- Email: [email protected]
Vanadium silicate (EVS)-supported silver nanoparticles: A novel catalytic sorbent for elemental mercury removal from flue gas
Accepted Manuscript Title: Vanadium silicate (EVS)-supported silver nanoparticles: a novel catalytic sorbent for elemental mercury removal from flue gas Authors: Zijian Zhou, Tiantian Cao, Xiaowei Liu, Shengming Xu, Zhenghe Xu, Minghou Xu PII: DOI: Reference:
S0304-3894(19)30490-X https://doi.org/10.1016/j.jhazmat.2019.04.062 HAZMAT 20579
To appear in:
Journal of Hazardous Materials
Received date: Revised date: Accepted date:
31 January 2019 9 April 2019 19 April 2019
Please cite this article as: Zhou Z, Cao T, Liu X, Xu S, Xu Z, Xu M, Vanadium silicate (EVS)-supported silver nanoparticles: a novel catalytic sorbent for elemental mercury removal from flue gas, Journal of Hazardous Materials (2019), https://doi.org/10.1016/j.jhazmat.2019.04.062 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.
Vanadium silicate (EVS)-supported silver nanoparticles: a novel catalytic sorbent for elemental mercury removal from flue gas
Zijian Zhoua, Tiantian Caob,c, Xiaowei Liua,*, Shengming Xuc, Zhenghe Xuc,d, *,
a
IP T
Minghou Xua
State Key Laboratory of Coal Combustion, Huazhong University of Science and
SC R
Technology, Wuhan 430074, China.
SINOPEC Research Institute of Petroleum Processing, Beijing 100083, China
c
Institute of Nuclear and New Energy Technology, Tsinghua University, Beijing
U
b
Department of Materials Science and Engineering, Southern University of Science
A
d
N
100084, China
M
and Technology, Shenzhen 518055, China
ED
*Corresponding Author
Prof. Zhenghe Xu, E-mail: [email protected]; Tel: +86 75588018968.
CC E
PT
Prof. Xiaowei Liu, E-mail: [email protected]; Tel: +86 2787546631
A
Graphical abstract
1
IP T SC R U
Highlights
Ag modified EVS was synthesized and used as catalytic sorbent for Hg0 capture.
The doped Ag nanoparticles can significantly improve Hg0 oxidation efficiency.
The interaction between Ag and V was responsible for the low temperature
M
A
N
activity.
The role of silver and vanadium on the Hg0 oxidation process was discussed.
PT
ABSTRACT
ED
CC E
Vanadium silicate (EVS) is a vanadium-substituted form of titanosilicate that has a high potential for use as a sorbent for mercury removal. In the present study, EVS with supported silver nanoparticles (EVS-Ag100) as the catalytic sorbent was
A
synthesized for elemental mercury (Hg0) capture. The physical and chemical properties of the sorbents were investigated. The raw EVS exhibited a poor Hg0 capture capacity (7.7 μg·g-1), because most of the vanadium species in the structure of EVS were V4+. The loading of the silver could significantly enhance the Hg0 capture 2
capacity (63.4 μg·g-1). EVS-Ag100 exhibited a superior Hg0 capture performance at temperatures of approximately 150 °C. Silver nanoparticles that formed on the EVS were the active sites. In addition, the vanadium species of EVS-Ag100 exhibited higher Hg0 oxidation activity than those in the framework of raw EVS. The XPS results revealed the activation of the vanadium species by the silver nanoparticles.
IP T
After the capture of Hg0 in the presence of O2, more V5+ was observed on the surface
of EVS-Ag100. Exposure of EVS-Ag100 to a continuous simulated flue gas at 150 °C
SC R
with a gas hourly space velocity of 220,000 h-1 led to Hg0 removal efficiency of >96% in a 1-hour test.
U
KEYWORDS: Catalytic sorbent, Elemental mercury removal, Vanadium silicate,
A
N
Silver nanoparticles, Coal-fired flue gas
M
1. Introduction
Mercury (Hg) and its compounds are regarded as pollutants that are hazardous to
ED
both the atmosphere and human health. Hg emissions from anthropogenic sources, such as the coal combustion process, have disturbed the biogeochemical cycle of
PT
mercury [1, 2]. Mercury mainly exists in three forms in coal combustion flue gas:
CC E
elemental mercury (Hg0), oxidized mercury (Hg2+) and particle-bound mercury (HgP) [3, 4]. Water soluble Hg2+ and solid HgP are easily removed from the flue gas by using wet flue gas desulfurization and dust removal devices [5-7]. However, the highly
A
volatile and water-insoluble Hg0 is challenging to capture and can escape into the atmosphere. A variety of sorbents have been proposed for efficient and cost-effective Hg0 capture from coal-fired power plants. Activated carbon (AC), chemically modified AC, 3
fly ash, transition metal oxides and noble metals, etc., have been used for mercury removal in labs and places of large-scale research [8-10]. AC has been widely investigated for years, but it has a limited Hg0 capture capacity [11, 12]. Bromine and sulfur modified AC (Br- and S-AC) have been reported as promising candidates [12, 13], however, bromine or sulfur species could lead to secondary pollution and
IP T
corrosion. Hence, utilizing Br- and S-AC is not very promising as a long term solution for mercury emission control in coal-fired power plants.
SC R
In recent years, much attention has been paid to transition metal oxides and noble metals based catalytic sorbents [14-17]. The transition metal oxides, such as MnOx
U
[18, 19], V2O5 [20], FeOx [21] and CuOx [22, 23], exhibit excellent Hg0 adsorption or
N
catalytic oxidation due to their extraordinary properties, including multiple valence
A
states and high chemical stability [24]. Vanadium oxide (V2O5), which is known as a selective catalytic reduction (SCR) catalyst for NOx removal, has been widely
M
investigated for Hg0 removal [25-27]. V2O5 is always loaded on the porous materials
ED
with large surface areas, such as TiO2 and SiO2 [26, 27], to obtain higher catalytic activity. However, it is difficult to disperse V2O5 uniformly on the inner surface of the
PT
support, especially into the micropores. Therefore, we proposed a vanadium-containing porous molecular sieve as a reliable substitute for the supported
CC E
V2O5. Vanadium silicate (EVS) is a derivative of a titanosilicate with pores and channels at molecular dimensions, in which the titanium was substituted by vanadium
A
[28, 29]. EVS is also a micro-porous molecular sieve with V4+/V5+ redox couple in its frameworks. Moreover, EVS has been used as a sorbent for some other gases [28, 29]. Therefore, considering its structure, composition and application, EVS seems to be a promising candidate to be used for Hg0 capture.
4
As reported, the V4+/V5+ redox couples in V2O5-based material are responsible for Hg0 adsorption and oxidation [25], but the high binding energy of V-O limited the low temperature Hg0 removal activity. Therefore, V2O5 exhibits better activity at >300 °C [30, 31] and is always being placed before the electrostatic precipitators. However, the alkali and alkaline earth metals in the fly ash could lead to deactivation
IP T
of V2O5, and efforts should be made to improve its low temperature activity. Silver (Ag) was found to enhance the Hg0 removal activity of transition metal oxides,
SC R
especially at low temperatures [16, 32]. The interaction of Ag and the metal oxides
could weaken the O-M (metal ions) bond strength, thereby decreasing the activation energy for the desorption of surface active oxygen species and promoting their
N
U
activity during Hg0 removal [32]. In addition, Ag loading could keep the transition
A
metal elements in their high oxidation states [16]. In this way, the reaction
at lower temperatures.
M
temperature window may become wider and the activity may be remarkably enhanced
ED
It is easy to load silver onto EVS [29]. The exchange of the alkali cations in EVS is a simple process, because EVS displays a strong preference for silver ions [17, 29].
PT
The exchange of silver for K+ and Na+ will obtain a material with powerful adsorption characteristics. In this way, the interaction between V and Ag will become stronger.
CC E
Furthermore, Ag has also been recognized as an effective sorbent for Hg0 capture with amalgamation, due to its ability to generate electrophilic oxygen, which facilitates the
A
redox process [17, 33]. Predictably, the Ag modified EVS will be a catalytic sorbent, which can adsorb and catalyse the Hg0 oxidation due to the existence of Ag and V4+/V5+. To verify the hypothesis, we synthesized EVS and silver modified EVS nanocomposites and we applied it to mercury removal for the first time, which is the first time this has been performed, to the best of our knowledge. The physical and 5
chemical properties of the manufactured material were characterized by various methods. We tested the Hg0 removal activity under different conditions, and the capture process was also discussed.
2. Materials and Methods 2.1. Material preparation
IP T
EVS and Ag modified EVS were hydrothermally synthesized [29]. In brief,
sodium silicate was dissolved in deionized water. Then, NaOH, KCl, NaF and NaCl
SC R
were added to the solution, and the obtained solution was named A. Next, VOSO4 was dissolved in deionized water to get a solution which was named B. The solutions were
U
mixed and stirred, and aged at room temperature. The aged mixture was transferred to
N
an autoclave for ageing at 200 °C. The synthesized product was washed with
A
deionized water and dried.
M
The cation exchange capacity (CEC) was calculated according to reference [29], and the calculated CEC was 5 meq/g. Fifty and One hundred percent silver exchange
ED
EVS were synthesized, and denoted as EVS-Ag50 and EVS-Ag100, meaning silver occupied 50% and 100% of the total CEC of the raw as-synthesized EVS, respectively.
PT
To make EVS-Ag50, we added 10 g of EVS to the AgNO3 solution (4.25 g of AgNO3
CC E
and 100 g of deionized water). In the dark, the suspension was stirred at room temperature for 28 hours. The resulting sample was then washed with deionized water and dried at room temperature, followed by thermal reduction at 350 °C under inert
A
gas flow.
2.2. Material characterizations X-ray fluorescence (XRF, EAGLE III, EDAX Inc.) spectrometry was used to determine the chemical composition of the samples. The morphologies of the samples were examined by a transmission electron microscope (TEM, JEOL, JEM 2100) 6
operated at a 200 kV accelerating voltage with an energy dispersive X-ray spectrometer (EDX). X-ray diffraction (XRD, Rigaku, D/max-2500/PC) with Cu-Kα as a radiation source was used to characterize the phases of the sorbents. N2 adsorption-desorption isotherms measured at -196 °C by an automatic gas sorption analyser (Quantachrome, Autosorb-IQ2) were also conducted. X-ray photoelectron spectra (XPS) were conducted by ESCALAB 250Xi (Thermo Fisher) to detect the chemical states of the elements. The binding energy was calibrated according to C 1s
IP T
band at 284.6 eV. Hg temperature-programmed desorption (Hg-TPD) was performed to obtain the information of mercury species on the sorbents. The spent EVS or
SC R
EVS-Ag100 used under N2+6% O2 atmosphere was put in the fixed-bed reactor and heated from room temperature to 600 °C with a heating rate of 15 °C·min-1 (the
results showed that all the mercury released when the temperature reached ~400°C). Pure nitrogen (1 L·min-1) was used as the carrier gas to transport the desorbed
U
mercury, and the mercury concentration was recorded by the RA-915M mercury
N
analyzer.
A
2.3. Activity tests
M
An Hg0 breakthrough test was conducted with a fixed-bed reactor and the inner diameter of the quartz reactor was 5 mm. A total of 50 mg of sorbent was used for
ED
each test. The Hg0 concentration was detected by an Hg0 analyser (RA-915M), and N2 flow was used as the carrier gas at 50 ml·min-1. Then, 200 μL of Hg0-saturated air was
PT
injected into the N2 flow. The Hg0 that broke through the sorbent will be captured by
CC E
the gold trap. The trap was heated, and the captured Hg0 was released and detected by RA-915M. The Hg0 breakthrough was determined by the ratios of the amount of Hg0 being released to the amount of Hg0 that was injected.
A
The mercury capture capacity was also tested. A mercury permeation device was
used to produce constant Hg0 in the flue gas and the carrier gas was N2 (flow rate: 50 mL·min-1). The Hg0 concentration was 520 μg·m-3. A total of 50 mg of sorbent was placed in a quartz reactor tube with an inner diameter of 5 mm. In the present study, 7
the mercury capture capacity represented the mass of Hg0 captured by the unit mass of the sorbent once the Hg0 breakthrough reached 1%, which is much stricter than other reported sorbents [34, 35]. The Hg0 removal activity in the simulated coal-fired flue gas was also tested. Three sets of experiments were conducted. The set I experiment aimed at determining
IP T
the Hg0 removal activity under the N2 and N2+6% O2 atmosphere at 150 °C, and the
test lasted 5 hours. A total of 200 mg of sorbent was used, and the flow rate was set at
SC R
300 ml·min-1, the equivalent to a gas hourly space velocity (GHVS) of approximately 100,000 h-1. The set II experiment was conducted to investigate the Hg0 removal
U
activity from a more practical perspective, under a more complex flue gas
environment and consisting of N2, 6% O2, 10% CO2, 0.05% NO, 0.1% SO2, 0.001%
N
HCl and 80 μg·m3 Hg0 vapor [36]. The set III experiment was carried out to study the
M
A
effect of individual acid gas on Hg0 removal. Tests were conducted for EVS-Ag100 in the presence of individual flue gases at 150 °C and the balance gas was N2 or N2+O2.
ED
The GHVS was approximately 220,000 h-1 for the set II and III experiments. The activity of the sorbent was calculated in terms of Hg breakthrough (X, %),
0
X
PT
Hg removal efficiency (η, %) and Hg adsorption capacity (Q, μg/g) by Eq. (1)-(3): H g R e le a s e 0
100%
CC E
H g In je c t
H g in H g o u t 0
(1)
0
100%
0
(2)
H g out
F H g in 0
A
Q
t
0
dt
(3)
M
where Hg0Release (μg) is the amount of mercury released from the gold trap, Hg0Inject (μg) is the amount of mercury injected into the flue gas, Hg0in (μg·m-3) and Hg0out (μg·m-3) are the Hg0 concentrations at the inlet and outlet of the fixed-bed reactor, F is the gas flow rate (m3·min-1), t is the reaction time (min) and M is the mass of the 8
sorbent (g), respectively.
3. Results and discussion 3.1. Characterizations The composition of different elements on the sorbent surface was determined by XRF, and the results were expressed as their mole ratios to that of vanadium. As
IP T
shown in Table 1, for the raw and Ag modified EVS, the values of Si to V were almost the same; this indicates that V and Si did not participate in the ion exchange. For
SC R
EVS-Ag50 and EVS-Ag100, the values of Ag to V were 0.87 and 1.41, respectively. It was also clear that with the increase of Ag for modified EVS, Na and K both
decreased. In addition, the total cation equivalents in the samples almost remained
U
constant when Ag increased. The observation suggested that silver replaced an equal
N
number of the alkali cations.
A
Table 1. XRF results expressed as molar ratios to vanadium. Si:V Ag:V Na:V K:V Cation equivalents
EVS
3.92 0
1.39
0.48 1.87
EVS-Ag50
3.83 0.87
0.51
0.43 1.81
0.18
0.26 1.85
ED
M
Samples
EVS-Ag100 3.99 1.41
PT
The surface area and porosity data calculated for the sorbents are summarized in Table 2. EVS is a microporous material with ordered structures [28], and the average
CC E
pore sizes of the three sorbents were all less than 1 nm. The BET surface area was approximately 234 m2/g for the original EVS, and decreased with the increase in the amount of Ag added to the EVS samples. For EVS-Ag50 and EVS-Ag100, the areas
A
decreased to 105 m2/g and 41 m2/g, respectively. The pore volume also reduced with the increasing Ag content, which could be due to the formation of Ag clusters or nanoparticles blocking some of the pores [37]. Table 2. BET surface area and pore analysis for the sorbents. 9
Samples EVS EVS-Ag50 EVS-Ag100
Mean pore diameter (nm) 0.98 0.78 0.71
Pore volume (cm3g-1) 0.26 0.22 0.15
BET surface area (m2g-1) 234 105 41
Figure 1 shows the XRD patterns of the EVS, EVS-Ag50, and EVS-Ag100. The
IP T
characteristic peaks for the as-synthesized EVS were almost the same as the reported ones [28, 29]. For EVS-Ag50 and EVS-Ag100, the other four diffraction peaks at
SC R
approximately 38.1°, 44.3°, 64.4° and 77.5° confirm the successful incorporation of
Ag (JCPDS NO. 04-0783) nanoparticles within the EVS channels [29]. The intensity of the four peaks of EVS-Ag100 was much stronger, indicating more silver was
PT
ED
M
A
N
U
exchanged.
CC E
Figure 1. XRD patterns of EVS, EVS-Ag50 and EVS-Ag100. XPS analysis was conducted to determine the chemical states of the elements and
their changes as a result of the silver ion exchange. The Ag 3d spectra are shown in
A
Figure 2(a). No signals of Ag were detected for the raw EVS. Both EVS-Ag50 and EVS-Ag100 had two peaks at 374.4 eV and 368.4 eV, which are attributed to Ag3d5/2 and Ag3d3/2, respectively [29]. The results suggested that the silver was loaded successfully on EVS. As shown in Figure 2(b), the EVS peaks observed at 524.5 eV 10
and 517.3 eV were assigned to V 2p1/2 and V 2p3/2, respectively. After the silver exchange, the peak position of V 2p showed a slight shift to a lower binding energy. The results suggested that the silver exchange probably weakened the bond strength of vanadium-oxygen, thereby reducing the barriers for surface oxygen species desorption [29, 38], which would be significant for Hg0 adsorption and oxidation. The
IP T
XPS Na 1s spectra are shown in Figure 2(c). There was only one peak at ~1072.0 eV, which was attributed to the Na+ [39]. The intensity decreased sharply after the silver
A
CC E
PT
ED
M
A
N
U
SC R
exchange, indicating that the majority of Na+ was exchanged by the Ag+.
11
IP T
SC R
Figure 2. Ag 3d (a), V 2p (b) and Na 1s (c) XPS spectra of the samples.
The TEM images of the EVS and silver exchanged nanocomposites were
performed to obtain the microscopic morphology information. As shown in Figure 3,
U
EVS is a nanomaterial with a diameter less than 300 nm and no dark spots indicating
N
metallic clusters were observed. Figure 3(a) and (b) showed EVS’s well-ordered
A
microstructures with uniform channel sizes. As shown in Figure 3(c) and (d), a large
M
number of spherical Ag nanoparticles (black dots) were visible on the surface of the
ED
EVS-Ag50, and they had an average diameter of 3-4 nm. In addition, there was also a tiny number of silver clusters with an average diameter of ~20 nm, as shown in Figure
PT
3(c), (e) and (f). In the case of EVS-Ag100, the silver nanoparticles remained uniformly distributed, and most of the particles were 3-4 nm in diameter, as shown in
CC E
Figure 3(e) and (f). The results suggested that the Ag nanocomposites were
A
successfully fabricated on the EVS.
12
IP T
Figure 3. TEM images of EVS (a & b), EVS-Ag50 (c & d) and EVS-Ag100 (e & f).
SC R
3.2. Mercury breakthrough test
To imitate the conditions of the sorbent injection, a Hg0 breakthrough test was
U
carried out, and the Hg0 removal performance over the sorbents with a short contact
N
time was investigated. As shown in Figure 4, 35.5% of Hg0 breakthrough was
A
observed for EVS at 50 °C. At the low temperature, the mercury was potentially
M
physically adsorbed on the EVS due to its good pore structure. The Hg0 breakthrough of EVS increased significantly with the temperature increase, and the Hg0
ED
breakthrough reached 69.6% at 150 °C. When the temperature exceeded 150 °C, the Hg0 removal efficiency increased slightly with the rise of reaction temperature. It was
PT
possible that the vanadium species in the EVS frameworks were responsible for the
CC E
Hg0 removal activity [26]. Hg0 removal over the EVS may follow the Mars-Maessen mechanism, in which the lattice oxygens and V5+ were consumed in the Hg0 oxidation process [25]. In previous studies, the vanadium-based material exhibited better Hg0
A
removal ability at a higher temperature (>300 °C) [31]. It should be noted that EVS exhibited better Hg0 removal activity than the removal obtained from a V2O5-based material at a lower temperature, and the reaction temperature window of EVS was
13
enlarged [31]. The results indicated that the vanadium species in the EVS framework had greater Hg0 removal activity at low temperatures. For Ag exchanged EVS, almost none of the Hg0 breakthrough in the temperature range of 50-150 °C was observed from Figure 4, suggesting that both EVS-Ag50 and EVS-Ag100 exhibit excellent Hg0 removal activity under these conditions. The
IP T
superior performance was mainly attributed to the formation of an Ag-Hg amalgam, a more stable and strong interaction than physical adsorption [14, 33]. When the
SC R
temperature increased to 300 °C, the Hg0 breakthrough increased sharply to about 50% for the Ag modified EVS, suggesting that the catalytic sorbent could still capture
U
mercury at this high temperature. This remarkable observation was largely due to the existence of the vanadium species in the framework of EVS, which was able to
CC E
PT
ED
M
A
N
capture Hg0 at a higher temperature.
Figure 4. Mercury breakthrough at different temperatures. The error bar is the standard deviation.
A
3.3. Mercury capture capacity The Hg0 capture capacity of the three catalytic sorbents was tested at 150 °C. As
shown in Figure 5, the Hg0 capture capacity of EVS was approximately 7.7 μg·g-1, which was much higher than that obtained from the raw SBA-15 [17], HZSM-5 [33], 14
graphene [40], etc. The good performance was mainly due to the active vanadium species in the frameworks, and the Hg0 was chemically adsorbed on the surface to form HgO [41]. The Hg0 capture capacity of EVS-Ag50 and EVS-Ag100 increased significantly to 40.3 μg·g-1 and 63.4 μg·g-1, respectively. The superior capacity was due to the silver loading. According to the results of XRF, the mass ratio of Ag in
IP T
EVS-Ag50 and EVS-Ag100 was 1:1.8. However, the increase of the capacity was not
in line with the Ag loading content, indicating that not all of the silver in the EVS was
SC R
active for Hg0 adsorption. From the results of TEM images of EVS-Ag50 and
EVS-Ag100, there were more silver nanoparticles on the EVS-Ag100. The Hg0
U
capture capacity was lower than that of Ag-SBA-15 in our previous study [17], but
N
still much higher than many noble metal sorbents, e.g., silver powder, gold powder,
CC E
PT
ED
M
A
Ag loaded magnetic zeolite composites, Ag loaded Fe-HZSM-5 [33, 35, 42].
Figure 5. Hg0 capture capacity of the samples. The error bar is the standard deviation.
A
3.4. Mercury removal efficiency in the continuous flue gas Hg0 removal activities under N2 and N2+6%O2 atmosphere were tested at 150 °C
with a high space velocity of 100,000 h-1. As shown in Figure 6, EVS exhibited poor activity on the Hg0 removal in the absence of O2. In the presence of O2, the efficiency was increased due to the supplementation of oxygen in the lattice. The results were 15
similar to those obtained for V2O5-based catalysts [41]. For the EVS-Ag50 and EVS-Ag100, the O2 significantly enhanced the mercury removal efficiency for the Ag loaded samples, indicating that O2 played a significant role in the Hg0 removal activity over for all the samples. In Zhao’s study, it was found that O2 was not significant for the Hg0 removal performance over Ag-Mo sorbents, but O2 could
IP T
obviously enhance Hg0 removal activity of an Ag-Mo modified V2O5 based SCR catalyst [15, 16]. The results indicated that some of the Hg0 was oxidized by the
M
A
N
U
SC R
active oxygen and the consumed oxygen was replenished by gaseous O2.
ED
Figure 6. Hg0 removal efficiency over the sorbents under N2 and N2+6% O2 at 150 °C. The error bar is the standard deviation.
PT
3.5. Effect of acid gases
CC E
Individual acid gases balanced in N2 or N2+6% O2 were used to investigate their effects on the Hg0 removal. HCl was the most important species responsible for Hg0 removal and could dramatically enhance the Hg0 removal activity over most of the
A
transition metal oxides in the presence of oxygen [43]. However, in the present study, as seen in Figure 7, 10 ppm HCl had only a slight enhancing effect on the Hg0 removal activity of EVS-Ag100. The enhancing effect of NO on Hg0 removal was observed from EVS-Ag100. In a pure N2 condition, 500 ppm NO in flue gas 16
significantly enhanced the removal performance. When O2 was injected in the flue gas, the Hg0 removal efficiency was further improved. The promotional effect was probably due to the formation of oxidizing nitrites [10], which were generated via NO oxidation by the surface oxygen species. SO2 showed an inhibitory effect on the Hg0 removal activity of EVS-Ag100 under N2+1000 ppm SO2 atmosphere. When O2 was
IP T
injected into the flue gas, the Hg0 removal activity did not decrease over time. The results indicated that O2 could weaken the inhibition effect [43], because O2 could
M
A
N
U
SC R
enhance the Hg0 removal efficiency (as shown in Figure 6).
ED
Figure 7. Effect of individual acid gas components on Hg0 removal efficiency. To evaluate its potential for practical application in coal-fired power plants,
PT
EVS-Ag100 was exposed to continuous flue gas at 150 °C. The flue gas consisted of
CC E
O2, CO2, NO, SO2, HCl and Hg0 with balance gas of N2. The space velocity was about 220,000 h-1, which was extremely higher than the typical space velocity of an SCR unit in coal-fired power plants. As shown in Figure 8, EVS-Ag100 exhibited a Hg0
A
removal efficiency of 96.4%.
17
IP T
SC R
Figure 8. Hg0 removal efficiency from EVS-Ag100 under a simulated flue gas at 150 °C.
3.6. Further discussions on Hg0 removal over EVS-Ag100
U
To further understand the mercury removal process over the sorbent, the XPS
N
analysis was conducted to determine the chemical states of the vanadium and silver
A
species before and after the reaction. As shown in Figure 9(a), the peak positions of
M
Ag 3d5/2 shifted to lower binding energy after being used for Hg0 removal, indicating the formation of an Ag-Hg amalgam. Figure 9(b) shows the XPS results of vanadium
ED
species on the surface of fresh and spent EVS-Ag100. As reported, V5+ was considered to be an active species for Hg0 removal, and it would be reduced to a low
PT
valence state in the Hg0 oxidation process. However, it should be noted in this study
CC E
that after being used for Hg0 capture in the presence of O2, more V5+ was generated. Two possible reasons could account for the phenomenon. One explanation is that the vanadium with low valence states was possibly reoxidized to V5+ rapidly. The other
A
possible reason could be due to the interactions between the vanadium species and the silver species via the reaction path Ag++V4+→Ag+V5+.
18
IP T
SC R
Figure 9. XPS spectra of Ag 3d (a) and V 2p (b) of EVS-Ag100 used for Hg0 adsorption under N2+6% O2.
The Hg-TPD was carried out to analyse the mercury species on the spent EVS and
U
EVS-Ag100 that were consumed under an N2+O2 atmosphere. As shown in Figure 10,
N
two desorption peaks were observed for EVS. The desorption peaks at lower
A
temperatures (<100 °C) indicated the decomposition of weakly adsorptive mercury
M
species on the outer and inner surface of the sorbents [44]. The peak at >230 °C denoted the decomposition of the strongly adsorbed mercury species [44, 45] which
ED
were probably the oxidized mercury binding with the vanadium species. Additional
PT
desorption peaks were observed from the TPD curve of EVS-Ag100. The peaks at 150-200 °C could be attributed to the Ag-Hg amalgam [16], decomposing Hg0 and
CC E
metallic silver metal as the temperature increased. Therefore, for the EVS-Ag100 material, both the vanadium species and the silver species were responsible for Hg0 removal. In addition, it could be noted that more Hg was released at higher
A
temperatures (>230 °C) on EVS-Ag100, indicating that the loading Ag nanoparticles enhanced the activity of the vanadium species by weakening the bond strength of V-O [15].
19
IP T
SC R
Figure 10. Hg-TPD spectra of EVS-Ag100 used for Hg0 adsorption under N2+6% O2. Based on the above discussion, the primary process for Hg0 capture over the
U
EVS-Ag100 at low temperatures is depicted in Figure 11. Hg0 favoured the adsorption
N
reaction on the silver nanoparticles to form a silver amalgam. Additionally, the
A
interaction between the silver nanoparticles and the vanadium species could weaken
M
the V-O bond strength, so that the transformation of the lattice oxygen to chemical adsorption oxygen became easy, and then the Hg oxidation activity could be
A
CC E
PT
ED
enhanced.
Figure 11. Primary reaction process for Hg0 capture over EVS and EVS-Ag100.
20
4. Conclusions In the present study, the Ag modified EVS was prepared and used for Hg0 capture. For the catalytic sorbent, the vanadium was in the framework of EVS and the Ag nanoparticles were dispersed on the EVS surface. The EVS-Ag100 exhibited a superior Hg0 capture performance at temperatures around 150 °C and the Hg0 capture
IP T
capacity was about 63.4 μg·g-1. Silver nanoparticles formed on the EVS surface and
the vanadium species in the framework of EVS were the active sites for Hg0 capture.
SC R
The superior activity was due to the synergy between Ag and Mo. Furthermore, the
bond strength of V–O might be weakened, which is beneficial for the oxidation ability.
U
O2 and NO could enhance the Hg0 removal activity, while SO2 inhibited the
A
N
performance.
M
Acknowledgements
Zijian Zhou and Tiantian Cao contributed equally to this work. This
ED
research is supported by the National Key R&D Program of China (2018YFB0605200), the National Natural Science Foundation of China
PT
(51661125011) and the China Postdoctoral Science Foundation Funded Project
CC E
(2018T110762, 2017M622439). We thank Dr. Xiangkun Cao for the writing modification. We appreciate the support from the Analytical and Testing Center
A
at Huazhong University of Science & Technology and Tsinghua University.
21
References
A
CC E
PT
ED
M
A
N
U
SC R
IP T
[1] D.P. Krabbenhoft, E.M. Sunderland, Global Change and Mercury, Science 341 (2013) 1457-1458. [2] M. Xu, R. Yan, C. Zheng, Y. Qiao, J. Han, C. Sheng, Status of trace element emission in a coal combustion process: a review, Fuel Process. Technol. 85 (2004) 215-237. [3] S.J. Lee, Y.-C. Seo, H.-N. Jang, K.-S. Park, J.-I. Baek, H.-S. An, K.-C. Song, Speciation and mass distribution of mercury in a bituminous coal-fired power plant, Atmos. Environ. 40 (2006) 2215-2224. [4] M. Xu, Y. Qiao, C. Zheng, L. Li, J. Liu, Modeling of homogeneous mercury speciation using detailed chemical kinetics, Combust. Flame 132 (2003) 208-218. [5] D. Pudasainee, S.J. Lee, S.-H. Lee, J.-H. Kim, H.-N. Jang, S.-J. Cho, Y.-C. Seo, Effect of selective catalytic reactor on oxidation and enhanced removal of mercury in coal-fired power plants, Fuel 89 (2010) 804-809. [6] K.S. Park, Y.C. Seo, S.J. Lee, J.H. Lee, Emission and speciation of mercury from various combustion sources, Powder Technol. 180 (2008) 151-156. [7] Z.-J. Zhou, X.-W. Liu, B. Zhao, Z.-G. Chen, H.-Z. Shao, L.-L. Wang, M.-H. Xu, Effects of existing energy saving and air pollution control devices on mercury removal in coal-fired power plants, Fuel Process. Technol. 131 (2015) 99-108. [8] D. Jampaiah, S.J. Ippolito, Y.M. Sabri, J. Tardio, P.R. Selvakannan, A. Nafady, B.M. Reddy, S.K. Bhargava, Ceria-zirconia modified MnOx catalysts for gaseous elemental mercury oxidation and adsorption, Catal. Sci. Technol. 6 (2016) 1792-1803. [9] 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, Int. J. Coal Geol. 90-91 (2012) 4-20. [10] Z. Zhou, X. Liu, Y. Hu, J. Xu, X.E. Cao, Z. Liao, M. Xu, Investigation on synergistic oxidation behavior of NO and Hg0 during the newly designed fast SCR process, Fuel 225 (2018) 134-139. [11] J. Wilcox, E. Sasmaz, A. Kirchofer, S.-S. Lee, Heterogeneous Mercury Reaction Chemistry on Activated Carbon, J. Air Waste Manage. Assoc. 61 (2011) 418-426. [12] E. Sasmaz, A. Kirchofer, A.D. Jew, A. Saha, D. Abram, T.F. Jaramillo, J. Wilcox, Mercury chemistry on brominated activated carbon, Fuel 99 (2012) 188-196. [13] J.A. Korpiel, R.D. Vidic, Effect of Sulfur Impregnation Method on Activated Carbon Uptake of Gas-Phase Mercury, Environ. Sci. Technol. 31 (1997) 2319-2325. [14] D.-H. Lim, S. Aboud, J. Wilcox, Investigation of Adsorption Behavior of Mercury on Au(111) from First Principles, Environ. Sci. Technol. 46 (2012) 7260-7266. [15] S. Zhao, Z. Li, Z. Qu, N. Yan, W. Huang, W. Chen, H. Xu, Co-benefit of Ag and Mo for the catalytic oxidation of elemental mercury, Fuel 158 (2015) 891-897. [16] S. Zhao, H. Xu, J. Mei, Y. Ma, T. Lou, Z. Qu, N. Yan, Ag-Mo modified SCR catalyst for a co-beneficial oxidation of elemental mercury at wide temperature range, 22
A
CC E
PT
ED
M
A
N
U
SC R
IP T
Fuel 200 (2017) 236-243. [17] T. Cao, Z. Li, Y. Xiong, Y. Yang, S. Xu, T. Bisson, R. Gupta, Z. Xu, Silica-Silver Nanocomposites as Regenerable Sorbents for Hg0 Removal from Flue Gases, Environ. Sci. Technol. 51 (2017) 11909-11917. [18] Z. Zhou, X. Liu, Z. Liao, H. Shao, C. Lv, Y. Hu, M. Xu, Manganese doped CeO2-ZrO2 catalyst for elemental mercury oxidation at low temperature, Fuel Process. Technol. 152 (2016) 285-293. [19] Y. Xia, Z. Liao, Y. Zheng, Z. Zhou, Highly Dispersed Mn-Ce Binary Metal Oxides Supported on Carbon Nanofibers for Hg0 Removal from Coal-Fired Flue Gas, Appl. Sci. 8 (2018) 2501. [20] Z. Wang, J. Liu, B. Zhang, Y. Yang, Z. Zhang, S. Miao, Mechanism of Heterogeneous Mercury Oxidation by HBr over V2O5/TiO2 Catalyst, Environ. Sci. Technol. 50 (2016) 5398-5404. [21] Y. Yang, J. Liu, Z. Wang, F. Liu, Heterogeneous reaction kinetics of mercury oxidation by HCl over Fe2O3 surface, Fuel Process. Technol. 159 (2017) 266-271. [22] H. Li, S. Wu, C.-Y. Wu, J. Wang, L. Li, K. Shih, SCR Atmosphere Induced Reduction of Oxidized Mercury over CuO-CeO2/TiO2 Catalyst, Environ. Sci. Technol. 49 (2015) 7373-7379. [23] H. Li, L. Zhu, S. Wu, Y. Liu, K. Shih, Synergy of CuO and CeO2 combination for mercury oxidation under low-temperature selective catalytic reduction atmosphere, Int. J. Coal Geol. 170 (2017) 69-76. [24] A.A. Presto, E.J. Granite, Survey of Catalysts for Oxidation of Mercury in Flue Gas, Environ. Sci. Technol. 40 (2006) 5601-5609. [25] Z. Zhou, X. Liu, Z. Liao, H. Shao, Y. Hu, Y. Xu, M. Xu, A novel low temperature catalyst regenerated from deactivated SCR catalyst for Hg0 oxidation, Chem. Eng. J. 304 (2016) 121-128. [26] H. Li, Y. Li, C.-Y. Wu, J. Zhang, Oxidation and capture of elemental mercury over SiO2-TiO2-V2O5 catalysts in simulated low-rank coal combustion flue gas, Chem. Eng. J. 169 (2011) 186-193. [27] S. Eswaran, H.G. Stenger, Effect of halogens on mercury conversion in SCR catalysts, Fuel Process. Technol. 89 (2008) 1153-1159. [28] F. Mani, J.A. Sawada, S.M. Kuznicki, Comparative adsorption study of EVS-10 and ETS-10, Microporous Mesoporous Mater. 204 (2015) 43-49. [29] F. Mani, J.A. Sawada, S.M. Kuznicki, Spontaneous formation of silver nanoparticles on the vanadium silicate EVS-10, Microporous Mesoporous Mater. 224 (2016) 208-216. [30] Y. Qiu, B. Liu, J. Du, Q. Tang, Z. Liu, R. Liu, C. Tao, The monolithic cordierite supported V2O5-MoO3/TiO2 catalyst for NH3-SCR, Chem. Eng. J. 294 (2016) 264-272. [31] J. Yang, Q. Yang, J. Sun, Q. Liu, D. Zhao, W. Gao, L. Liu, Effects of mercury oxidation on V2O5-WO3/TiO2 catalyst properties in NH3-SCR process, Catal. Commun. 59 (2015) 78-82. 23
A
CC E
PT
ED
M
A
N
U
SC R
IP T
[32] S. Zhao, Y. Ma, Z. Qu, N. Yan, Z. Li, J. Xie, W. Chen, The performance of Ag doped V2O5-TiO2 catalyst on the catalytic oxidation of gaseous elemental mercury, Catal. Sci. Technol. 4 (2014) 4036-4044. [33] T. Cao, Z. Zhou, Q. Chen, Z. Li, S. Xu, J. Wang, M. Xu, T. Bisson, Z. Xu, Magnetically responsive catalytic sorbent for removal of Hg0 and NO, Fuel Process. Technol. 160 (2017) 158-169. [34] Y. Liu, D.J.A. Kelly, H. Yang, C.C.H. Lin, S.M. Kuznicki, Z. Xu, Novel Regenerable Sorbent for Mercury Capture from Flue Gases of Coal-Fired Power Plant, Environ. Sci. Technol. 42 (2008) 6205-6210. [35] J. Dong, Z. Xu, S.M. Kuznicki, Mercury Removal from Flue Gases by Novel Regenerable Magnetic Nanocomposite Sorbents, Environ. Sci. Technol. 43 (2009) 3266-3271. [36] Y. Zheng, A.D. Jensen, C. Windelin, F. Jensen, Review of technologies for mercury removal from flue gas from cement production processes, Prog. Energy Combust. Sci. 38 (2012) 599-629. [37] H. Li, S. Wu, L. Li, J. Wang, W. Ma, K. Shih, CuO-CeO2/TiO2 catalyst for simultaneous NO reduction and Hg0 oxidation at low temperatures, Catal. Sci. Technol. 5 (2015) 5129-5138. [38] X. Ge, H.-l. Zhang, Temperature-Programmed Reduction Studies of V-Ag Catalysts, J. Solid State Chem. 141 (1998) 186-190. [39] C. Heske, R. Fink, E. Umbach, W. Riedl, F. Karg, Na-induced effects on the electronic structure and composition of Cu(In,Ga)Se2 thin-film surfaces, Appl. Phys. Lett. 68 (1996) 3431-3433. [40] Y. Liu, C. Tian, B. Yan, Q. Lu, Y. Xie, J. Chen, R. Gupta, Z. Xu, S.M. Kuznicki, Q. Liu, H. Zeng, Nanocomposites of graphene oxide, Ag nanoparticles, and magnetic ferrite nanoparticles for elemental mercury (Hg0) removal, RSC Adv. 5 (2015) 15634-15640. [41] B. Zhao, X. Liu, Z. Zhou, H. Shao, C. Wang, J. Si, M. Xu, Effect of molybdenum on mercury oxidized by V2O5-MoO3/TiO2 catalysts, Chem. Eng. J. 253 (2014) 508-517. [42] G. Luo, H. Yao, M. Xu, X. Cui, W. Chen, R. Gupta, Z. Xu, Carbon Nanotube-Silver Composite for Mercury Capture and Analysis, Energy Fuels 24 (2010) 419-426. [43] H. Li, C.-Y. Wu, Y. Li, L. Li, Y. Zhao, J. Zhang, Role of flue gas components in mercury oxidation over TiO2 supported MnOx-CeO2 mixed-oxide at low temperature, J. Hazard. Mater. 243 (2012) 117-123. [44] Q. Wan, L. Duan, K. He, J. Li, Removal of gaseous elemental mercury over a CeO2-WO3/TiO2 nanocomposite in simulated coal-fired flue gas, Chem. Eng. J. 170 (2011) 512-517. [45] Q. Wan, L. Duan, J. Li, L. Chen, K. He, J. Hao, Deactivation performance and mechanism of alkali (earth) metals on V2O5-WO3/TiO2 catalyst for oxidation of 24
A
CC E
PT
ED
M
A
N
U
SC R
IP T
gaseous elemental mercury in simulated coal-fired flue gas, Catal. Today 175 (2011) 189-195.
25
Figure captions
Figure 1. XRD patterns of EVS, EVS-Ag50 and EVS-Ag100. Ag 3d (a), V 2p (b) and Na 1s (c) XPS spectra of the samples.
Figure 3.
TEM images of EVS (a & b), EVS-Ag50 (c & d) and EVS-Ag100 (e & f).
Figure 4.
Mercury breakthrough at different temperatures. The error bar is the
IP T
Figure 2.
standard deviation.
Hg capture capacity of the samples. The error bar is the standard
SC R
Figure 5. deviation.
Hg0 removal efficiency over the sorbents under N2 and N2+6% O2 at
U
Figure 6.
150 °C. The error bar is the standard deviation.
Effect of individual acid gas components on Hg0 removal efficiency.
Figure 8.
Hg0 removal efficiency from EVS-Ag100 under a simulated flue gas at
M
A
N
Figure 7.
150 °C.
XPS spectra of Ag 3d (a) and V 2p (b) of EVS-Ag100 used for Hg0
ED
Figure 9.
adsorption under N2+6% O2.
O2.
Primary reaction process for Hg0 capture over EVS and EVS-Ag100.
A
CC E
Figure 11.
Hg-TPD spectra of EVS-Ag100 used for Hg0 adsorption under N2+6%
PT
Figure 10.
26
Table legends
Table 1. XRF results expressed as molar ratios to vanadium.
A
CC E
PT
ED
M
A
N
U
SC R
IP T
Table 2. BET surface area and pore analysis for the sorbents.
27
S0304-3894(19)30490-X https://doi.org/10.1016/j.jhazmat.2019.04.062 HAZMAT 20579
To appear in:
Journal of Hazardous Materials
Received date: Revised date: Accepted date:
31 January 2019 9 April 2019 19 April 2019
Please cite this article as: Zhou Z, Cao T, Liu X, Xu S, Xu Z, Xu M, Vanadium silicate (EVS)-supported silver nanoparticles: a novel catalytic sorbent for elemental mercury removal from flue gas, Journal of Hazardous Materials (2019), https://doi.org/10.1016/j.jhazmat.2019.04.062 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.
Vanadium silicate (EVS)-supported silver nanoparticles: a novel catalytic sorbent for elemental mercury removal from flue gas
Zijian Zhoua, Tiantian Caob,c, Xiaowei Liua,*, Shengming Xuc, Zhenghe Xuc,d, *,
a
IP T
Minghou Xua
State Key Laboratory of Coal Combustion, Huazhong University of Science and
SC R
Technology, Wuhan 430074, China.
SINOPEC Research Institute of Petroleum Processing, Beijing 100083, China
c
Institute of Nuclear and New Energy Technology, Tsinghua University, Beijing
U
b
Department of Materials Science and Engineering, Southern University of Science
A
d
N
100084, China
M
and Technology, Shenzhen 518055, China
ED
*Corresponding Author
Prof. Zhenghe Xu, E-mail: [email protected]; Tel: +86 75588018968.
CC E
PT
Prof. Xiaowei Liu, E-mail: [email protected]; Tel: +86 2787546631
A
Graphical abstract
1
IP T SC R U
Highlights
Ag modified EVS was synthesized and used as catalytic sorbent for Hg0 capture.
The doped Ag nanoparticles can significantly improve Hg0 oxidation efficiency.
The interaction between Ag and V was responsible for the low temperature
M
A
N
activity.
The role of silver and vanadium on the Hg0 oxidation process was discussed.
PT
ABSTRACT
ED
CC E
Vanadium silicate (EVS) is a vanadium-substituted form of titanosilicate that has a high potential for use as a sorbent for mercury removal. In the present study, EVS with supported silver nanoparticles (EVS-Ag100) as the catalytic sorbent was
A
synthesized for elemental mercury (Hg0) capture. The physical and chemical properties of the sorbents were investigated. The raw EVS exhibited a poor Hg0 capture capacity (7.7 μg·g-1), because most of the vanadium species in the structure of EVS were V4+. The loading of the silver could significantly enhance the Hg0 capture 2
capacity (63.4 μg·g-1). EVS-Ag100 exhibited a superior Hg0 capture performance at temperatures of approximately 150 °C. Silver nanoparticles that formed on the EVS were the active sites. In addition, the vanadium species of EVS-Ag100 exhibited higher Hg0 oxidation activity than those in the framework of raw EVS. The XPS results revealed the activation of the vanadium species by the silver nanoparticles.
IP T
After the capture of Hg0 in the presence of O2, more V5+ was observed on the surface
of EVS-Ag100. Exposure of EVS-Ag100 to a continuous simulated flue gas at 150 °C
SC R
with a gas hourly space velocity of 220,000 h-1 led to Hg0 removal efficiency of >96% in a 1-hour test.
U
KEYWORDS: Catalytic sorbent, Elemental mercury removal, Vanadium silicate,
A
N
Silver nanoparticles, Coal-fired flue gas
M
1. Introduction
Mercury (Hg) and its compounds are regarded as pollutants that are hazardous to
ED
both the atmosphere and human health. Hg emissions from anthropogenic sources, such as the coal combustion process, have disturbed the biogeochemical cycle of
PT
mercury [1, 2]. Mercury mainly exists in three forms in coal combustion flue gas:
CC E
elemental mercury (Hg0), oxidized mercury (Hg2+) and particle-bound mercury (HgP) [3, 4]. Water soluble Hg2+ and solid HgP are easily removed from the flue gas by using wet flue gas desulfurization and dust removal devices [5-7]. However, the highly
A
volatile and water-insoluble Hg0 is challenging to capture and can escape into the atmosphere. A variety of sorbents have been proposed for efficient and cost-effective Hg0 capture from coal-fired power plants. Activated carbon (AC), chemically modified AC, 3
fly ash, transition metal oxides and noble metals, etc., have been used for mercury removal in labs and places of large-scale research [8-10]. AC has been widely investigated for years, but it has a limited Hg0 capture capacity [11, 12]. Bromine and sulfur modified AC (Br- and S-AC) have been reported as promising candidates [12, 13], however, bromine or sulfur species could lead to secondary pollution and
IP T
corrosion. Hence, utilizing Br- and S-AC is not very promising as a long term solution for mercury emission control in coal-fired power plants.
SC R
In recent years, much attention has been paid to transition metal oxides and noble metals based catalytic sorbents [14-17]. The transition metal oxides, such as MnOx
U
[18, 19], V2O5 [20], FeOx [21] and CuOx [22, 23], exhibit excellent Hg0 adsorption or
N
catalytic oxidation due to their extraordinary properties, including multiple valence
A
states and high chemical stability [24]. Vanadium oxide (V2O5), which is known as a selective catalytic reduction (SCR) catalyst for NOx removal, has been widely
M
investigated for Hg0 removal [25-27]. V2O5 is always loaded on the porous materials
ED
with large surface areas, such as TiO2 and SiO2 [26, 27], to obtain higher catalytic activity. However, it is difficult to disperse V2O5 uniformly on the inner surface of the
PT
support, especially into the micropores. Therefore, we proposed a vanadium-containing porous molecular sieve as a reliable substitute for the supported
CC E
V2O5. Vanadium silicate (EVS) is a derivative of a titanosilicate with pores and channels at molecular dimensions, in which the titanium was substituted by vanadium
A
[28, 29]. EVS is also a micro-porous molecular sieve with V4+/V5+ redox couple in its frameworks. Moreover, EVS has been used as a sorbent for some other gases [28, 29]. Therefore, considering its structure, composition and application, EVS seems to be a promising candidate to be used for Hg0 capture.
4
As reported, the V4+/V5+ redox couples in V2O5-based material are responsible for Hg0 adsorption and oxidation [25], but the high binding energy of V-O limited the low temperature Hg0 removal activity. Therefore, V2O5 exhibits better activity at >300 °C [30, 31] and is always being placed before the electrostatic precipitators. However, the alkali and alkaline earth metals in the fly ash could lead to deactivation
IP T
of V2O5, and efforts should be made to improve its low temperature activity. Silver (Ag) was found to enhance the Hg0 removal activity of transition metal oxides,
SC R
especially at low temperatures [16, 32]. The interaction of Ag and the metal oxides
could weaken the O-M (metal ions) bond strength, thereby decreasing the activation energy for the desorption of surface active oxygen species and promoting their
N
U
activity during Hg0 removal [32]. In addition, Ag loading could keep the transition
A
metal elements in their high oxidation states [16]. In this way, the reaction
at lower temperatures.
M
temperature window may become wider and the activity may be remarkably enhanced
ED
It is easy to load silver onto EVS [29]. The exchange of the alkali cations in EVS is a simple process, because EVS displays a strong preference for silver ions [17, 29].
PT
The exchange of silver for K+ and Na+ will obtain a material with powerful adsorption characteristics. In this way, the interaction between V and Ag will become stronger.
CC E
Furthermore, Ag has also been recognized as an effective sorbent for Hg0 capture with amalgamation, due to its ability to generate electrophilic oxygen, which facilitates the
A
redox process [17, 33]. Predictably, the Ag modified EVS will be a catalytic sorbent, which can adsorb and catalyse the Hg0 oxidation due to the existence of Ag and V4+/V5+. To verify the hypothesis, we synthesized EVS and silver modified EVS nanocomposites and we applied it to mercury removal for the first time, which is the first time this has been performed, to the best of our knowledge. The physical and 5
chemical properties of the manufactured material were characterized by various methods. We tested the Hg0 removal activity under different conditions, and the capture process was also discussed.
2. Materials and Methods 2.1. Material preparation
IP T
EVS and Ag modified EVS were hydrothermally synthesized [29]. In brief,
sodium silicate was dissolved in deionized water. Then, NaOH, KCl, NaF and NaCl
SC R
were added to the solution, and the obtained solution was named A. Next, VOSO4 was dissolved in deionized water to get a solution which was named B. The solutions were
U
mixed and stirred, and aged at room temperature. The aged mixture was transferred to
N
an autoclave for ageing at 200 °C. The synthesized product was washed with
A
deionized water and dried.
M
The cation exchange capacity (CEC) was calculated according to reference [29], and the calculated CEC was 5 meq/g. Fifty and One hundred percent silver exchange
ED
EVS were synthesized, and denoted as EVS-Ag50 and EVS-Ag100, meaning silver occupied 50% and 100% of the total CEC of the raw as-synthesized EVS, respectively.
PT
To make EVS-Ag50, we added 10 g of EVS to the AgNO3 solution (4.25 g of AgNO3
CC E
and 100 g of deionized water). In the dark, the suspension was stirred at room temperature for 28 hours. The resulting sample was then washed with deionized water and dried at room temperature, followed by thermal reduction at 350 °C under inert
A
gas flow.
2.2. Material characterizations X-ray fluorescence (XRF, EAGLE III, EDAX Inc.) spectrometry was used to determine the chemical composition of the samples. The morphologies of the samples were examined by a transmission electron microscope (TEM, JEOL, JEM 2100) 6
operated at a 200 kV accelerating voltage with an energy dispersive X-ray spectrometer (EDX). X-ray diffraction (XRD, Rigaku, D/max-2500/PC) with Cu-Kα as a radiation source was used to characterize the phases of the sorbents. N2 adsorption-desorption isotherms measured at -196 °C by an automatic gas sorption analyser (Quantachrome, Autosorb-IQ2) were also conducted. X-ray photoelectron spectra (XPS) were conducted by ESCALAB 250Xi (Thermo Fisher) to detect the chemical states of the elements. The binding energy was calibrated according to C 1s
IP T
band at 284.6 eV. Hg temperature-programmed desorption (Hg-TPD) was performed to obtain the information of mercury species on the sorbents. The spent EVS or
SC R
EVS-Ag100 used under N2+6% O2 atmosphere was put in the fixed-bed reactor and heated from room temperature to 600 °C with a heating rate of 15 °C·min-1 (the
results showed that all the mercury released when the temperature reached ~400°C). Pure nitrogen (1 L·min-1) was used as the carrier gas to transport the desorbed
U
mercury, and the mercury concentration was recorded by the RA-915M mercury
N
analyzer.
A
2.3. Activity tests
M
An Hg0 breakthrough test was conducted with a fixed-bed reactor and the inner diameter of the quartz reactor was 5 mm. A total of 50 mg of sorbent was used for
ED
each test. The Hg0 concentration was detected by an Hg0 analyser (RA-915M), and N2 flow was used as the carrier gas at 50 ml·min-1. Then, 200 μL of Hg0-saturated air was
PT
injected into the N2 flow. The Hg0 that broke through the sorbent will be captured by
CC E
the gold trap. The trap was heated, and the captured Hg0 was released and detected by RA-915M. The Hg0 breakthrough was determined by the ratios of the amount of Hg0 being released to the amount of Hg0 that was injected.
A
The mercury capture capacity was also tested. A mercury permeation device was
used to produce constant Hg0 in the flue gas and the carrier gas was N2 (flow rate: 50 mL·min-1). The Hg0 concentration was 520 μg·m-3. A total of 50 mg of sorbent was placed in a quartz reactor tube with an inner diameter of 5 mm. In the present study, 7
the mercury capture capacity represented the mass of Hg0 captured by the unit mass of the sorbent once the Hg0 breakthrough reached 1%, which is much stricter than other reported sorbents [34, 35]. The Hg0 removal activity in the simulated coal-fired flue gas was also tested. Three sets of experiments were conducted. The set I experiment aimed at determining
IP T
the Hg0 removal activity under the N2 and N2+6% O2 atmosphere at 150 °C, and the
test lasted 5 hours. A total of 200 mg of sorbent was used, and the flow rate was set at
SC R
300 ml·min-1, the equivalent to a gas hourly space velocity (GHVS) of approximately 100,000 h-1. The set II experiment was conducted to investigate the Hg0 removal
U
activity from a more practical perspective, under a more complex flue gas
environment and consisting of N2, 6% O2, 10% CO2, 0.05% NO, 0.1% SO2, 0.001%
N
HCl and 80 μg·m3 Hg0 vapor [36]. The set III experiment was carried out to study the
M
A
effect of individual acid gas on Hg0 removal. Tests were conducted for EVS-Ag100 in the presence of individual flue gases at 150 °C and the balance gas was N2 or N2+O2.
ED
The GHVS was approximately 220,000 h-1 for the set II and III experiments. The activity of the sorbent was calculated in terms of Hg breakthrough (X, %),
0
X
PT
Hg removal efficiency (η, %) and Hg adsorption capacity (Q, μg/g) by Eq. (1)-(3): H g R e le a s e 0
100%
CC E
H g In je c t
H g in H g o u t 0
(1)
0
100%
0
(2)
H g out
F H g in 0
A
Q
t
0
dt
(3)
M
where Hg0Release (μg) is the amount of mercury released from the gold trap, Hg0Inject (μg) is the amount of mercury injected into the flue gas, Hg0in (μg·m-3) and Hg0out (μg·m-3) are the Hg0 concentrations at the inlet and outlet of the fixed-bed reactor, F is the gas flow rate (m3·min-1), t is the reaction time (min) and M is the mass of the 8
sorbent (g), respectively.
3. Results and discussion 3.1. Characterizations The composition of different elements on the sorbent surface was determined by XRF, and the results were expressed as their mole ratios to that of vanadium. As
IP T
shown in Table 1, for the raw and Ag modified EVS, the values of Si to V were almost the same; this indicates that V and Si did not participate in the ion exchange. For
SC R
EVS-Ag50 and EVS-Ag100, the values of Ag to V were 0.87 and 1.41, respectively. It was also clear that with the increase of Ag for modified EVS, Na and K both
decreased. In addition, the total cation equivalents in the samples almost remained
U
constant when Ag increased. The observation suggested that silver replaced an equal
N
number of the alkali cations.
A
Table 1. XRF results expressed as molar ratios to vanadium. Si:V Ag:V Na:V K:V Cation equivalents
EVS
3.92 0
1.39
0.48 1.87
EVS-Ag50
3.83 0.87
0.51
0.43 1.81
0.18
0.26 1.85
ED
M
Samples
EVS-Ag100 3.99 1.41
PT
The surface area and porosity data calculated for the sorbents are summarized in Table 2. EVS is a microporous material with ordered structures [28], and the average
CC E
pore sizes of the three sorbents were all less than 1 nm. The BET surface area was approximately 234 m2/g for the original EVS, and decreased with the increase in the amount of Ag added to the EVS samples. For EVS-Ag50 and EVS-Ag100, the areas
A
decreased to 105 m2/g and 41 m2/g, respectively. The pore volume also reduced with the increasing Ag content, which could be due to the formation of Ag clusters or nanoparticles blocking some of the pores [37]. Table 2. BET surface area and pore analysis for the sorbents. 9
Samples EVS EVS-Ag50 EVS-Ag100
Mean pore diameter (nm) 0.98 0.78 0.71
Pore volume (cm3g-1) 0.26 0.22 0.15
BET surface area (m2g-1) 234 105 41
Figure 1 shows the XRD patterns of the EVS, EVS-Ag50, and EVS-Ag100. The
IP T
characteristic peaks for the as-synthesized EVS were almost the same as the reported ones [28, 29]. For EVS-Ag50 and EVS-Ag100, the other four diffraction peaks at
SC R
approximately 38.1°, 44.3°, 64.4° and 77.5° confirm the successful incorporation of
Ag (JCPDS NO. 04-0783) nanoparticles within the EVS channels [29]. The intensity of the four peaks of EVS-Ag100 was much stronger, indicating more silver was
PT
ED
M
A
N
U
exchanged.
CC E
Figure 1. XRD patterns of EVS, EVS-Ag50 and EVS-Ag100. XPS analysis was conducted to determine the chemical states of the elements and
their changes as a result of the silver ion exchange. The Ag 3d spectra are shown in
A
Figure 2(a). No signals of Ag were detected for the raw EVS. Both EVS-Ag50 and EVS-Ag100 had two peaks at 374.4 eV and 368.4 eV, which are attributed to Ag3d5/2 and Ag3d3/2, respectively [29]. The results suggested that the silver was loaded successfully on EVS. As shown in Figure 2(b), the EVS peaks observed at 524.5 eV 10
and 517.3 eV were assigned to V 2p1/2 and V 2p3/2, respectively. After the silver exchange, the peak position of V 2p showed a slight shift to a lower binding energy. The results suggested that the silver exchange probably weakened the bond strength of vanadium-oxygen, thereby reducing the barriers for surface oxygen species desorption [29, 38], which would be significant for Hg0 adsorption and oxidation. The
IP T
XPS Na 1s spectra are shown in Figure 2(c). There was only one peak at ~1072.0 eV, which was attributed to the Na+ [39]. The intensity decreased sharply after the silver
A
CC E
PT
ED
M
A
N
U
SC R
exchange, indicating that the majority of Na+ was exchanged by the Ag+.
11
IP T
SC R
Figure 2. Ag 3d (a), V 2p (b) and Na 1s (c) XPS spectra of the samples.
The TEM images of the EVS and silver exchanged nanocomposites were
performed to obtain the microscopic morphology information. As shown in Figure 3,
U
EVS is a nanomaterial with a diameter less than 300 nm and no dark spots indicating
N
metallic clusters were observed. Figure 3(a) and (b) showed EVS’s well-ordered
A
microstructures with uniform channel sizes. As shown in Figure 3(c) and (d), a large
M
number of spherical Ag nanoparticles (black dots) were visible on the surface of the
ED
EVS-Ag50, and they had an average diameter of 3-4 nm. In addition, there was also a tiny number of silver clusters with an average diameter of ~20 nm, as shown in Figure
PT
3(c), (e) and (f). In the case of EVS-Ag100, the silver nanoparticles remained uniformly distributed, and most of the particles were 3-4 nm in diameter, as shown in
CC E
Figure 3(e) and (f). The results suggested that the Ag nanocomposites were
A
successfully fabricated on the EVS.
12
IP T
Figure 3. TEM images of EVS (a & b), EVS-Ag50 (c & d) and EVS-Ag100 (e & f).
SC R
3.2. Mercury breakthrough test
To imitate the conditions of the sorbent injection, a Hg0 breakthrough test was
U
carried out, and the Hg0 removal performance over the sorbents with a short contact
N
time was investigated. As shown in Figure 4, 35.5% of Hg0 breakthrough was
A
observed for EVS at 50 °C. At the low temperature, the mercury was potentially
M
physically adsorbed on the EVS due to its good pore structure. The Hg0 breakthrough of EVS increased significantly with the temperature increase, and the Hg0
ED
breakthrough reached 69.6% at 150 °C. When the temperature exceeded 150 °C, the Hg0 removal efficiency increased slightly with the rise of reaction temperature. It was
PT
possible that the vanadium species in the EVS frameworks were responsible for the
CC E
Hg0 removal activity [26]. Hg0 removal over the EVS may follow the Mars-Maessen mechanism, in which the lattice oxygens and V5+ were consumed in the Hg0 oxidation process [25]. In previous studies, the vanadium-based material exhibited better Hg0
A
removal ability at a higher temperature (>300 °C) [31]. It should be noted that EVS exhibited better Hg0 removal activity than the removal obtained from a V2O5-based material at a lower temperature, and the reaction temperature window of EVS was
13
enlarged [31]. The results indicated that the vanadium species in the EVS framework had greater Hg0 removal activity at low temperatures. For Ag exchanged EVS, almost none of the Hg0 breakthrough in the temperature range of 50-150 °C was observed from Figure 4, suggesting that both EVS-Ag50 and EVS-Ag100 exhibit excellent Hg0 removal activity under these conditions. The
IP T
superior performance was mainly attributed to the formation of an Ag-Hg amalgam, a more stable and strong interaction than physical adsorption [14, 33]. When the
SC R
temperature increased to 300 °C, the Hg0 breakthrough increased sharply to about 50% for the Ag modified EVS, suggesting that the catalytic sorbent could still capture
U
mercury at this high temperature. This remarkable observation was largely due to the existence of the vanadium species in the framework of EVS, which was able to
CC E
PT
ED
M
A
N
capture Hg0 at a higher temperature.
Figure 4. Mercury breakthrough at different temperatures. The error bar is the standard deviation.
A
3.3. Mercury capture capacity The Hg0 capture capacity of the three catalytic sorbents was tested at 150 °C. As
shown in Figure 5, the Hg0 capture capacity of EVS was approximately 7.7 μg·g-1, which was much higher than that obtained from the raw SBA-15 [17], HZSM-5 [33], 14
graphene [40], etc. The good performance was mainly due to the active vanadium species in the frameworks, and the Hg0 was chemically adsorbed on the surface to form HgO [41]. The Hg0 capture capacity of EVS-Ag50 and EVS-Ag100 increased significantly to 40.3 μg·g-1 and 63.4 μg·g-1, respectively. The superior capacity was due to the silver loading. According to the results of XRF, the mass ratio of Ag in
IP T
EVS-Ag50 and EVS-Ag100 was 1:1.8. However, the increase of the capacity was not
in line with the Ag loading content, indicating that not all of the silver in the EVS was
SC R
active for Hg0 adsorption. From the results of TEM images of EVS-Ag50 and
EVS-Ag100, there were more silver nanoparticles on the EVS-Ag100. The Hg0
U
capture capacity was lower than that of Ag-SBA-15 in our previous study [17], but
N
still much higher than many noble metal sorbents, e.g., silver powder, gold powder,
CC E
PT
ED
M
A
Ag loaded magnetic zeolite composites, Ag loaded Fe-HZSM-5 [33, 35, 42].
Figure 5. Hg0 capture capacity of the samples. The error bar is the standard deviation.
A
3.4. Mercury removal efficiency in the continuous flue gas Hg0 removal activities under N2 and N2+6%O2 atmosphere were tested at 150 °C
with a high space velocity of 100,000 h-1. As shown in Figure 6, EVS exhibited poor activity on the Hg0 removal in the absence of O2. In the presence of O2, the efficiency was increased due to the supplementation of oxygen in the lattice. The results were 15
similar to those obtained for V2O5-based catalysts [41]. For the EVS-Ag50 and EVS-Ag100, the O2 significantly enhanced the mercury removal efficiency for the Ag loaded samples, indicating that O2 played a significant role in the Hg0 removal activity over for all the samples. In Zhao’s study, it was found that O2 was not significant for the Hg0 removal performance over Ag-Mo sorbents, but O2 could
IP T
obviously enhance Hg0 removal activity of an Ag-Mo modified V2O5 based SCR catalyst [15, 16]. The results indicated that some of the Hg0 was oxidized by the
M
A
N
U
SC R
active oxygen and the consumed oxygen was replenished by gaseous O2.
ED
Figure 6. Hg0 removal efficiency over the sorbents under N2 and N2+6% O2 at 150 °C. The error bar is the standard deviation.
PT
3.5. Effect of acid gases
CC E
Individual acid gases balanced in N2 or N2+6% O2 were used to investigate their effects on the Hg0 removal. HCl was the most important species responsible for Hg0 removal and could dramatically enhance the Hg0 removal activity over most of the
A
transition metal oxides in the presence of oxygen [43]. However, in the present study, as seen in Figure 7, 10 ppm HCl had only a slight enhancing effect on the Hg0 removal activity of EVS-Ag100. The enhancing effect of NO on Hg0 removal was observed from EVS-Ag100. In a pure N2 condition, 500 ppm NO in flue gas 16
significantly enhanced the removal performance. When O2 was injected in the flue gas, the Hg0 removal efficiency was further improved. The promotional effect was probably due to the formation of oxidizing nitrites [10], which were generated via NO oxidation by the surface oxygen species. SO2 showed an inhibitory effect on the Hg0 removal activity of EVS-Ag100 under N2+1000 ppm SO2 atmosphere. When O2 was
IP T
injected into the flue gas, the Hg0 removal activity did not decrease over time. The results indicated that O2 could weaken the inhibition effect [43], because O2 could
M
A
N
U
SC R
enhance the Hg0 removal efficiency (as shown in Figure 6).
ED
Figure 7. Effect of individual acid gas components on Hg0 removal efficiency. To evaluate its potential for practical application in coal-fired power plants,
PT
EVS-Ag100 was exposed to continuous flue gas at 150 °C. The flue gas consisted of
CC E
O2, CO2, NO, SO2, HCl and Hg0 with balance gas of N2. The space velocity was about 220,000 h-1, which was extremely higher than the typical space velocity of an SCR unit in coal-fired power plants. As shown in Figure 8, EVS-Ag100 exhibited a Hg0
A
removal efficiency of 96.4%.
17
IP T
SC R
Figure 8. Hg0 removal efficiency from EVS-Ag100 under a simulated flue gas at 150 °C.
3.6. Further discussions on Hg0 removal over EVS-Ag100
U
To further understand the mercury removal process over the sorbent, the XPS
N
analysis was conducted to determine the chemical states of the vanadium and silver
A
species before and after the reaction. As shown in Figure 9(a), the peak positions of
M
Ag 3d5/2 shifted to lower binding energy after being used for Hg0 removal, indicating the formation of an Ag-Hg amalgam. Figure 9(b) shows the XPS results of vanadium
ED
species on the surface of fresh and spent EVS-Ag100. As reported, V5+ was considered to be an active species for Hg0 removal, and it would be reduced to a low
PT
valence state in the Hg0 oxidation process. However, it should be noted in this study
CC E
that after being used for Hg0 capture in the presence of O2, more V5+ was generated. Two possible reasons could account for the phenomenon. One explanation is that the vanadium with low valence states was possibly reoxidized to V5+ rapidly. The other
A
possible reason could be due to the interactions between the vanadium species and the silver species via the reaction path Ag++V4+→Ag+V5+.
18
IP T
SC R
Figure 9. XPS spectra of Ag 3d (a) and V 2p (b) of EVS-Ag100 used for Hg0 adsorption under N2+6% O2.
The Hg-TPD was carried out to analyse the mercury species on the spent EVS and
U
EVS-Ag100 that were consumed under an N2+O2 atmosphere. As shown in Figure 10,
N
two desorption peaks were observed for EVS. The desorption peaks at lower
A
temperatures (<100 °C) indicated the decomposition of weakly adsorptive mercury
M
species on the outer and inner surface of the sorbents [44]. The peak at >230 °C denoted the decomposition of the strongly adsorbed mercury species [44, 45] which
ED
were probably the oxidized mercury binding with the vanadium species. Additional
PT
desorption peaks were observed from the TPD curve of EVS-Ag100. The peaks at 150-200 °C could be attributed to the Ag-Hg amalgam [16], decomposing Hg0 and
CC E
metallic silver metal as the temperature increased. Therefore, for the EVS-Ag100 material, both the vanadium species and the silver species were responsible for Hg0 removal. In addition, it could be noted that more Hg was released at higher
A
temperatures (>230 °C) on EVS-Ag100, indicating that the loading Ag nanoparticles enhanced the activity of the vanadium species by weakening the bond strength of V-O [15].
19
IP T
SC R
Figure 10. Hg-TPD spectra of EVS-Ag100 used for Hg0 adsorption under N2+6% O2. Based on the above discussion, the primary process for Hg0 capture over the
U
EVS-Ag100 at low temperatures is depicted in Figure 11. Hg0 favoured the adsorption
N
reaction on the silver nanoparticles to form a silver amalgam. Additionally, the
A
interaction between the silver nanoparticles and the vanadium species could weaken
M
the V-O bond strength, so that the transformation of the lattice oxygen to chemical adsorption oxygen became easy, and then the Hg oxidation activity could be
A
CC E
PT
ED
enhanced.
Figure 11. Primary reaction process for Hg0 capture over EVS and EVS-Ag100.
20
4. Conclusions In the present study, the Ag modified EVS was prepared and used for Hg0 capture. For the catalytic sorbent, the vanadium was in the framework of EVS and the Ag nanoparticles were dispersed on the EVS surface. The EVS-Ag100 exhibited a superior Hg0 capture performance at temperatures around 150 °C and the Hg0 capture
IP T
capacity was about 63.4 μg·g-1. Silver nanoparticles formed on the EVS surface and
the vanadium species in the framework of EVS were the active sites for Hg0 capture.
SC R
The superior activity was due to the synergy between Ag and Mo. Furthermore, the
bond strength of V–O might be weakened, which is beneficial for the oxidation ability.
U
O2 and NO could enhance the Hg0 removal activity, while SO2 inhibited the
A
N
performance.
M
Acknowledgements
Zijian Zhou and Tiantian Cao contributed equally to this work. This
ED
research is supported by the National Key R&D Program of China (2018YFB0605200), the National Natural Science Foundation of China
PT
(51661125011) and the China Postdoctoral Science Foundation Funded Project
CC E
(2018T110762, 2017M622439). We thank Dr. Xiangkun Cao for the writing modification. We appreciate the support from the Analytical and Testing Center
A
at Huazhong University of Science & Technology and Tsinghua University.
21
References
A
CC E
PT
ED
M
A
N
U
SC R
IP T
[1] D.P. Krabbenhoft, E.M. Sunderland, Global Change and Mercury, Science 341 (2013) 1457-1458. [2] M. Xu, R. Yan, C. Zheng, Y. Qiao, J. Han, C. Sheng, Status of trace element emission in a coal combustion process: a review, Fuel Process. Technol. 85 (2004) 215-237. [3] S.J. Lee, Y.-C. Seo, H.-N. Jang, K.-S. Park, J.-I. Baek, H.-S. An, K.-C. Song, Speciation and mass distribution of mercury in a bituminous coal-fired power plant, Atmos. Environ. 40 (2006) 2215-2224. [4] M. Xu, Y. Qiao, C. Zheng, L. Li, J. Liu, Modeling of homogeneous mercury speciation using detailed chemical kinetics, Combust. Flame 132 (2003) 208-218. [5] D. Pudasainee, S.J. Lee, S.-H. Lee, J.-H. Kim, H.-N. Jang, S.-J. Cho, Y.-C. Seo, Effect of selective catalytic reactor on oxidation and enhanced removal of mercury in coal-fired power plants, Fuel 89 (2010) 804-809. [6] K.S. Park, Y.C. Seo, S.J. Lee, J.H. Lee, Emission and speciation of mercury from various combustion sources, Powder Technol. 180 (2008) 151-156. [7] Z.-J. Zhou, X.-W. Liu, B. Zhao, Z.-G. Chen, H.-Z. Shao, L.-L. Wang, M.-H. Xu, Effects of existing energy saving and air pollution control devices on mercury removal in coal-fired power plants, Fuel Process. Technol. 131 (2015) 99-108. [8] D. Jampaiah, S.J. Ippolito, Y.M. Sabri, J. Tardio, P.R. Selvakannan, A. Nafady, B.M. Reddy, S.K. Bhargava, Ceria-zirconia modified MnOx catalysts for gaseous elemental mercury oxidation and adsorption, Catal. Sci. Technol. 6 (2016) 1792-1803. [9] 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, Int. J. Coal Geol. 90-91 (2012) 4-20. [10] Z. Zhou, X. Liu, Y. Hu, J. Xu, X.E. Cao, Z. Liao, M. Xu, Investigation on synergistic oxidation behavior of NO and Hg0 during the newly designed fast SCR process, Fuel 225 (2018) 134-139. [11] J. Wilcox, E. Sasmaz, A. Kirchofer, S.-S. Lee, Heterogeneous Mercury Reaction Chemistry on Activated Carbon, J. Air Waste Manage. Assoc. 61 (2011) 418-426. [12] E. Sasmaz, A. Kirchofer, A.D. Jew, A. Saha, D. Abram, T.F. Jaramillo, J. Wilcox, Mercury chemistry on brominated activated carbon, Fuel 99 (2012) 188-196. [13] J.A. Korpiel, R.D. Vidic, Effect of Sulfur Impregnation Method on Activated Carbon Uptake of Gas-Phase Mercury, Environ. Sci. Technol. 31 (1997) 2319-2325. [14] D.-H. Lim, S. Aboud, J. Wilcox, Investigation of Adsorption Behavior of Mercury on Au(111) from First Principles, Environ. Sci. Technol. 46 (2012) 7260-7266. [15] S. Zhao, Z. Li, Z. Qu, N. Yan, W. Huang, W. Chen, H. Xu, Co-benefit of Ag and Mo for the catalytic oxidation of elemental mercury, Fuel 158 (2015) 891-897. [16] S. Zhao, H. Xu, J. Mei, Y. Ma, T. Lou, Z. Qu, N. Yan, Ag-Mo modified SCR catalyst for a co-beneficial oxidation of elemental mercury at wide temperature range, 22
A
CC E
PT
ED
M
A
N
U
SC R
IP T
Fuel 200 (2017) 236-243. [17] T. Cao, Z. Li, Y. Xiong, Y. Yang, S. Xu, T. Bisson, R. Gupta, Z. Xu, Silica-Silver Nanocomposites as Regenerable Sorbents for Hg0 Removal from Flue Gases, Environ. Sci. Technol. 51 (2017) 11909-11917. [18] Z. Zhou, X. Liu, Z. Liao, H. Shao, C. Lv, Y. Hu, M. Xu, Manganese doped CeO2-ZrO2 catalyst for elemental mercury oxidation at low temperature, Fuel Process. Technol. 152 (2016) 285-293. [19] Y. Xia, Z. Liao, Y. Zheng, Z. Zhou, Highly Dispersed Mn-Ce Binary Metal Oxides Supported on Carbon Nanofibers for Hg0 Removal from Coal-Fired Flue Gas, Appl. Sci. 8 (2018) 2501. [20] Z. Wang, J. Liu, B. Zhang, Y. Yang, Z. Zhang, S. Miao, Mechanism of Heterogeneous Mercury Oxidation by HBr over V2O5/TiO2 Catalyst, Environ. Sci. Technol. 50 (2016) 5398-5404. [21] Y. Yang, J. Liu, Z. Wang, F. Liu, Heterogeneous reaction kinetics of mercury oxidation by HCl over Fe2O3 surface, Fuel Process. Technol. 159 (2017) 266-271. [22] H. Li, S. Wu, C.-Y. Wu, J. Wang, L. Li, K. Shih, SCR Atmosphere Induced Reduction of Oxidized Mercury over CuO-CeO2/TiO2 Catalyst, Environ. Sci. Technol. 49 (2015) 7373-7379. [23] H. Li, L. Zhu, S. Wu, Y. Liu, K. Shih, Synergy of CuO and CeO2 combination for mercury oxidation under low-temperature selective catalytic reduction atmosphere, Int. J. Coal Geol. 170 (2017) 69-76. [24] A.A. Presto, E.J. Granite, Survey of Catalysts for Oxidation of Mercury in Flue Gas, Environ. Sci. Technol. 40 (2006) 5601-5609. [25] Z. Zhou, X. Liu, Z. Liao, H. Shao, Y. Hu, Y. Xu, M. Xu, A novel low temperature catalyst regenerated from deactivated SCR catalyst for Hg0 oxidation, Chem. Eng. J. 304 (2016) 121-128. [26] H. Li, Y. Li, C.-Y. Wu, J. Zhang, Oxidation and capture of elemental mercury over SiO2-TiO2-V2O5 catalysts in simulated low-rank coal combustion flue gas, Chem. Eng. J. 169 (2011) 186-193. [27] S. Eswaran, H.G. Stenger, Effect of halogens on mercury conversion in SCR catalysts, Fuel Process. Technol. 89 (2008) 1153-1159. [28] F. Mani, J.A. Sawada, S.M. Kuznicki, Comparative adsorption study of EVS-10 and ETS-10, Microporous Mesoporous Mater. 204 (2015) 43-49. [29] F. Mani, J.A. Sawada, S.M. Kuznicki, Spontaneous formation of silver nanoparticles on the vanadium silicate EVS-10, Microporous Mesoporous Mater. 224 (2016) 208-216. [30] Y. Qiu, B. Liu, J. Du, Q. Tang, Z. Liu, R. Liu, C. Tao, The monolithic cordierite supported V2O5-MoO3/TiO2 catalyst for NH3-SCR, Chem. Eng. J. 294 (2016) 264-272. [31] J. Yang, Q. Yang, J. Sun, Q. Liu, D. Zhao, W. Gao, L. Liu, Effects of mercury oxidation on V2O5-WO3/TiO2 catalyst properties in NH3-SCR process, Catal. Commun. 59 (2015) 78-82. 23
A
CC E
PT
ED
M
A
N
U
SC R
IP T
[32] S. Zhao, Y. Ma, Z. Qu, N. Yan, Z. Li, J. Xie, W. Chen, The performance of Ag doped V2O5-TiO2 catalyst on the catalytic oxidation of gaseous elemental mercury, Catal. Sci. Technol. 4 (2014) 4036-4044. [33] T. Cao, Z. Zhou, Q. Chen, Z. Li, S. Xu, J. Wang, M. Xu, T. Bisson, Z. Xu, Magnetically responsive catalytic sorbent for removal of Hg0 and NO, Fuel Process. Technol. 160 (2017) 158-169. [34] Y. Liu, D.J.A. Kelly, H. Yang, C.C.H. Lin, S.M. Kuznicki, Z. Xu, Novel Regenerable Sorbent for Mercury Capture from Flue Gases of Coal-Fired Power Plant, Environ. Sci. Technol. 42 (2008) 6205-6210. [35] J. Dong, Z. Xu, S.M. Kuznicki, Mercury Removal from Flue Gases by Novel Regenerable Magnetic Nanocomposite Sorbents, Environ. Sci. Technol. 43 (2009) 3266-3271. [36] Y. Zheng, A.D. Jensen, C. Windelin, F. Jensen, Review of technologies for mercury removal from flue gas from cement production processes, Prog. Energy Combust. Sci. 38 (2012) 599-629. [37] H. Li, S. Wu, L. Li, J. Wang, W. Ma, K. Shih, CuO-CeO2/TiO2 catalyst for simultaneous NO reduction and Hg0 oxidation at low temperatures, Catal. Sci. Technol. 5 (2015) 5129-5138. [38] X. Ge, H.-l. Zhang, Temperature-Programmed Reduction Studies of V-Ag Catalysts, J. Solid State Chem. 141 (1998) 186-190. [39] C. Heske, R. Fink, E. Umbach, W. Riedl, F. Karg, Na-induced effects on the electronic structure and composition of Cu(In,Ga)Se2 thin-film surfaces, Appl. Phys. Lett. 68 (1996) 3431-3433. [40] Y. Liu, C. Tian, B. Yan, Q. Lu, Y. Xie, J. Chen, R. Gupta, Z. Xu, S.M. Kuznicki, Q. Liu, H. Zeng, Nanocomposites of graphene oxide, Ag nanoparticles, and magnetic ferrite nanoparticles for elemental mercury (Hg0) removal, RSC Adv. 5 (2015) 15634-15640. [41] B. Zhao, X. Liu, Z. Zhou, H. Shao, C. Wang, J. Si, M. Xu, Effect of molybdenum on mercury oxidized by V2O5-MoO3/TiO2 catalysts, Chem. Eng. J. 253 (2014) 508-517. [42] G. Luo, H. Yao, M. Xu, X. Cui, W. Chen, R. Gupta, Z. Xu, Carbon Nanotube-Silver Composite for Mercury Capture and Analysis, Energy Fuels 24 (2010) 419-426. [43] H. Li, C.-Y. Wu, Y. Li, L. Li, Y. Zhao, J. Zhang, Role of flue gas components in mercury oxidation over TiO2 supported MnOx-CeO2 mixed-oxide at low temperature, J. Hazard. Mater. 243 (2012) 117-123. [44] Q. Wan, L. Duan, K. He, J. Li, Removal of gaseous elemental mercury over a CeO2-WO3/TiO2 nanocomposite in simulated coal-fired flue gas, Chem. Eng. J. 170 (2011) 512-517. [45] Q. Wan, L. Duan, J. Li, L. Chen, K. He, J. Hao, Deactivation performance and mechanism of alkali (earth) metals on V2O5-WO3/TiO2 catalyst for oxidation of 24
A
CC E
PT
ED
M
A
N
U
SC R
IP T
gaseous elemental mercury in simulated coal-fired flue gas, Catal. Today 175 (2011) 189-195.
25
Figure captions
Figure 1. XRD patterns of EVS, EVS-Ag50 and EVS-Ag100. Ag 3d (a), V 2p (b) and Na 1s (c) XPS spectra of the samples.
Figure 3.
TEM images of EVS (a & b), EVS-Ag50 (c & d) and EVS-Ag100 (e & f).
Figure 4.
Mercury breakthrough at different temperatures. The error bar is the
IP T
Figure 2.
standard deviation.
Hg capture capacity of the samples. The error bar is the standard
SC R
Figure 5. deviation.
Hg0 removal efficiency over the sorbents under N2 and N2+6% O2 at
U
Figure 6.
150 °C. The error bar is the standard deviation.
Effect of individual acid gas components on Hg0 removal efficiency.
Figure 8.
Hg0 removal efficiency from EVS-Ag100 under a simulated flue gas at
M
A
N
Figure 7.
150 °C.
XPS spectra of Ag 3d (a) and V 2p (b) of EVS-Ag100 used for Hg0
ED
Figure 9.
adsorption under N2+6% O2.
O2.
Primary reaction process for Hg0 capture over EVS and EVS-Ag100.
A
CC E
Figure 11.
Hg-TPD spectra of EVS-Ag100 used for Hg0 adsorption under N2+6%
PT
Figure 10.
26
Table legends
Table 1. XRF results expressed as molar ratios to vanadium.
A
CC E
PT
ED
M
A
N
U
SC R
IP T
Table 2. BET surface area and pore analysis for the sorbents.
27