO removal at high temperature (350–420 °C) over V2O5–WO3(MoO3)/TiO2 [2]. Many measures have been considered to improve the removal ability of SO2 and NO. Liu et al. [3] have found that Fe modified activated carbon shows excellent removal ability of SO2. When rare earth oxides are loaded onto the activated semi-coke surface, a high NO conversion is obtained at low temperature [4]. However, separate control of SO2 and NO results in high operational costs. Catalyst deactivation and NH3 escape will seriously affect the life time and catalytic ability of catalysts. So, simultaneous removal of SO2 and NO has caught great interesting, including the electron beam with ammonia (EBA), the pulse corona-induced plasma chemical process

⇑ Corresponding author at: College of Architecture and Environment, Sichuan University, Chengdu 610065, China. E-mail address: [email protected] (J.-X. Guo). http://dx.doi.org/10.1016/j.fuel.2017.04.035 0016-2361/Ó 2017 Elsevier Ltd. All rights reserved.

(PPCP), the activated coke (AC) methods. Regretfully, when catalyst is installed at upstream of dust control and desulfurization, these technologies do not fundamentally solve the catalyst poisoning because of dust and SO2 with high concentration. Therefore, separate deSO2/NOx at low temperature in one device will be promising and gradually gets more attention. Recently, Mn oxides, as active components, can highly disperse on an inert carrier with a large surface to remove SO2 and NOx because of labile oxygen atoms [5]. Activated carbon (AC) materials have attracted much attention due to huge surface area, rich pore structure and oxygen-containing functional groups on the carbon surface [3]. AC can selectively adsorb SO2 and NOx and achieve a high performance even at low concentrations [6]. When Mn oxides loaded onto the carbon materials, they can improve SO2 removal ability [7,8] and exhibit better NH3–SCR activity at low temperature and a more extensive operating temperature window [7]. However, they easily lose activity because NH3 can react with SO2 to form ammonium sulfates in the presence of O2 and H2O and deposits on the catalyst surface [8]. Therefore, further efforts must be carried out to improve SO2 resistance of SCR catalysts. Ceria (CeO2), as an oxygen storage material, can store/release oxygen during the redox process due to the Ce4+ M Ce3+ as well as the transfer of electrons, ions and oxygen [9,10]. Ceria-based materials

329

N.-J. Fang et al. / Fuel 202 (2017) 328–337

are commonly used as catalysts and promoters in several heterogeneous reactions [11,12]. Some studies have shown that Ce–W catalyst exhibits high catalytic performance of NO [13] and CeO2 shows good SO2 resistance during NH3–SCR process [14]. In order to obtain better performance of denitration, Mn and Ce are simultaneously loaded on TiO2 to remove NOx. They achieve a 92% conversion of NO at 120 °C [15], and their interactions are critical for sulfur resistance [16]. The sulfur-resistant behavior is improved remarkably because of Ce doping [17]. When AC is used as carrier to load Mn and Ce, NO conversion maintains at 99% in the range of 120–250 °C and the synergistic effects of Mn and Ce improve greatly the catalytic ability of NO comparing to singe metal oxides [18]. However, to the best of our knowledge, there are few reports for the removal of SO2 by Ce and Mn bimetal modified AC, and the main role of Ce and Mn in desulfurization and denitration is scarce. Therefore, this work is to explore the removal capacity of SO2 or NO on Ce and Mn bimetal modified AC treated by nitric acid. The influence factors on catalytic ability are also discussed when catalyst bed temperatures are only changed, including oxygencontaining functional groups, the adsorption diameter of gas molecules and metal oxides species. The main role of Ce and Mn in desulfurization and denitration is also involved. The results will significantly improve the SO2 and NO removal technology in flue gas. In this paper, manganese oxides and cerium oxides with different Ce/Mn ratios were impregnated on AC treated by nitric acid. The physical and chemical properties of samples were characterized by N2 adsorption/desorption isotherms, X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR) and X-ray photoelectron spectroscopy (XPS) before and after SO2 and NO removal. 2. Experimental 2.1. Pretreatment of activated carbon All samples were prepared by impregnation method. All chemicals used in this work were analytic reagent (AR). The original activated carbon, purchasing from Henan Changge Chemical Co., Ltd., China, was ground and sieved to 10–20 mesh. These carbons were immersed in distilled water and boiled in an electric furnace for 30 min. After that, the sample was filtered and washed in distilled water until the washing liquid became neutral. The sample was dried in an oven at 105 °C for 12 h, and the washed carbon was named as AC. The washed AC was completely immersed into 39 wt% HNO3 solution and heated at 60 °C for 2 h in water bath. Then, the sample was filtered and washed with distilled water until the washing liquid became neutral. Finally, the sample treated by HNO3 was dried in an oven at 105 °C for 12 h and denoted as NAC. 2.2. Catalyst preparation Manganese nitrate and cerium nitrate were used as precursors. The NAC was completely immersed into manganese nitrate solution with an appropriate concentration to achieve 5 wt% Mn loading. During the impregnation, the mixture was stirred constantly for 30 min and placed stilly for 12 h. The solution was heated at 60 °C with stirring until the liquid was totally eliminated. The sample was dried at 110 °C for 12 h and then calcined at 650 °C for 3 h in pure N2 atmosphere. The obtained sample was named as Mn(5)/ NAC. The obtained sample was immersed into the Ce(NO3)3 solution with different cerium concentration to achieve 1, 3, 5, 7 and 9 wt% Ce loading and calcined at 450 °C in a pure N2 atmosphere for 3 h. The other treatment process is the same as the Mn(5)/NAC. They were

labeled as Ce(1)-Mn/NAC, Ce(3)-Mn/NAC, Ce(5)-Mn/NAC, Ce(7)Mn/NAC and Ce(9)-Mn/NAC. As a comparative sample, 5 wt% Ce loaded on NAC was prepared and named as Ce(5)/NAC. 2.3. Activity evaluation SO2 or NO removal testing was carried out at a fixed-bed reactor under atmosphere pressure for all samples. The inner diameter of the reactor is 18 mm. For SO2 removal testing, 8 g samples were filled into the reactor, and the filling height was 70 mm. The simulated flue gas contained 2700 ppm SO2, 10% O2, 10% water vapor and N2 as the balance. These gases were fully mixed in the mixing bottle before entering the reactor. The gas flow rate was 800 mL/ min. The reaction temperature was 80 °C. The gas hourly space velocity (GHSV) was 2700 h
1. NaOH solution was used to absorb the SO2 in the tail gas. When the concentration of outlet SO2 reaches 200 ppm, it could be considered as breakthrough. Breakthrough sulfur capacity is usually defined as the accumulated quantity of SO2 removal per unit mass of catalyst at breakthrough point. The corresponding working time is regarded as the breakthrough time. Breakthrough capacity of each sample is calculated from the SO2 inlet concentration, working time, flow rate and mass of sample. The concentration of SO2, and O2 in the stimulated gas before and after the reactor were analyzed on-line by Gasboarb 3000 gas analyzer (Wuhan Cubic Optoelectronics Co., Ltd. Wuhan, China). For NO removal testing, 3.5 g samples were filled into the reactor, and the filling height was 28 mm. The simulated flue gas mixture contained 530 ppm NO, 530 ppm NH3, 5% O2 and N2 as the balance. The total flow rate was 600 mL/min, corresponding to a GHSV of 5000 h
1. The simulated gas was fully mixed in the mixing bottle. The range of reaction temperature was from 80 to 220 °C, and at each temperature point, it was maintained for half an hour. When the reaction proceeded steadily at each temperature point, the data of outlet NO concentration were collected. The NOx in tail gas was absorbed by NaOH solution. The concentration of NO and O2 in the stimulated gas before and after the reactor were analyzed on-line by Gasboarb 3000 gas analyzer (Wuhan Cubic Optoelectronics Co., Ltd. Wuhan, China). The removal efficiency (%) of NO is obtained by analyzing the NO inlet and outlet concentration as the following formula:

NO removal efficiencyð%Þ ¼

½NOin - ½NOout  100% ½NOin

2.4. Catalyst characterization N2 adsorption–desorption isotherms of the samples at
196 °C were measured on AUTOSORB-IQ adsorption apparatus (Quantachrome Instruments, USA) and high purity N2 was used as adsorbate. Prior to analysis, each sample was outgassed at 250 °C and 10
5 torr for 6 h. The adsorption isotherms were used to calculate the surface area (SBET) of each sample using the Brunauer–Emmet t–Teller equation in the range of P/P0 from 0.05 to 0.35. The total pore volume (Vtotal) was calculated corresponding to the nitrogen adsorption amount at P/P0 = 0.95. The micropore volume (Vmicro) and pore size distribution were calculated by Dubinin–Radushkevich (D–R) equation and Horvath–Kawazoe (H–K) model, respectively [19]. Crystal structure of the samples was detected by a powder X-ray diffraction on a DX-2700 diffractometer (Dandong Haoyuan Instrument Co., Ltd., China) using Cu Ka radiation (k = 0.15418 nm) and operating at 40 kV and 30 mA. The samples were scanned in the range of 2h from 10° to 80° with the step of 0.03°. The crystalline phases were identified by comparing the

330

N.-J. Fang et al. / Fuel 202 (2017) 328–337

reference data from the international center for diffraction data (JCPDs). Surface functional groups of the samples were investigated through a Nicolet 6700 spectrometer (Thermo Electron Co., USA). The wave number was in the range of 4000–400 cm
1, using KBr as a tablet. X-ray photoelectron spectroscopy (XPS) experiments were conducted on a spectrometer (Thermo Scientific EscaLab250Xi) using Mg Ka radiation under ultra high vacuum (UHV) operating at 12 kV and 15 mA. Energy calibration was done by recording the core level spectra of Au 4f7/2 (84.0 eV) and Ag 3d5/2 (368.30 eV). Peak areas including satellites were computed using a program which assumed Gaussian–line shapes and flat background subtraction. 3. Results and discussion 3.1. Desulfurization performance The relationships between outlet SO2 concentration and working time at 80 °C are shown in Fig. 1. The breakthrough sulfur capacity of all samples is summarized in Table 1. As shown in Fig. 1, for all samples, the outlet SO2 concentration gradually increases with working time prolonging, showing that the deactivation of samples is a gradual progress. A reasonable explanation is that reactive molecules, such as SO2 and O2, are adsorbed on the active sites to generate SO3. The generated SO3 react with adsorbed H2O and form H2SO4. Some of the generated H2SO4 will deposit on the surface or into the pore of catalysts to cover active sites or block the reaction channel. Some is removed by excessive condensed H2O to release active sites. Both factors will lead to a gradual deactivation of the catalyst. When working time is extended, more H2SO4 is deposited, leading to the complete deactivation of the catalyst. It is found that NAC shows poor SO2 removal ability and has breakthrough sulfur capacity of 58 mg/g. This is because NAC treated by nitric acid has abundant oxygencontaining functional groups which can act as active centers in multifunctional samples because of their acid–base or red–ox properties [20,21]. When Mn or Ce is loaded onto NAC, SO2 removal capacity has been greatly enhanced and breakthrough sulfur capacity increases to 107 mg/g for Mn(5)/NAC and 67 mg/g for Ce(5)/NAC. This result is probably related to the catalytic activity of metal oxide species (active sites) during the SO2 oxidation process. One possible explanation is that the lattice oxygen atoms in metal

NAC Mn(5)/NAC Ce(5)/NAC Ce(1)-Mn/NAC Ce(3)-Mn/NAC Ce(5)-Mn/NAC Ce(7)-Mn/NAC Ce(9)-Mn/NAC

Outlet SO2 concentration (ppm)

200

150

100

0 10

20

30

40

50

60

Samples

Breakthrough sulfur capacity (mg/g)

NAC Mn(5)/NAC Ce(5)/NAC Ce(1)-Mn/NAC Ce(3)-Mn/NAC Ce(5)-Mn/NAC Ce(7)-Mn/NAC Ce(9)-Mn/NAC

58 107 67 113 101 90 76 104

oxides can act as a supplier of active oxygen which firstly reacted with the absorbed SO2. Oxygen in gas phase is absorbed to the reductive metal oxides and converted into lattice oxygen. In this way, the metal oxides promote the oxygen transfer to improve the removal capacity of SO2. Furthermore, the sulfur capacity of Mn(5)/NAC is almost twice of NAC. This is due to various oxide species of Mn, such as MnO, Mn2O3 and Mn3O4. These oxides play an important role in the catalytic oxidation of SO2. The sulfur capacity of Ce(5)/NAC increases slightly compared to NAC, which may be caused by the following reaction: 2CeO2 + 3SO2 + O2 ? Ce2(SO4)3 [22]. The generated Ce2(SO4)3 will be deposited on catalyst surface to cover active sites or block channels. In a word, Mn- or Ce-doping into NAC can improve the sulfur capacity. The former is better than the latter, suggesting that Mn oxides are more conducive to improving the SO2 removal ability of NAC. When Ce is added to Mn modified NAC, the SO2 removal ability of Ce-Mn/NAC is better than Ce/NAC but similar to or worse than Mn/NAC. The breakthrough sulfur capacity of Ce(1)-Mn/NAC is 113 mg/g, which is about 1.7 times of Ce(5)-NAC and close to the sulfur capacity of Mn(5)/NAC. It indicates that Mn oxides play a dominant role for SO2 removal in Ce-Mn/NAC. When the Ce loading increases from 1 to 7 wt%, the sulfur capacity of samples decreases obviously. The breakthrough sulfur capacity of Ce(7)-Mn/NAC is only 76 mg/g, which is much lower than Mn(5)/NAC and close to Ce (5)/NAC. The excessive Ce is considered as the hamper for SO2 adsorption by covering the carbon surface, resulting in poor catalytic efficiency. It is worth noticing that the sulfur capacity of Ce (9)-Mn/NAC increases to 104 mg/g but is worse than Ce(1)-Mn/ NAC. This may be concluded that the CeO2 in Ce(9)-Mn/NAC plays a crucial role and more Ce2(SO4)3 is regenerated in SO2 removal, causing the increase of sulfur capacity. The results show that the SO2 removal order of samples is as follows: NAC < Ce(5)/NAC < Ce (7)-Mn/NAC < Ce(5)-Mn/NAC < Ce(3)-Mn/NAC < Ce(9)-Mn/NAC < Mn(5)/NAC < Ce(1)-Mn/NAC. 3.2. Denitration performance

50

0

Table 1 Breakthrough sulfur capacity of all samples.

70

80

90 100 110 120 130 140

Breakthrough time (min) Fig. 1. Relationship between outlet SO2 concentration and working time of all samples at 80 °C.

In Fig. 2, NAC shows poor NO catalytic activity and NO removal rate of 47% at 80 °C. When the temperature increases, NO removal efficiency of NAC increases gradually and reaches 90% at 220 °C. This suggests that NAC has certain catalytic ability at low temperature because of surface functional groups. When Mn or Ce is added to NAC, NO removal efficiency has been improved greatly. Mn(5)/NAC achieves a NO removal efficiency of 75% at 80 °C and 90% at 140 °C. When the temperature increases to 220 °C, NO removal efficiency increases to 98%. In the case of Ce(5)/NAC, NO removal efficiency is higher than 93% at 80 °C and maintains 100% when temperature is from 150 to 220 °C, indicating that Ce oxide has excellent NO removal ability. Therefore, Mn or Ce oxides can improve the denitration efficiency, especially Ce oxides. This may be that NO is first oxidized to NO2 and then reduced by NH3 to N2 and H2O. Here, Mn or Ce oxides play a role as an oxygen supplier, and oxygen transfer in Ce oxides is more efficient.

N.-J. Fang et al. / Fuel 202 (2017) 328–337

331

3.3. Characterizations

(NO3)3 deposited on NAC are decomposed into metal oxides under high calcination temperature and releases some gases. These released gases can expand pore structures of NAC. Besides, the generated oxides are deposited on NAC surface, preventing N2 entering the channel interior. These factors cause the difference of pore structure. When 1 wt% Ce is loaded onto Mn/NAC, SBET, VTotal and VMicro of Ce(1)-Mn/NAC decease to 683 m2/g, 0.487 cm3/g and 0.280 cm3/g. The SBET and VTotal of Ce(7)-Mn/NAC increase to 668 m2/g and 0.681 cm3/g, and it possesses a large mesopore volume and average pore size, 0.405 cm3/g and 4.08 nm. Ce(1)-Mn/ NAC has better SO2 removal ability than Ce(7)-Mn/NAC in desulfurization, but the latter shows better NO removal efficiency than the former in denitration, indicating that micropores are in favor of desulfurization while mesopores can promote NO adsorption. According to the potential energy theory, smaller pore size corresponds to stronger adsorption capacity. Metal oxides and reaction gases are easily absorbed in the pores. When the loading of metal oxides increases, the surface of carriers will be covered gradually. When they exceed to a contain value, pores will be blocked, which hinders the gases entering the active sites during reaction. When pore size becomes larger, the adsorption capacity will reduce. If there is no blockage in the pores, the gas molecules can enter into the active sites and a high catalytic ability is achieved. The catalytic ability has a close relationship with pore size because of pore size effects. Catalysts with porous structures can promote gas diffusion into internal spaces, which plays an important role for SO2 or NO removal [24]. When pore diameter is small, more carbon surface will contact with SO2 and shorten the diffusion distance, leading to a fast adsorption. There exists an appropriate ratio between the adsorption of gas molecules and the pore size during SO2 or NO adsorption process. The pore size distribution of the samples is illustrated in Fig. 3. All samples have a similar pore size distribution and the primary peak appears at about 0.6 nm, showing that micropores are dominant. In addition, two secondary peaks located at 1.1 and 2.6 nm are also observed, implying that there are some mesopores. For Mn(5)/NAC and Ce (1)-Mn/NAC, the peak height of micropore is similar and larger than the others. They show a good SO2 removal efficiency, indicating that micropore is conducive for the removal of SO2. SO2 molecules are absorbed on the pore walls and oxidized to SO3 in the presence of O2, followed by generating H2SO4 with H2O. At the same time, the H2SO4 is easy to be dissolved in water and taken away by gas flow [25], such as excessive water vapor. In this way, the active sites are released and more SO2 can be adsorbed, leading to an excellent SO2 adsorption capacity. Ce(7)-Mn/NAC shows its highest peak at 2.6 nm and a good NO removal efficiency, indicating that mesopore is helpful for NO adsorption. This is because mesopores are mainly used for the diffusion channel of gaseous molecules and the storage for the byproducts. When NO is oxidized to NO2, it will be reduced to N2 by NH3 and taken away by gas flow.

3.3.1. Textural properties At present, the removal mechanism of SO2 on AC surface is as follows: SO2 + C ? C-SO2; C-SO2 + O2 ? C-SO3; C-SO3 + H2O ? C-H2SO4. These reactions are closely related to porous structure of activated carbon. The N2 adsorption–desorption isotherms of all samples are presented in Fig. 3. The surface area (SBET), total pore volume (VTotal), micropore volume (VMicro) and average pore size are listed in Table 2. As shown in Fig. 3, the shapes of all isotherms show the characteristic of microporous materials, and a small hysteresis loop is observed, meaning the existence of mesopores. In Table 2, NAC has surface area of 769 m2/g and total pore volume of 0.516 cm3/g with micropore volume of 0.300 cm3/g. For Mn(5)/NAC and Ce(5)/NAC, the SBET and VTotal decrease while average pore size shows an increase. This may be that Mn(NO3)2 and Ce

3.3.2. Oxygen containing functional groups In order to explore the changes of oxygen-containing functional groups of the samples in SO2 catalytic oxidation and NO reduction with NH3, FTIR and XPS are carried out. As shown in Fig. 4(a), all fresh samples have a broad band in the range of 3050– 3600 cm
1 with a maximum near 3423 cm
1, corresponding to the OAH stretching vibration of hydroxyl groups in chemically adsorbed water, which is originated from KBr. The peaks at 1560 cm
1 is attributed to C@C stretching vibration. The observed broad bands in the range of 1300–900 cm
1 correspond to CAO stretching in the alcohols, carboxylic acid or carboxylic groups. A weak peak at 760–630 cm
1 is still observed, implying that AOH stretching of alcohol and phenol exists on the activated carbon. CAOA oxygen play a role in the desulfurization reaction and may

NO removal efficiency (%)

100

90

80

NAC Mn(5)/NAC Ce(1)-Mn/NAC Ce(3)-Mn/NAC Ce(5)-Mn/NAC Ce(7)-Mn/NAC Ce(9)-Mn/NAC Ce(5)/NAC

70

60

50 80

100

120

140

160

180

200

220

Time (min) Fig. 2. Relationship between NO conversion and temperature of all samples.

The introduction of Ce to Mn(5)/NAC can obviously improve the NO removal efficiency. When the Ce loading increases from 1 to 9 wt%, the NO removal ability increases firstly and then declines. For Ce(5)-Mn/NAC, the removal efficiency of NO achieves 85% at 80 °C and increases to 100% at 130 °C. When Ce loading increases to 7 wt %, Ce(7)-Mn/NAC exhibits a good NO removal ability, 93% at 80 °C and 99% at 110 °C. At the same time, it can maintain 100% removal efficiency of NO when temperature is from 130 to 220 °C. It displays that Ce oxides could improve the catalytic ability and decrease the activation temperature. This is because Ce4+ in Ce oxides will lose oxygen atom during NO removal to form oxygen defects and converts to Ce3+. The released oxygen can be captured by NO to form NO2. The generated NO2 is reduced by the reductive agent, such as NH3. The reducible Ce oxides are to absorb oxygen in gas phase and form oxidative Ce oxides. In this way, a redox cycle is established. However, when Ce loading increases to 9 wt%, NO removal ability of Ce(9)-Mn/NAC decreases. This is because there is a monolayer dispersion capacity when metals oxide dispersed on a support [23]. When its loading is lower than the capacity, the oxide disperses as a monolayer. When its loading exceeds the monolayer dispersion capacity, the surplus oxide will cover the catalyst surface to block the pores or to occupy the active sites. They influence the adsorption of reactive gas, resulting in poor NO removal efficiency. However, there exists a synergistic effect between Mn and Ce, which can improve NO removal ability of Ce-Mn/NAC. The NO removal order from excellent to poor is as follows: Ce(7)-Mn/NAC > Ce(5)-Mn/NAC > Ce(5)/NAC > Ce(9)-Mn/ NAC > Ce(3)-Mn/NAC > Ce(1)-Mn/NAC > Mn(5)/NAC > NAC.

332

N.-J. Fang et al. / Fuel 202 (2017) 328–337 500 400

0.4

NACMn(5)/NAC

Mn(5)/NAC

NAC

0.3

300

0.2 200

0.1

100

0.0

500 0 400

Ce(5)/NAC

Ce(1)-Mn/NAC

Ce(3)-Mn/NAC

Ce(5)-Mn/NAC

Ce(5)/NAC

Ce(1)-Mn/NAC

Ce(3)-Mn/NAC

Ce(5)-Mn/NAC

Ce(7)-Mn/NAC

Ce(9)-Mn/NAC

300

100 500 0 400

0.4

Pore size distribution (cm 3/g.nm)

3

Pore volume (cm /g)

200

300 200 100 0 500 400

Ce(9)-Mn/NAC

Ce(7)-Mn/NAC

300 200 100 0 0.0

0.2

0.4

0.6

0.8

1.0 0.0

0.2

0.4

0.6

0.8

0.3 0.2 0.1 0.0 0.4 0.3 0.2 0.1 0.0

1.0

1

Relative pressure (P/P)

2

3

4

5

6

0

7

8

9 0

1

2

3

4

5

6

7

8

9

Pore width (nm)

Fig. 3. N2 adsorption-desorption isotherms (left) and pore size distribution (right) of all samples.

Table 2 Surface area, total pore volume, micropore volume and average pore size of all samples. Samples

SBET (m2/g)

VTotal (cm3/g)

VMicro (cm3/g)

VMeso (cm3/g)

Average pore size (nm)

NAC Mn(5)/NAC Ce(5)/NAC Ce(1)-Mn/NAC Ce(3)-Mn/NAC Ce(5)-Mn/NAC Ce(7)-Mn/NAC Ce(9)-Mn/NAC

769 739 719 683 660 645 668 595

0.516 0.512 0.501 0.487 0.418 0.410 0.681 0.41

0.300 0.301 0.291 0.280 0.256 0.260 0.276 0.241

0.216 0.211 0.210 0.207 0.162 0.150 0.405 0.169

2.62 2.77 2.79 2.85 2.61 3.68 4.08 2.76

provide an electron-rich surface that can transfer electrons, which are chemically bonded to the carbon surface, to the oxygen [29]

and forming O
2 . This O2 can be further dissociated to O on the catalyst surface, because the dissociation energy barrier of oxygen molecular on oxygen-doped carbon materials is very low [26]. These oxygen species are very active and can be captured by the absorbed SO2 to form SO3. The OAH is conducive for SO2 removal because SO2 can be absorbed by surface OAH through the strong H-bonding interaction and the reaction barrier of SO2 oxidization to SO3 is reduced [36]. Besides, there is an electron transfer channel between the OH, the adsorbed SO2 and the epoxy groups, resulting in enhanced ability for SO2 removal. The SO2 removal capacity of NAC in Fig. 1 proves this point. For Mn(5)/NAC, a small peak at 1382 cm
1 is observed, which is assigned to C@O stretching vibrations in carboxylic acid. The possible explanation is that the p
p⁄ transition causes a conjugation effect, resulting in the shift of C@O stretching vibrations to the low wavenumber, corresponding to a good SO2 removal ability. This C@O group in carboxyl group is

not found or very weak in other samples. Combined with the results of SO2 removal testing, Ce(1)-Mn/NAC exhibits the best desulfurization performance. It is reasonable to deduce the presence of C@O, because C@O with Brønsted basic properties is the active center for SO2 oxidation. Since SO2 is an acidic gas, it is first adsorbed on the basic active sites when it enters the catalyst bed, and then transformed into H2SO4 in the presence of oxygen and water vapor. After desulfurization, as shown in Fig. 4(b), the functional groups on catalyst surface appear changes. The characteristic
1 peaks of SO2
) are observed for all samples, proving 4 (1120 cm that SO2 is converted to the sulfate eventually. New weak peaks at 1383 cm
1 attributed to the vibration of C@O are observed, which is due to the dissociative adsorption of O2. This may be that after the dissociation of O2, one O atom of the O2 molecule is connected with the two C atoms to form an epoxy group, and the other O atom with one C atom to form a carbonyl group [26] or to oxidize reductive metal species. For Mn(5)/NAC after desulfurization, the peak at 1382 cm
1 disappears but a new peak at 1026 cm
1 is

N.-J. Fang et al. / Fuel 202 (2017) 328–337

a

-OH

Transmiance (%)

NO. The removal of NO at low temperature is divided into two steps: NO and O2 react with each other to form NO2, and the produced NO2 is further reduced to N2 by NH3 [28]. During the oxidation of NO, the adsorption of NO is the key step. CAO promotes the oxygen chemically bonded to the carbon surface [29] and form O
2. This O
2 can be further dissociated. NO is firstly adsorbed on the active sites and then oxidized to NO2 by the dissociative oxygen. The kinetic mechanism for the oxidation of NO is as follows (Cf represents for active site) [30]:

C=C

b

C-O

c d e f g

C=O

NO þ Cf ! C
NO

(a)

O2 þ 2Cf ! 2C
O -1

Transmiance (%)

Wave number (cm )

a b c d e f g

C=C

-OH

C
NO þ C
O ! C
NO2 þ Cf C
NO2 þ C
NO2 ! C
NO3 þ NO þ Cf

C=O SO 42-

C
NO3 þ C
NO ! C
NO
NO3 þ Cf C
NO
NO3 ! 2NO2 þ Cf SO 42C-H C=C C-O

(b) 400 0

3500

300 0

250 0

200 0

150 0

1000

500

Wave number (cm )

Transmiance (%)

b c d

2NO2 ðadÞ þ 8H ! N2 þ 4H2 O Thus, the NO is removed and the total chemical reaction is as following:

C-O

4NO þ 4NH3 þ O2 ! 2N2 þ 6H2 O:

e f g

(c) 400 0

NH2 ðadÞ ! NHðadÞ þ H 2NHðadÞ ! N2 þ 2H

-NH 2 -NH 3

-OH

The generated NO2 is reduced by H originated from N–H cleavage of the adsorbed NH3 and form N2 and H2O. The possible chemical reactions are as follows:

NH3 ðadÞ ! NH2 ðadÞ þ H

C=O SO 4 2-1

a

333

3500

3000

2500

2000

1500

1000

500

Wave number (cm -1 ) Fig. 4. FTIR spectra of all samples before (a) and after (b) desulfurization as well as after denitration (c). (a) Ce(5)/NAC; (b) Ce(9)-Mn/NAC; (c) Ce(7)-Mn/NAC; (d) Ce (5)-Mn/NAC; (e) Ce(3)-Mn/NAC; (f) Ce(1)-Mn/NAC; and (g) Mn(5)/NAC.

observed, assigned to the CAO stretching in RAO, indicating that it may promote the conversion of SO2 to SO3 because it can reduce the reaction barrier [26]. There are two strong SO2
absorption 4 peaks (1176 and 590 cm
1), which shows that the SO2
4 ions concentration on catalyst surface is higher than the other samples. This may be attributed to sulfuric acid or sulfate, corresponding to its excellent SO2 conversion. After NO removal, as shown in Fig. 4(c), the surface oxygencontaining functional groups show some changes compared to the fresh samples. The peak at 1557 cm
1 is attributed to the scissoring vibration mode of NH2 species [27]. It indicates that NH3 is absorbed on the catalyst surface and NAH bond is broken to produce NH2 and H. This H is very important and active, which is able to reduce quickly the generated NO2. For Ce(3)-Mn/NAC and Ce(5)/ NAC, a strong diffraction peak due to ANH3 is observed at 1384 cm
1, showing that NH3 is first absorbed on the catalyst surface and does not directly participate in the reduction reaction of

C 1s and O 1s XPS spectra used to furtherly investigate the chemical property of samples are presented in Fig. 5. Deconvolution of C 1s spectra of all samples gives four peaks located at 284.3–284.6, 285.6–285.7, 287.3–287.4 and 290.1–290.3 eV, corresponding to CAC [31], CAO, C@O [32] and O@CAOH [33], respectively. It suggests that Ce doping will not change the types of functional groups. The content of CAC and CAO carbon in Ce(1)Mn/NAC has no difference before and after SO2 removal, but the C@O and O@CAOH carbon before SO2 removal is higher than the latter, meaning that they are consumed in the SO2 removal. It indicates that C@O takes part in the oxidation of SO2. This is because it can transfer electrons to the oxygen chemically bonded to the carbon surface and influence the surface attraction force and the adsorption capacity. O@CAOH can react with SO2 to form a hydrogen bond, which will reduce the energy barrier of SO2 to SO3 [34]. This may be one reason of Ce(1)-Mn/NAC with high SO2 removal. Ce(7)-Mn/NAC before NO removal has more CAC carbon but less CAO, but C@O and O@CAOH carbon is more in Ce(7)-Mn/NAC after NO removal, indicating oxygen-bonded carbon can be generated during NO removal. This is because NO is adsorbed on C atom to form CANO and react with the CAO to produce CANO2, which can be reduced by NH3 finally. As shown in Fig. 5, the deconvolution of O 1s XPS spectra presents three peaks centered at 530–530.5, 532–533 and 533.4– 533.5 eV, which are assigned to oxygen in oxides [35], C@O or CAO oxygen and AOH oxygen [25], respectively. The oxides content in Ce(1)-Mn/NAC increase compared to Mn(5)/NAC and Ce (5)/NAC. When Ce loading increases to 7 wt%, the oxides content becomes higher than Ce(1)-Mn/NAC. This is caused by the increase of Ce loading. The content of oxides and OH groups decrease from

334

N.-J. Fang et al. / Fuel 202 (2017) 328–337

Mn/NAC

284.4eV; C-C; 66.6%

Ce(1)-Mn/NAC

285.6eV; C-O, 15.6%

Intensity (a.u.)

Ce(1)-Mn/NAC-S

284.4eV; C-C; 69%

Ce(7)-Mn/NAC-N 284.6eV; C-C; 65.8%

284.4eV; C-C; 69%

285.7eV; C-O, 16.4%

285.7eV; C-O, 16.4%

287.4eV; C=O; 6.3%

285.7eV; C-O, 15.3% 287.3eV; C=O; 9.2% 290.3eV; O=C-OH; 9.7%

287.4eV; C=O; 6.3% 290.1eV; O=C-OH; 8.3%

290.1eV; O=C-OH; 8.3% 292

285.7eV; C-O, 12.3% 287.3eV; C=O; 7.9% 290.1eV; O=C-OH; 7.2%

287.4eV; C=O; 7.7% 290.3eV; O=C-OH; 7.9%

290.3eV; O=C-OH; 7.9%

296

284.3eV; C-C; 72.6%

285.6eV; C-O, 16.5%

287.4eV; C=O; 9.9%

Ce/NAC

Ce(7)-Mn/NAC

284.4eV; C-C; 69.3%

288

284

280

296

292

288

284

280

296

292

288

284

280

Binging energy (eV) Mn/NAC

Intensity (a.u.)

533.5eV; OH; 31.1%

Ce/NAC

534eV; OH;19%

532.5eV; C-O and C=O; 45.4% 530.5eV; oxides; 23.5%

Ce(1)-Mn/NAC

533.4eV; OH; 19.9%

532.2eV; Ce(7)-Mn/NAC C-O and C=O; 532eV; 44.4% C-O and C=O; 41.8% 530.5eV; oxides; 533.4eV; 35.7% OH;17.3%

Ce(1)-Mn/NAC-S 532.9eV; C-O and C=O; 66.9% 531.1eV; oxides; 14.1%

538 536 534 532 530 528

Ce(7)-Mn/NAC-N 533eV; C-O and C=O; 78.3%

534.3eV; OH;12.4%

531.4eV; oxides;9.3%

538 536 534 532 530 528

530eV; oxides; 40.9%

A

532eV; C-O and C=O; 47.9%

530.1eV; oxides; 38.1%

533.5eV; OH;14%

538 536 534 532 530 528

Binging energy (eV) Fig. 5. C 1s and O 1s XPS spectra of samples.

35.7 and 19.9% in Ce(1)-Mn/NAC to 9.3 and 12.4% in Ce(1)-Mn/ NAC-S. Instead, CAO or C@O oxygen is obvious increase. It indicates that some MnOx has converted into other substances. The OAH content of Ce(1)-Mn/NAC after desulfurization decreases, showing that OAH is involved in SO2 removal. When SO2 is adsorbed on the surface of NAC, electrons will be transferred from hydroxyl groups to the adsorbed SO2, followed to the epoxy groups, resulting in enhanced ability for SO2 removal. When O2 is adsorbed on the surface of NAC, it will react with the C surface to form CAO group. Subsequently, the formed CAO will oxidize the adsorbed SO2 to SO3 with a lower barrier. The adsorption and oxidation mechanism of SO2 on Ce-Mn/NAC is as follows:

O2 þ 2Cf ! 2C
O;SO2 þ Cf ! C
SO2 ; Cf þ H2 O ! C
H2 O

C
SO2 þ C
O ! C
SO3 þ Cf C
SO3 þ C
H2 O ! C
H2 SO4 þ Cf : After NO removal, for Ce(7)-Mn/NAC, the same results are observed. The OH can improve NO and C surface to form a strong NAC bond, resulting a good adsorption and oxidation of NO [37]. Hou et al. [38] have found that the hydroxyl group can promote the adsorption of NO on catalyst surface, which is caused by the weak hydrogen-bonding interaction between hydroxyl group and NO. The H atom in hydroxyl group will be captured by the adsorbed NO2 to form a nitrous acid like structure. Subsequently, a newly OAH bond is formed [39], which enhances the charge transfer.

335

N.-J. Fang et al. / Fuel 202 (2017) 328–337

3.3.3. Metal oxides species In order to explore the changes of Mn and Ce species during desulfurization and denitration, XRD and XPS analyses are carried out. As shown in Fig. 6(a), all fresh samples have two broad peaks at 20°–30° and 40°–50° in 2h range, corresponding to the characteristic peaks of carbon structure [40]. For Mn(5)/NAC, diffraction peaks at 32.9° and 35.6° are attributed to Mn2O3 [41] and the peak at 29.1° belongs to Mn3O4 [35], indicating that Mn2O3 and Mn3O4 coexist in our experiment condition. For Ce(5)/NAC, CeO2 and Ce2O3 are observed, indicating that the Ce(NO3)3 is decomposed into CeO2 and Ce2O3 at 400 °C. When Ce is added into Mn/NAC, the peaks of Mn2O3 species decrease or disappear. This may be that the strong interaction between Mn and Ce can further improve the

(a)

Mn3O4

o

Ce(5)/NAC

* CeO2

*

Mn2O3

o Ce2O3

Ce(9)-Mn/NAC Ce(7)-Mn/NAC Ce(5)-Mn/NAC Ce(3)-Mn/NAC

Ce(1)-Mn/NAC Mn(5)/NAC

( )

(b)

• MnO2

Ce(5)/NAC

* CeO2

Mn3O4

& Ce2(SO4)3

Ce(9)-Mn/NAC Ce(7)-Mn/NAC

dispersion of metal oxides. It is worth noticing that the characteristic peaks of CeOx in Ce-Mn/NAC catalysts are not obviously observed, showing that they are in poor crystallinity. As shown in Fig. 6(b), two broad peaks at 20°–30° and 40°–50° indicate that the carbon structure has not been destroyed in SO2 removal. Only diffraction peaks ascribed to MnO2 (2h = 28.7°, 37.1° and 42.2°) are observed in Mn(5)/NAC [42], implying that the oxidation state of Mn appears changes during SO2 removal. This is because reductive Mn oxides (such as Mn2O3 and Mn3O4) can be oxidized by O2 in the gas phase. The SO2 removal of Ce (5)/NAC is poor. This may be that oxygen transfer of Ce oxides is difficult at too low temperature (80 °C). For Ce(1)-Mn/NAC, a peak is detected at 2h = 36.5°, which corresponds to CeO2. According to Fig. 1, the SO2 removal efficiency of Ce-Mn/NAC samples are poorer than that of Mn/NAC but better than Ce/NAC, showing that CeOx has some oxidation ability for SO2, but it’s not as strong as MnOx, which is due to the poor oxygen transfer at low temperature. It suggests that Mn oxides are more conducive to SO2 removal than Ce oxides. The characteristic peaks of CeOx and MnOx in Ce-Mn/ NAC catalysts are still not obviously observed. While characteristic diffraction peaks attributed to Ce2(SO4)3 are observed at 2h = 42.0° in Ce-containing samples after desulfurization except Ce(1)-Mn/ NAC, indicating some Ce species have reacted with the generated H2SO4 when the generated H2SO4 is not timely removed from catalysts. Peaks centered at 2h = 31.4° corresponding to Ce2(SO4)3 are found in Ce(9)-Mn/NAC and Ce(5)/NAC, showing that more Ce2(SO4)3 are generated. The generation of Ce2(SO4)3 is one of the reasons for the decrease of SO2 removal efficiency. The S 2p XPS spectra of Mn/NAC, Ce/NAC and Ce(7)-Mn/NAC are shown in Fig. 7. The binding energy peaks of S 2p1/2 and 2p3/2 over all samples are at around 164.4 and 169.2 eV. The former is attributed to the sulfur in carbon structure [34,43] and the latter corresponds to the S6+ in SO2
4 [23,44]. It is obvious that SO2 is oxidized to sulfuric acid or sulfate. During the SO2 removal, the possible reactions are as follows:

Mn2 O3 þ Mn3 O4 þ SO2 þ 2O2 þ H2 O ! 5MnO2 þ H2 SO4 &

Ce(5)-Mn/NAC

Ce2 O3 þ SO2 þ O2 þ H2 O ! CeO2 þ H2 SO4

&

Ce(3)-Mn/NAC

2CeO2 þ 3SO2 þ O2 ! Ce2 ðSO4 Þ3 : Ce(1)-Mn/NAC Mn(5)/NAC

The XRD patterns after NO removal are shown in Fig. 6(c). MnO2 appears after denitration in Mn(5)/NAC and Ce(1)-Mn/NAC, but it disappears gradually with the increase of Ce loading. This may be that CeO2 is more sensitive to NO oxidation than Mn oxides. The peaks of CeO2 are not detected in all samples, indicating CeO2 is

• •

•&

(c) Ce(5)/NAC

Mn3O4

Mn2O3

• ΜnO2

* CeO2

169.2

Ce(1)-Mn/NAC

Ce(9)-Mn/NAC

164.4

Intensity (a.u.)

Ce(7)-Mn/NAC Ce(5)-Mn/NAC Ce(3)-Mn/NAC Ce(1)-Mn/NAC

Mn(5)/NAC

• * •

•

( ) Fig. 6. XRD patterns before (a), after (b) SO2 removal and after NO removal (c) of samples.

169.2

164.4 Ce/NAC

169

172

170

168

164.4

166

Mn/NAC

164

Binging energy (eV) Fig. 7. S 2p XPS spectra of samples after SO2 removal.

162

336

N.-J. Fang et al. / Fuel 202 (2017) 328–337

Ce/NAC h

Mn/NAC

f

g

e

642.5

b

Ce 3+ : 33.8%

d 641

a

c

644.5

Ce(1)-Mn/NAC Ce(1)-Mn/NAC

g

e

h

Ce 3+ : 18.3% a

b

c

641

644.4

Ce(7)-Mn/NAC-S

Ce(7)-Mn/NAC-S

644.8 641

Ce 3+ : 36.2%

f

h

642.6

b

e

g

Intensity

Intensity (a.u.)

d

f

642.4

a

d c

Ce(7)-Mn/NAC Ce(7)-Mn/NAC

h 642.5

e

g

641

d

f

a c

644.6

d

h

624.4

e g

641

644.5

655

b

Ce(7)-Mn/NAC-N

Ce(7)-Mn/NAC-N

660

Ce 3+ : 19.3%

650

f

a c

645

640

635

920

910

Binging energy (eV)

900

Ce 3+ : 19.8%

b

890

880

870

Binding Energy (eV)

Fig. 8. Mn 2p (a) and Ce 3d (b) XPS spectra of samples.

in amorphous state. NO is prone to be absorbed on CeO2 and form NO2 and Ce2O3. The latter is easily absorbed O2 to form CeO2. The formed NO2 is then reduced by NH3. NO removal rate of Ce(1)-Mn/ NAC is poor, which is due to the low content of CeO2, because the NO removal efficiency is correlated with the surface active components [42,45]. Ce can change the electronic state of Mn to promote the transfer of electrons and enhance MnOx dispersion on the catalyst surface [46]. This synergistic effect can further improve catalytic ability, leading to a high NO removal efficiency. The Mn 2p and Ce 3d XPS spectra are shown in Fig. 8. The Mn 2p3/2 peaks can be deconvoluted into three peaks at 641.0, 642.4–642.6 and 644.4–644.8 eV. They belong to Mn2+, Mn3+ and Mn4+, respectively [47,48]. It indicates that MnOx with different valence states coexist in the catalyst. For Ce(7)-Mn/NAC, the ratio of Mn4+, Mn3+ and Mn2+ are 12.9, 43.2 and 43.9%, but after desulfurization, the ratio changes to 35.8, 44.1 and 20.1%, respectively. It means the increase of Mn4+ and the decrease of Mn2+. Mn3+ is the major species because Mn3O4 can absorb or lose oxygen to form MnO2 or Mn2O3. SO2 can react with the lattice oxygen atoms (O2
) in Mn3O4 and Mn2O3 to form SO3. When O2
is consumed, the oxygen in the gaseous phase is adsorbed on the surface of Mn3O4 and Mn2O3 to form chemical bond and dissociated to convert lattice oxygen [49]. After denitration, Mn4+ also shows an increase, but the increase amount is smaller compared to the samples after desulfurization. It indicates that Mn oxides play an important role in the catalytic reaction. In Fig. 8b, deconvolution of Ce 3d XPS peaks gives eight peaks, among which the binding energies located at 903.9 (f) and 886.5–886.8 eV (b) attribute to Ce3+ state [23,44,50] and the others are assigned to Ce4+ species [51]. It indicates that Ce is a mixture state of Ce3+ and Ce4+. The

percentages of Ce4+ and Ce3+ are calculated by the peak area. It is obvious that the Ce4+ content in Ce(7)-Mn/NAC show a decrease from 80.7 to 63.8%, accompanying with an increase of Ce3+ from 19.3 to 36.2% after SO2 removal, proving that a portion of Ce4+ has converted into Ce3+ during SO2 removal. This is because Ce3+ can easily react with SO2 and O2 to generate Ce2(SO4)3, which lose the oxygen transfer between Ce3+ and Ce4+, causing the decrease of desulfurization activity. The generated Ce2(SO4)3 is deposited in pore or surface to increase transfer resistance, leading to poor SO2 removal. No obvious difference is observed in Ce(7)-Mn/NAC before and after NO removal. This is because the low strength of Mn3+AO and Mn4+AO bond favors the catalytic oxidation of NO, which will promote the generation and release of the product NO2 [52]. The generated NO2 will react with H from the cleavage of NAH in NH3 to form N2 and H2O. When the temperature rises, CeO2 is more able to transfer electrons to the adsorbed O2, which accelerates the transfer of oxygen, resulting in a good catalytic ability. The reversible adsorption/desorption cycles of lattice oxygen could further promote the oxidation of NO [28]. CeO2 can strengthen MnAO bond between O2 and MnOx to form a manganese – oxide – type phase, resulting in higher metal oxides dispersion. From above analysis, one can know that Mn species are more conducive to SO2 removal, while Ce species take the dominant role during denitration process.

4. Conclusions Ce and Mn bimetal modified activated carbon show great catalytic ability for SO2 and NO removal. Mn oxides are more con-

N.-J. Fang et al. / Fuel 202 (2017) 328–337

ducive to desulfurization, while Ce oxide is more beneficial for NO removal. Mn3O4 and Mn2O3 species are favorable for the oxidation of SO2 and NO because of the lattice oxygen, but CeO2 can improve charge transfer. The micropore facilitates SO2 removal while the mesopore improves NO removal. Oxygen functional groups play an important role both for SO2 and NO removal. C@O and O@CAOH groups are favorable for SO2 oxidation, and CAO can promote the oxidation of SO2 and NO. The hydroxyl groups can promote the adsorption and oxidation of SO2 and NO, but the role of OH is different in SO2 removal and NO removal. Acknowledgements This work is financially supported by the Sichuan Provincial Science and Technology Agency Support Projects (No. 2016GZ0047 and 2014GZ0134) and the Chengdu Science and Technology Bureau Huimin Projects (No. 2014-HM01-00263-SF). References [1] Wang W, Li C, Yan Z. Study on molding semi-coke used for flue-gas desulphurization. Catal Today 2010;158:235–40. [2] Yu W, Wu X, Si Z, Weng D. Influences of impregnation procedure on the SCR activity and alkali resistance of V2O5–WO3/TiO2 catalyst. Appl Surf Sci 2013;283:209–14. [3] Liu XL, Guo JX, Chu YH, Luo DM, Yin HQ, Sun MC, et al. Desulfurization performance of iron supported on activated carbon. Fuel 2014;123:93–100. [4] Wang J, Yan Z, Liu L, Zhang Y, Zhang Z, Wang X. Low-temperature SCR of NO with NH3 over activated semi-coke composite-supported rare earth oxides. Appl Surf Sci 2014;309:1–10. [5] Halina Martyniuk JW. Adsorbents and catalysts from brown coal for flue gas desulfurization. Fuel 1995;74:1716–8. [6] Izquierdo Marı’a Teresa, Rubio B, Mayoral Carmen, Andres Jose’ Manuel. Low cost coal-based carbons for combined SO2 and NO removal from exhaust gas. Fuel 2003;82:147–51. [7] Zhang D, Zhang L, Shi L, Fang C, Li H, Gao R, et al. In situ supported MnOx-CeOx on carbon nanotubes for the low-temperature selective catalytic reduction of NO with NH3. Nanoscale 2013;5:1127–36. [8] Tian W, Yang H, Fan X, Zhang X. Catalytic reduction of NOx with NH3 over different-shaped MnO2 at low temperature. J Hazard Mater 2011;188:105–9. [9] Chu Y, Zhang T, Guo J, Liu C, Yin H, Zhu X, et al. Low temperature selective catalytic reduction of NO by C3H6 over CeOx loaded on AC treated by HNO3. J Rare Earth 2015;33:371–81. [10] Peña DA, Uphade BS, Smirniotis PG. TiO2-supported metal oxide catalysts for low-temperature selective catalytic reduction of NO with NH3 I. Evaluation and characterization of first row transition metals. J Catal 2004;221:421–31. [11] Trovarelli Alessandro, Llorca Jordi, de Leitenburg Carla, Dolcetti Giuliano, Kiss Janos T. Nanophase fluorite-structured CeO2–ZrO2 catalysts prepared by highenergy mechanical milling. J Catal 1997;169:490–502. [12] Laosiripojana N, Assabumrungrat S. The effect of specific surface area on the activity of nano-scale ceria catalysts for methanol decomposition with and without steam at SOFC operating temperatures. Chem Eng Sci 2006;61:2540–9. [13] Shan W, Liu F, He H, Shi X, Zhang C. Novel cerium-tungsten mixed oxide catalyst for the selective catalytic reduction of NOx with NH3. Chem Commun 2011;47:8046–8. [14] Gu T, Liu Y, Weng X, Wang H, Wu Z. The enhanced performance of ceria with surface sulfation for selective catalytic reduction of NO by NH3. Catal Commun 2010;12:310–3. [15] Sheng Z, Hu Y, Xue J, Wang X, Liao W. SO2 poisoning and regeneration of MnCe/TiO2 catalyst for low temperature NOx reduction with NH3. J Rare Earth 2012;30:676–82. [16] Reddy GK, He J, Thiel SW, Pinto NG, Smirniotis PG. Sulfur-tolerant Mn-Ce-Ti sorbents for elemental mercury removal from flue gas: mechanistic investigation by XPS. J Phys Chem C 2015;119:8634–44. [17] Wu Z, Jin R, Wang H, Liu Y. Effect of ceria doping on SO2 resistance of Mn/TiO2 for selective catalytic reduction of NO with NH3 at low temperature. Catal Commun 2009;10:935–9. [18] Chu YH, Wang TZ, Yao Y, Song P, Yin HQ, Guo JX. Desulfurization performance of iron supported on activated carbon. J Sichuan Univ: Nat Sci Ed 2013;45:127–32. [19] Valladares FR, Zgrablich G. Characterization of active carbons: the influence of the method in the determination of the pore size distribution. Carbon 1998;36:1491–9. [20] Tseng HH, Wey MY. Study of SO2 adsorption and thermal regeneration over activated carbon-supported copper oxide catalysts. Carbon 2004;42:2269–78. [21] Román Martínez M C AD, Linares Solano A, Salinas Martínez de Lecea C. Metalsupport interaction in Pt/C samples: Influence of the support surface chemistry and the metal precursor. Carbon 1995; 33: 3–12.

337

[22] Li L, Diao Y, Liu X. Ce-Mn mixed oxides supported on glass-fiber for lowtemperature selective catalytic reduction of NO with NH3. J Rare Earth 2014;32:409–15. [23] Gao X, Liu S, Zhang Y, Luo Z, Cen K. Physicochemical properties of metal-doped activated carbons and relationship with their performance in the removal of SO2 and NO. J Hazard Mater 2011;188:58–66. [24] Ma K, Zou W, Zhang L, Li L, Yu S, Tang C, et al. Construction of hybrid multishell hollow structured CeO2–MnOx materials for selective catalytic reduction of NO with NH3. RSC Adv 2017;7:5989–99. [25] Guo JX, Liang J, Chu YH, Sun MC, Yin HQ, Li JJ. Desulfurization activity of nickel supported on acid-treated activated carbons. Appl CatalA: Gen 2012;421– 422:142–7. [26] Hou ML, Cen WL, Nan F, Li J, Chu Yh, Yin HQ. Dissociation of O2 and its reactivity on O/S doped graphene. RSC Adv 2016;6:7015–21. [27] Qi G, Yang RT, Chang R. MnOx-CeO2 mixed oxides prepared by co-precipitation for selective catalytic reduction of NO with NH3 at low temperatures. Appl Catal B: Environ 2004;51:93–106. [28] Qi G, Yang RT. A superior catalyst for low-temperature NO reduction with NH3. Chem Commun 2003:848–9. [29] Brown TC, Haynes BS. Oxygen chemisorption on carbon. Proc Int Symp Combust 1992;24:1199–206. [30] Klose W, Rincón S. Adsorption and reaction of NO on activated carbon in the presence of oxygen and water vapour. Fuel 2007;86:203–9. [31] Chen H, Hashisho Z. Fast preparation of activated carbon from oil sands coke using microwave-assisted activation. Fuel 2012;95:178–82. [32] Vautard F, Grappe H, Ozcan S. Stability of carbon fiber surface functionality at elevated temperatures and its influence on interfacial adhesion. Appl Surf Sci 2013;268:61–72. [33] Durán Valle CJ, Madrigal Martínez M, Martínez Gallego M, Fonseca IM, Matos I, Botelho do Rego AM. Activated carbon as a catalyst for the synthesis of Nalkylimidazoles and imidazolium ionic liquids. Catal Today 2012;187:108–14. [34] Liu YJ, Qu YF, Guo JX, Wang XJ, Chu YH, Yin HQ, et al. Thermal regeneration of manganese supported on activated carbons treated by HNO3 for Desulfurization. Energ Fuel 2015;29:1931–40. [35] Guo JX, Qu YF, Shu S, Wang XJ, Yin HQ, Chu YH. Effects of preparation conditions on Mn-based activated carbon catalysts for desulfurization. New J Chem 2015;39:5997–6015. [36] Zhang HJ, Cen WL, Liu J, Guo JX, Yin HQ, Ning P. Adsorption and oxidation of SO2 by graphene oxides: a van der Waals density functional theory study. Appl Surf Sci 2015;324:61–7. [37] Cen WL, Hou ML, Liu J, Yuan S, Liu YJ, Chu YH. Oxidation of SO2 and NO by epoxy groups on graphene oxides: the role of the hydroxyl group. RSC Adv 2015;5:22802–10. [38] Hou ML, Cen WL, Zhang HJ, Liu J, Yin HQ, Wei FS. Adsorption and oxidation of NO on graphene oxides: a dispersion corrected density functional theory investigation. Appl Surf Sci 2015;339:55–61. [39] Tang S, Cao Z. Adsorption of nitrogen oxides on graphene and graphene oxides: insights from density functional calculations. J Chem Phys 2011;134:044710. [40] Yamashita T, Hayes P. Analysis of XPS spectra of Fe2+ and Fe3+ ions in oxide materials. Appl Surf Sci 2008;254:2441–9. [41] Qu YF, Guo JX, Chu YH, Sun MC, Yin HQ. The influence of Mn species on the SO2 removal of Mn-based activated carbon catalysts. Appl Surf Sci 2013;282:425–31. [42] Boningari T, Ettireddy PR, Somogyvari A, Liu Y, Vorontsov A, McDonald CA, et al. Influence of elevated surface texture hydrated titania on Ce-doped Mn/ TiO2 catalysts for the low-temperature SCR of NOx under oxygen-rich conditions. J Catal 2015;325:145–55. [43] Sun CQ. Surface and nanosolid core-level shift: impact of atomic coordinationnumber imperfection. Phys Rev B 2004;69. [44] Andreoli S, Deorsola FA, Pirone R. MnOx-CeO2 catalysts synthesized by solution combustion synthesis for the low-temperature NH3-SCR. Catal Today 2015;253:199–206. [45] Liu C, Gao G, Shi JW, He C, Li G, Bai N, et al. MnOx-CeO2 shell-in-shell microspheres for NH3-SCR de-NOx at low temperature. Catal Commun 2016;86:36–40. [46] Zhong Y, Ding LL, Daniel W, Earnest FG. Supercritical water oxidation of NH3 over a MnO2/CeO2 catalyst. Ind Eng Chem Res 1998;37:1707–16. [47] Park E, Kim M, Jung H, Chin S, Jurng J. Effect of sulfur on Mn/Ti catalysts prepared using chemical vapor condensation (CVC) for low-temperature NO reduction. ACS Catal 2013;3:1518–25. [48] Banerjee HD. Interpretation of XPS Mn(2p) spectra of Mn oxyhydroxides and constraints on the mechanism of MnO2 precipitation. Am Mineral 1998;83:305–15. [49] Doornkamp VP. The universal character of the Mars and Van Krevelen mechanism. J Mol Catal A: Chem 2000;162:19–32. [50] Yang S, Zhu W, Jiang Z, Chen Z, Wang J. The surface properties and the activities in catalytic wet air oxidation over CeO2–TiO2 catalysts. Appl Surf Sci 2006;252:8499–505. [51] Baylet A, Capdeillayre C, Retailleau L, Vernoux P, Figueras F, Giroir-Fendler A. Relation between partial propene oxidation, sulphate content and selective catalytic reduction of NOx by propene on ceria/sulphated titania. Appl Catal B: Environ 2010;96:434–40. [52] Wu Z, Tang N, Xiao L, Liu Y, Wang H. MnOx/TiO2 composite nanoxides synthesized by deposition-precipitation method as a superior catalyst for NO oxidation. J Colloid Interf Sci 2010;352:143–8.