In-situ catalytic oxidation of Hg0 via a gas diffusion electrode

Chemical Engineering Journal 310 (2017) 170–178

Contents lists available at ScienceDirect

Chemical Engineering Journal journal homepage: www.elsevier.com/locate/cej

In-situ catalytic oxidation of Hg0 via a gas diffusion electrode Yi Xu 1, Limei Cao 1, Wei Sun, Ji Yang ⇑ School of Resources and Environmental Engineering, State Environmental Protection Key Laboratory of Environmental Risk Assessment and Control on Chemical Process, East China University of Science and Technology, Shanghai 200237, PR China

h i g h l i g h t s

g r a p h i c a l a b s t r a c t


A novel diffusion electrochemical

reactor was proposed to remove Hg0.
The reaction mechanism of Hg

0

removal was investigated.
Hydrogen peroxide and hydroxyl

radicals play a dominant role for Hg0 removal.

a r t i c l e

i n f o

Article history: Received 21 July 2016 Received in revised form 17 October 2016 Accepted 20 October 2016 Available online 21 October 2016 Keywords: Elemental mercury Gas diffusion electrode H2O2 HO

a b s t r a c t A diffusion electrochemical reactor was proposed to remove elemental mercury (Hg0) from coal-fired flue gas. The experiments were carried out in an undivided column reactor with a self-made gas diffusion electrode (GDE) as cathode and Ti/IrO2 as anode. Hg0 was oxidized when the simulated gas passed through GDE. It turned out that the removal of Hg0 in electrochemical process was dominated by electro-generated H2O2 and free radicals on GDE interface or in the electrolyte. Under 70 °C, Hg0 removal efficiency exceeded 90% after 40 min electrolysis. The effects of operation parameters were investigated and the results demonstrated that voltage, gas flow rate, and initial concentration of Hg0 had significant influences on Hg0 removal, especially the reaction temperature, electrolyte concentration and pH. H2O2 and HO in the electrolyte were measured by UV–vis spectrophotometer and electron spin resonance (ESR), respectively. The above results show that electrochemical technique is a promising method for emission control of Hg0 from coal-fired flue gas. Ó 2016 Elsevier B.V. All rights reserved.

1. Introduction Mercury, a hazardous air pollutant (HAP) as designated by US EPA, has received a considerable attention due to its high toxicity, long range transport, persistent and bioaccumulation [1]. It was estimated that about 2320 t of Hg was emitted to the globe atmosphere annually [2]. Coal combustion accounting for more than ⇑ Corresponding author at: East China University of Science and Technology, Shanghai 200237, PR China. E-mail address: [email protected] (J. Yang). 1 These authors contributed equally to this work. http://dx.doi.org/10.1016/j.cej.2016.10.095 1385-8947/Ó 2016 Elsevier B.V. All rights reserved.

60% of total mercury release, was regarded as the main source of mercury emissions to the atmosphere [3]. The mercury released to the environment can be converted through biological processes into methyl-mercury (Me-Hg), which is highly toxic and may cause neurological damage depending on its concentration in food or water [4]. Therefore, it is significant to control the emission of mercury to the atmosphere. In the flue gas derived from coal-fired power plants, particulate mercury (HgP), oxidized gaseous mercury (Hg2+) and elemental mercury (Hg0) are the main forms. The HgP and Hg2+ can be effectively removed through particulate matter collection devices and

Y. Xu et al. / Chemical Engineering Journal 310 (2017) 170–178

wet flue gas desulfurization facilities, respectively. However, Hg0 is difficult to be removed due to its high volatility and insolubility [5]. Hence, two main strategies are used to remove Hg0 from flue gas: Hg0 adsorption and oxidize Hg0 to Hg2+. The former technology is costly by injecting activated carbon [6] and novel sorbents [7] into the flue gas. The latter one includes the addition of strong oxidants like halogens, ozone or H2O2 to the flue gas [8], photochemical oxidation, using catalysts [9] or membranes with catalytic oxidation system [10]. However, some technical problems such as the application costs, safety, reliability and secondary pollution of reaction products can’t be effectively resolved yet [11]. Therefore, the development of high efficiency, low pollution and economic technology is urgently needed nowadays. Recently, the advanced oxidation technology has been widely studied to solve environmental problems. The highly active and non-selective hydroxyl radicals (HO) produced from H2O2, UV/H2O2 and metal/H2O2 can oxidize many kinds of pollutants. So far, various reports are concerning the use of commercial H2O2 to remove Hg0 from flue gas. Korell, J. et al. [12] applied H2O2 solution to oxidize Hg0 to Hg2+ from the flue gas in the wet scrubber. Yangxian Liu et al. [11] remove Hg0 from flue gas containing SO2/NO by UV/H2O2 process in a novel photochemical reactor. Nevertheless, the cost and hazards associated with the transport and handling of commercial concentrated H2O2 can’t be neglected [13]. Moreover, a notable shortcoming exists in this chemical process is the fast consumption of H2O2 and the need of adding H2O2 continuously. The above problems can be solved if H2O2 can be generated in a simple system efficiently and continuously. In recent decades, lots of researches have proved that H2O2 could be in-situ electro-generated by the two electron reduction of oxygen in an electrochemical system [14]. However, traditional cathode materials such as graphite felt and ACF are not efficient in H2O2 production due to the low solubility of oxygen. This limitation can be overcome when O2 is supplied through gas diffusion electrode (GDE). It has been demonstrated that GDE can enhance H2O2 productivity due to its excellent electrical conductivity, high surface area and porous structures, leading to a faster reduction of oxygen. Moreover, the porous structures contain abundant hydrophobic channels, which enormously supply unlimited oxygen to the electrode/electrolyte interface rather than dissolve into the solution [15]. In these experiments, pure air containing mercury vapor was used as the simulated gas and it was diverted to GDE directly. GDE catalyzed the reduction of oxygen to produce H2O2, which was from simulated flue gas without other pumped oxygen. The high concentration of electro-generated H2O2 was electroactivated and large amount of highly reactive radicals like OH was produced. H2O2 and OH existing in the electrolyte could oxidize Hg0 directly and kept it in the solution when mercury vapor passed through the porous structure of GDE and the electrolyte. This is the design idea to convert Hg0 by electrochemical process and with our effort, desirable results were achieved. The innovative approach developed in our present work consists of investigating into the feasibility of using GDE for the in situ electro-generation of H2O2 and the indirect electrooxidation of Hg0 from flue gas in a well-designed electrochemical reactor. In current study, a Ti/IrO2 anode and a self-made gas diffusion electrode as the cathode were used for the electrochemical oxidation of Hg0. The cooperative oxidation of electro-generated H2O2 with reactive oxygen radicals and active anode is supposed to increase the oxidation Hg0 rate markedly. Sequences of experiments were investigated to explore the relevant operational parameters effect on H2O2 production as well as Hg0 removal efficiency.

171

2. Experimental section 2.1. Chemical and materials The carbon black powder (CB, Vulcan XC-72R) was purchased from Cabot Corporation and used without any further treatment. Graphite (Sinopharm Chemical Reagent Co., Ltd, China.) was pretreated: boiled in deionized water for 2 h, rinsed twice with deionized water, filtered with a filter, dried at 80 °C and then stored for later use. Polytetrafluoroethylene emulsion solution (PTFE, 301B, 60 wt%, Shanghai 3F New Materials Co., Ltd, China.) 5,5-dimethy l-1-pyrroline N-oxide (DMPO, P97%) was bought from Aladdin Chemical Reagent (Shanghai) Co., Ltd, China. Mercury permeation tube was purchased from Suzhou Qingan Instrument Co., Ltd, China. The commercial Ti/IrO2 was purchased from Baoji Yichen Technology Co., Ltd, China. The loading of Ti/IrO2 anode is 1 mg IrO2 per cm2. 2.2. Preparation of gas diffusion electrode (GDE) The gas diffusion electrode was consisted of three parts: the stainless wire was employed as substrate, the graphite gas diffusion layer and carbon black (XC-72) catalyst layer. The fabrication procedure was as follows: 0.2 g pretreated graphite and 0.1 g carbon black (Vulcan XC-72) were added into 60 mL 3% ethanol in a conical flask, respectively. And the conical flasks were put in an ultrasonic bath for 15 min at room temperature, and then appropriate amount of PTFE was added into the mixtures and kept ultrasonic dispersion for another 15 min. Next, the mixtures were stirred at 80 °C until they became desirable flocks and the flocks were filtered on a circulate stainless steel wire in order to form a paste. The resulting paste was put into a mold to be pressed under 20 MPa for 3 min to form a circulate sheet (R = 22.5 mm, 1.7 mm thickness). Via calcination at 350 °C for 60 min, a GDE was fabricated. 2.3. Experimental apparatus The experimental installations are consist of gas cylinder (purity, 99.99%), mass flow meter, water bath, mercury vapor generator (Suzhou Qingan Instrument Co., Ltd, China), self-designed reactor, direct current constant voltage power supply with currentvoltage monitor, PC and mercury vapor analyzer (SG-921, Jiangsu Jiangfen Electro-analytical Instrument Co., Ltd, China.). All the reactions were carried out in an undivided self-designed column reactor with effective sectional area of 7.07 cm2 and volume of 100 mL as shown in Fig. 1(a). Fig. 1(b and c) shows the Ti/IrO2 anode and gas diffusion cathode, respectively. Mercury permeation tube was placed in a sealed U shaped glass tube which was immersed in a temperature controlled water bath. By adjusting the temperature of the water bath, Hg0 Concentration in the simulated gas could be controlled. A laboratory direct current constant voltage power supply and current-voltage monitor was used to provide the electric power and detect the current. A continuous mercury analyzer was used to detect the inlet and outlet Hg0 Concentration. Vent gases were treated with potassium permanganate saturated solution to avoid air pollution. 2.4. Electrolytic procedures 80 mL Na2SO4 solution was used as supporting electrolyte and the initial pH was adjusted by H2SO4 and NaOH. The prepared GDE (Section 2.2) was used as cathode and Ti/IrO2 (R = 15 mm) as the anode. The distance between Ti/IrO2 anode and GDE cathode was 10 mm. Pure air (21% O2 with N2 left) was used as oxygen

172

Y. Xu et al. / Chemical Engineering Journal 310 (2017) 170–178

Fig. 1. Schematic diagram of electrochemical system: (a) reactor, (b) Ti/IrO2 anode, (c) gas diffusion cathode, (d) SEM of diffusion layer, (e) SEM of catalyst layer.

source and gas carrier was firstly transferred to GDE and then diverted into the electrolyte. During the electrolytic process, the electrolyte was periodically sampled and analyzed. The electrochemical oxidation of Hg0 process without adding any catalyst was performed in the same apparatus and the simulated flue gas was composed of 350 lgm3 Hg0 and 21%O2 with N2 left. Before the electrochemical removal of Hg0 was conducted, the mixed gas was led to the mercury analyzer to determine the inlet Hg0 Concentration. Then it was diverted to the reactor until the adsorption had been reached a balance and the balanced Hg0 Concentration recorded by the mercury analyzer was denoted as the initial Hg0 Concentration.

Hg0 removal efficiency g was calculated according to the following equation (Eq. (2)):



g¼ 1

 C  100% C0

ð2Þ

where C and C0 represent the outlet Hg0 Concentration and initial Hg0 Concentration (lgm3), respectively. The DMPO free radicals adducts were determined at ambient temperature on an ESR spectrometer (Bruker BioSpin Gmbh) combining with 5,5-Dimethyl-1-Pyrroline N-Oxide (DMPO). The settings of EPR spectrometer were: center field: 3510 G; sweep width, 150 G; microwave frequency, 9.873 GHz; modulation frequency, 100 kHz; power, 6.390 mW; modulation amplitude, 1 G.

2.5. Analytical methods 3. Results and discussion The surface morphology of GDE was analyzed by scanning electron microscopy (SEM). Linear sweep voltammetry (LSV) was carried out to compare the electrochemical activity of GDE under different conditions. The performance was recorded by the E5000 workstation at a scan rate of 50 mV/s in a three-electrode system, using GDE as the working electrode, Ti/IrO2 as the counter electrode and a saturated calomel electrode as reference electrode in 0.1 M Na2SO4 solution at ambient temperature. The concentration of H2O2 was determined by UV–vis spectrophotometer (UV-4802, Unico (Shanghai) Instrument Co., Ltd.) using (NH4)6Mo7O24 in H2SO4 [16]. The current efficiency (CE) of H2O2 production was calculated as the following formula (Eq. (1)) [17]:

nFCV CE ¼ R t  100% Idt 0

ð1Þ

Here, n is the number of electrons transferred for oxygen reduction to H2O2, F is the Faraday constant (96,485 C/mol), C is the concentration of H2O2 (mol/L), V is the bulk volume (L), I is the current (A) and t is the electrolysis time (s). The Hg0 Concentration in the gas was monitored by the continuous mercury analyzer with cold atomic absorption spectrometry.

3.1. Characterization of GDE The amount of hydrogen peroxide produced at the cathode was a decisive parameter and was directly dependent on the electrode characteristics (electrical conductivity, porosity and electroactivity of the surface) [18]. Therefore, GDE was characterized by SEM, BET, LSV and ESR. The prepared gas diffusion cathode for electro-generation of H2O2 was composed of three parts: the stainless wire was used as substrate, the graphite with flake structure employed as gas diffusion layer and carbon black (XC-72) served as catalyst to produce H2O2 by oxygen reduction. The morphology of cathode was characterized by SEM. As shown in Fig. 1(d), graphite flake was quite large and the surface was very smooth, which prevented from immersing the electrode. The surface of diffusion layer was composed of PTFE and graphite with flake structure, different sizes and irregular gap between them. The morphology of carbon black was quite different from graphite and it was presented in Fig. 1(e). The numerous pellets were carbon black particles with much smaller size and mesoporous structure, which would enhance its surface area. The performed BET also revealed that graphite flake gave a 6.9 m2g1 while XC-72 had a larger surface area of

Y. Xu et al. / Chemical Engineering Journal 310 (2017) 170–178

235.5 m2g1. It was obvious that there were many interspaces existing in GDE. The performance of prepared cathode for electro-generation of H2O2 was tested by line scan voltammetry (LSV). It could be observed from Fig. 2 that only hydrogen evolution reaction (HER) happened under N2 saturated solution between 0.1 V and 1.3 V vs. SCE in 0.1 M Na2SO4 solution. When tested in O2 saturated solution, higher current response was obtained due to the contribution of oxygen reduction reaction (ORR), and here it was mainly from the oxygen reduction to H2O2. Many researchers reported that carbon black (XC-72) preferred to go through two electrons transfer route to generate H2O2 rather than four electrons transfer process for H2O [13,19]. Another feature was the fact that before 0.35 V, the voltammetry curves in N2 and O2 saturated solution were roughly similar. It meant that the onset-potential of XC-72 catalyst for generating H2O2 might be 0.35 V. In here, we employed XC-72 to in-suit generate H2O2 for removing Hg0 from simulated flue gas. Thus, the performance of GDE to electro-generate H2O2 is a key point. We will demonstrate the capacity of H2O2 production on gas diffusion cathode and its Hg0 removing performance under different conditions in below discussions. The ESR spectrum was obtained after 30 min electrolysis in neutral medium with DMPO to trap free radicals. As shown in Fig. 3, DMPO-HO adducts were formed in the electrolyte as a quartet of signals with intensities of 1:2:2:1 was detected. It can be seen from the characterized spectrum that the hyperfine splitting constants of DMPO free adducts were aN = aH = 15 G, which were in accordance with the literature [20] reported and conforming that HO were exist in our electrochemical system. According to Fig. 3(a), there was little HO formed by feeding N2, while, DMPO-HO adducts were largely formed with the adding of oxygen. Furthermore, the intensity was shoot up with the increasing electrolyte concentration as shown in Fig. 3(b). 3.2. Hydrogen peroxide electro-generated on GDE In an undivided electrolysis appliance, the gas diffusion electrode catalyzed the two-electron reduction of O2 to H2O2 (Eq. (3)) [21], and the 4-electron process of ORR (Eq. (4)), should be avoided as much as possible. Several operation parameters were carried out to find out the optimal conditions for H2O2 electro-generation.

173

Fig. 2. LSV obtained at Pt electrode with Na2SO4 (0.1 M) as supporting electrolyte in the absence and presence of O2; scan rate of 50 mV/s.

The enhanced air flow rate could promote the mass transfer rate of O2 to the surface of GDE, which was beneficial for the electrogeneration of H2O2 [23]. As shown in Fig. 4(b2), the current efficiencies were 75.1%, 79.8%, 81.9%, 85.4%, correspondingly. The current efficiencies were improved with the increased flow rate as well since the mass transfer was enhanced by the agitation. It can be seen from Fig. 4(c1 and c2) that when the pH values varied from 3 to 11, the accumulation of H2O2 in 120 min were 22.3, 24.3, 31.7, 28.9, 26.7 mmol/L and current efficiencies were 45.8%, 52.0%, 77.68%, 67.7%, 62.5%, accordingly. H2O2 concentration rose slightly with the increasing pH from 3 to 7, however, declined when the solution was more alkaline. In acid solution, the redundant hydrogen ions would compete with the existing H2O2 and consume oxygen to form H2O (Eqs. (4) and (7)). As for the basic solution, the abundant OH and OH 2 could decompose H2O2 (Eqs. (8) and (9)), resulting in the drop of H2O2 concentration and current efficiency [17].

Acidic medium : H2 O2 þ 2Hþ þ 2e ! 2H2 O

ð7Þ

Alkaline medium : H2 O2 þ OH ! HO2 þ H2 O

ð8Þ ð9Þ

O2 þ 2Hþ þ 2e ! H2 O2

ð3Þ

H2 O2 þ HO2 ! HO þ  O2 þ H2 O

O2 þ 4Hþ þ 4e ! 2H2 O

ð4Þ

The effect of voltage on H2O2 production are shown in Fig. 4(a1) The yield of H2O2 with Na2SO4 of 0.1 mol/L, air flow rate of 50 mL/min after 120 min electrolysis at the voltage varied from 3 V to 6 V were 11.8, 24.3, 33.3, 45.6 mmol/L, respectively. It indicated that the H2O2 accumulation increased with the increased voltage. As for the current efficiency, the trend was just the opposite and this was agreement with literature [22]. The according current efficiencies were 84.1%, 81.6%, 79.9%, 76.5%, respectively. As seen in Fig. 4(a2), the current efficiency reduced with the increased voltage, on account of some side reactions, for instance, H2O2 evolution on the cathode (Eq. (5)) and oxidation on the anode (Eq. (6)) may occur and become more apparent with the increased applied voltage.

The H2O2 electro-generation in neutral medium at constant potential of 4 V for 120 min in order to study the effect of electrolyte concentration was depicted in Fig. 4(d1). When the Na2SO4 concentrations were 0.05, 0.1, 0.5, 0.8 mol/L, the accumulations of H2O2 were 13.7, 24.7, 44.2, 51.6 mmol/L and the trend was accordance with the HO amount in the electrolyte. It could be inferred that the amount of HO was positively related to the H2O2 concentration and the HO maybe produced by electro-activated H2O2 (Eq. (10)). The according current efficiencies shown in Fig. 4(d2) were 65.1%, 77.9%, 80.7%, 85.7%, respectively. It was denoted that with the increment of Na2SO4 concentration, H2O2 concentration increased as well as the current efficiency. On account of the increased Na2SO4 concentration enhanced the conductivity and more electrons were captured for the reduction of O2 to H2O2.

H2 O2 þ 2e þ 2Hþ ! H2 O

ð5Þ

H2 O2 þ e ! HO þ OH

H2 O2  2e þ OH ! O2 þ 2H2 O

ð6Þ

As shown in Fig. 4(e1 and e2), both of the H2O2 concentrations and the current efficiencies were dropped dramatically with the elevated temperature. The phenomenon could be interpreted that H2O2 was extremely unstable when reaction temperature was above to 50 °C and most of the accumulated H2O2 were

Fig. 4(b1) showed the effect of air flow rate on H2O2 production. The accumulation of H2O2 in 120 min at the air flow rate of 25, 50, 75, 100 mL/min were 21.7, 24.3, 28.1, 31.8 mmol/L, respectively.

ð10Þ

174

Y. Xu et al. / Chemical Engineering Journal 310 (2017) 170–178

Fig. 3. ESR spectrum of OH radicals trapped by DMPO: (a) Effects of different oxygen concentration. (b) Effects of different electrolyte concentration. Conditions: 4 V, 50 mL/ min gas flow rate, pH 7.0, 0.5 M Na2SO4, 21% O2, 10 °C.

decomposed at high temperature (Eq. (11)). On the other hand, the solubility of O2 was declined with temperature went up, which may affect the production of H2O2 as well. heat

2H2 O2 ! 2H2 O þ O2

ð11Þ

3.3. Electrochemical catalytic oxidation of Hg0 with GDE Distinctly, the standard electrode potential of H2O2 (1.763 V) is higher than that of Hg2+/Hg0 (0.796 V), which means that Hg0 Can be oxidized by H2O2 (Eq. (12)) [24]. Moreover, when the GDE catalyzes the two-electron reduction of O2 to H2O2 and it can be converted to HO and HO2(Eqs. (10) and (13)), whose oxidation capacity are even stronger than H2O2 (Eqs. (14) and (15)) [25]. The reaction can be described as follows [26]:

H2 O2 þ 2Hg0 þ 2Hþ ! Hg2þ 2 þ 2H2 O

ð12Þ

H2 O2 þ HO ! HO2 þ H2 O

ð13Þ

2Hg0 þ 2HO þ 2Hþ ! Hg2þ 2 þ 2H2 O

ð14Þ

 þ 2þ þ 2H2 O Hg2þ 2 þ 2HO2 þ 2H ! 2Hg

ð15Þ

0

Therefore, the performance of Hg removal in indirect electrochemical technology is essentially dependent on the existing H2O2 and oxidative free radicals. The high H2O2 yielded in situ presented above inaugurating the possibility for Hg0 oxidation with high-efficiency in electrochemical process. 3.3.1. Effect of voltage on Hg0 removal Hg0 removal may be depended on the capacity of electron transfer as well as the electro-generation of H2O2 and free radicals. In order to confirm the effects of voltage on Hg0 removal, a series of electrolysis was carried out in selected voltages ranging from 3 V to 6 V and the results were presented in Fig. 5(a). The Hg0 removal efficiency surged dramatically in the initial 10 min as H2O2 and free radicals were increased in the electrolyte. While, after 10 min, it was almost remained constant at 3 V and 4 V, but moderately increased at 5 V and 6 V. The Hg0 removal efficiency was enhanced with the increment of voltage and this is in agreement with H2O2 yield as well as HO, which was exactly verified that the Hg0 removal is directly related to H2O2 and HO concentration. Electro-generation of H2O2 with the combination of the side

reactions are occurring simultaneously during the electrolysis process. Furthermore, the side reactions could be intensified with time went by due to the accumulation of H2O2. This is the reason why 10 min later, Hg0 removal was slightly increased or remains stable. In summary, the increment of voltage was beneficial to the removal of Hg0. 3.3.2. Effect of gas flow rate on Hg0 removal To verify the effect of gas flow rate on Hg0 removal, four flow rates (25, 50, 75, 100 mL/min) were used and the results were provided in Fig. 5(b). Hg0 removal efficiency kept an upward tendency with the growth of gas flow rate and this phenomenon could be explained by two mechanisms [27]. One is the oxygen demand, increasing gas flow rate enhances oxygen content in the electrolyte as well as the surface of GDE, which can accelerate the production of H2O2. Another is the mass transfer, the agitation of the electrolyte becomes severe with the gas flow rate increased, which can facilitate the contact between the Hg0 and electrolyte as well as the active site on GDE. In brief, the increased gas flow rate could enhance the Hg0 removal. 3.3.3. Effect of initial pH on Hg0 removal For the effect of electrochemical performance, pH is one of the major factors. Therefore, the effect of pH varying from 3 to 11 on Hg0 removal was investigated. H2SO4 and NaOH solutions were used to adjust pH. It can be seen from Fig. 5(c), the alkaline electrolyte was more beneficial for Hg0 removal than that in neutral or acidic medium. It is known that the removal of Hg0 was ascribed to the cooperative oxidation processes of anode and electrogenerated H2O2 with free radicals. Moreover, in alkaline medium, H2O2 was unstable and could be converted to strong oxidizing active radicals like HO, HO2 and O 2 more easily. However, H2O2 in acid electrolyte remains stable, especially when pH is below 3.5. For this reason, the Hg0 removal in alkaline electrolyte was more favorable than that in acid or neutral media. 3.3.4. Effect of electrolyte concentration on Hg0 removal The effect of electrolyte concentration on Hg0 removal was investigated since electro-conductivity was a significant factor in the electrochemical process. Well electro-conductivity conduced to quicker electron transfer and better Hg0 removal rate. The concentration of Na2SO4 was set from 0.05 mol/L to 0.8 mol/L. As shown in Fig. 5(d), it was apparent that an increase of Na2SO4 concentration led to the enhancement of Hg0 removal. This could be

Y. Xu et al. / Chemical Engineering Journal 310 (2017) 170–178

175

Fig. 4. Effects of various factors on H2O2 electro-generation and its current efficiency: (a) voltage; (b) gas flow rate; (c) pH; (d) Na2SO4 concentration; (e) reaction temperature. Conditions: 4 V, 50 mL/min gas flow rate, pH 7.0, 0.1 M Na2SO4, 21% O2, 10 °C.

176

Y. Xu et al. / Chemical Engineering Journal 310 (2017) 170–178

Fig. 5. Effects of diversified parameters on Hg0 removal: (a) voltage; (b) gas flow rate; (c) pH; (d) Na2SO4 concentration; (e) initial Hg0 Concentration; (f) reaction temperature. Conditions: 4 V, 50 mL/min gas flow rate, pH 7.0, 0.1 M Na2SO4, 21% O2, 10 °C, 350 lg/m3 initial Hg0 concentration.

explained from two reasons. First, the increased electrolyte concentration could enhance the conductivity and resulted in faster mass transfer and lower energy consumption [28]. Second, SO2 4 could react with HO to form SO-4 (the redox potential up to 2.6 V) (Eq. (16)) [29], whose oxidation capacity was weaker than HO (the redox potential was 2.8 V), however, it can also oxidize Hg0 (Eq. (17)) [30]. Therefore, the first reason should be the primary one.  þ   SO2 4 þ OH þ H ! SO4 þ H2 O

ð16Þ

2  SO4 þ Hg0 ! Hg2þ þ 2SO2 4

ð17Þ

3.3.5. Effect of Hg0 Concentration on Hg0 removal As this work is a fundamental work for the electro-oxidation of industrial Hg0 pollution control, the initial concentration of Hg0 in the range of 50–350 lg/m3 was carried out to investigate. It could be observed from Fig. 5(e) that with the initial concentration of Hg0

increased from 50 lg/m3 to 350 lg/m3, the Hg0 removal efficiency decreased from 63% to 35% after 40 min electrolysis. This phenomenon could be explained from two reasons. First, increasing the Hg0 initial concentration will increase the number of Hg0 through GDE per unite area and time, meanwhile, the relative molecule ratio of H2O2 and HO will be decreased [30]. Second, more intermediate like Hg+ was produced as the reaction proceeded under high Hg0 Concentration, which will compete for the oxidizing agents and influence the removal efficiency. However, it can be inferred from the above results that the first aspect is the major reason, which results in the decrease Hg0 removal in high initial Hg0 Concentration. Nevertheless, the removed quantity of Hg0 increased with the increment of initial Hg0 Concentration. In considering of the quantum yield [31], increased Hg0 Concentration leads to higher probability that Hg0 atoms reacted with the H2O2 and free radicals. This promoted the initial rate and the numbers of Hg0 atoms reacted.

Y. Xu et al. / Chemical Engineering Journal 310 (2017) 170–178

3.3.6. Effect of reaction temperature on Hg0 removal For the reason that the outlet of flue gas always carries high quantity of heat, the temperature of solution will increase inevitably. Therefore, the influence of temperature on Hg0 removal is carried out in a series temperature of selected ranging from 10 °C to 70 °C and the results were given in Fig. 5(f) The Hg0 removal rates were 34.5%, 52.9%, 71.4%, 90.6%, respectively, indicating that increasing reaction temperature has an apparent positive impact on Hg0 removal. This phenomenon can be explained from two aspects. First, according to Arrhenius equation [32], the kinetic energy of molecules as Hg0, H2O2, O2 were promoted and the activated molecules were increased in high temperature, thus, the chance of one molecule to react with another is boost and leads to desirable Hg0 removal rate. Second, H2O2 is unstable and easily to be decomposed at high temperature [33], which leads to the increment of free radicals like HO2, HO and O 2 , so that Hg0 removal rate was improved. Although, the solubility of O2 decreased with the increased temperature, the GDE offered enough gas-water-electrode contacting area for the generation of H2O2 and active radicals. For the existing large amount of free radicals, the declined solubility of O2 could be neglected. As a consequence, desirable Hg0 removal rate can be obtained at high temperature.

3.3.7. Mechanism of electrochemical Hg0 removal via GDE To explore Hg0 removal mechanism on self-made GDE during the electrolysis, a series of comparative experiments were carried out: (a) air as gaseous source; (b) 6% oxygen with nitrogen as gaseous source; (c) nitrogen as gaseous source; (d) sole-H2O2 and no electrolysis; (e) proton exchange membrane used to separate the anode and cathode and the simulated gas was led to the anode compartment only. According to Fig. 6, negligible Hg0 removal was achieved when the simulated gas was treated by sole-H2O2 as oxidant and only anode compartment due to their limited oxidation power. This could be explained that the dominant oxidant to remove Hg0 was free radicals like HO rather than sole-H2O2 and the simulated gas couldn’t be thoroughly exposed to the anode because there was a big hole on it to let gas pass. It was clearly that less than 12% Hg0 removal was obtained by the electrochemical process feeding with nitrogen. This low removal efficiency was mostly due to the oxidation of the anode and the scarcely oxidation substances generated from cathode. In comparison, it was obvious that a significant increase of Hg0 removal occurred when feeding 6% oxygen or air, which proved that the ORR process could be beneficial to remove Hg0 from flue gas. It was apparently that the Hg0 removal rate looked alike with 6% oxygen and air. This phenomenon could be explained that before each experiment started, the gas was led to the electrolyte for the Hg0 absorption; meanwhile, the oxygen was saturated during the absorption process. From the above results, H2O2 and free radicals like HO played leading roles in the oxidation of Hg0 process. In summary, electrochemical technology can be a promising method for the Hg0 removal and worth further investigation. Therefore, higher Hg0 removal efficiency can be achieved with increased concentration of H2O2 and HO as the probabilities of Hg0 to contact with H2O2 and free radicals were also increased. In conclusion, the mechanism of Hg0 removal can be explained from two aspects. First, H2O2 was electro-generated through two-electron reduction of oxygen on the interface of GDE. Second, H2O2 was electro-activated and decomposed to free radicals like HO2, HO and O in the electrolyte or on the 2 interface of GDE. As shown in Eqs. (13) and (14), H2O2 was immediately electro-activated through electron transfer to form free radicals, therefore, Hg0 was removed in this electrochemical system [22].

177

Fig. 6. Hg0 removal performance in the absence and presence of O2. Condition: voltage, 4 V; gas flow rate, 50 mL/min; pH, 4; Na2SO4 concentration, 0.1 M; Hg0 Concentration, 350 lg/m3; reaction temperature, 10 °C.

4. Conclusion By introducing the electrochemical process without any catalyst, Hg0 from simulated flue gas was removed. The electrochemical removal of Hg0 was explored in an undivided electrolysis reactor with a Ti/IrO2 anode and a self-designed gas diffusion electrode as the cathode. On the interface of GDE, H2O2 was electrogenerated and then electro-activated to free radicals. Due to its porous structures along with hydrophobic characteristics, the interface of gas-electrolyte-electrode was largely increased, thus there’s more chance for Hg0 to react with free radicals. It can be drawn that the optimal conditions for Hg0 removal were voltage of 6 V, gas flow rate of 100 mL/min, pH of 11, Na2SO4 of 0.8 mol/L, temperature of 70 °C. Compared the experiments, the results confirmed that H2O2 and free radicals electro-activated from it were the principle oxidizing power of Hg0 removal. In other words, it suggested that the electrochemical process containing H2O2 and free radicals on the interface of GDE is a promising method for Hg0 removal from flue gas. Acknowledgement This research is based upon work supported by the National Natural Science Foundation of China (Project No. 21307032), the Postdoctoral Science Foundation (Project No. 2012M511062) and ‘‘the Fundamental Research Funds for the Central Universities.” Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the authors and do not necessarily reflect the view of the supporting organizations. References [1] US EPA. Clean Air Act (As Amended Through P.L. 108–201, February 24, 2004). Q:nCOMPnENVIR1nCLEANAIR.001. [2] E.G. Pacyna, J.M. Pacyna, K. Sundseth, J. Munthe, K. Kindbom, S. Wilson, F. Steenhuisen, P. Maxson, Global emission of mercury to the atmosphere from anthropogenic sources in 2005 and projections to 2020, Atmos. Environ. 44 (2010) 2487–2499. [3] S. Wang, L. Zhang, G. Li, Y. Wu, J. Hao, N. Pirrone, F. Sprovieri, M. Ancora, Mercury emission and speciation of coal-fired power plants in China, Atmos. Chem. Phys. 10 (2010) 1183–1192. [4] E. Pacyna, J. Pacyna, Global emission of mercury from anthropogenic sources in 1995, Water. Air. Poll. 137 (2002) 149–165. [5] Y. Ma, Z. Qu, H. Xu, W. Wang, N. Yan, Investigation on mercury removal method from flue gas in the presence of sulfur dioxide, J. Hazard. Mater. 279 (2014) 289–295. [6] H. Zeng, F. Jin, J. Guo, Removal of elemental mercury from coal combustion flue gas by chloride-impregnated activated carbon, Fuel 83 (2004) 143–146.

178

Y. Xu et al. / Chemical Engineering Journal 310 (2017) 170–178

[7] W. Xu, H. Wang, T. Zhu, J. Kuang, P. Jing, Mercury removal from coal combustion flue gas by modified fly ash, J. Environ. Sci. 25 (2013) 393–398. [8] Z. Qu, N. Yan, P. Liu, Y. Chi, J. Jia, Bromine chloride as an oxidant to improve elemental mercury removal from coal-fired flue gas, Environ. Sci. Technol. 43 (2009) 8610–8615. [9] 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. [10] Y. Guo, N. Yan, S. Yang, P. Liu, J. Wang, Z. Qu, J. Jia, Conversion of elemental mercury with a novel membrane catalytic system at low temperature, J. Hazard. Mater. 213–214 (2012) 62–70. [11] Y. Liu, J. Pan, Q. Wang, Removal of Hg0 from containing-SO2/NO flue gas by ultraviolet/H2O2 process in a novel photochemical reactor, AIChE J. 60 (2014) 2275–2285. [12] J. Korell, H. Seifert, H.R. Paur, S. Andersson, P. Bolin, Flue gas cleaning with the MercOx process, Chem. Eng. Technol. 26 (2003) 737–740. [13] N. Li, J. An, L. Zhou, T. Li, J. Li, C. Feng, X. Wang, A novel carbon black graphite hybrid air-cathode for efficient hydrogen peroxide production in bioelectrochemical systems, J. Power Sources 306 (2016) 495–502. [14] W.R.P. Barros, T. Ereno, A.C. Tavares, M.R.V. Lanza, In situ electrochemical generation of hydrogen peroxide in alkaline aqueous solution by using an unmodified gas diffusion electrode, ChemElectroChem 2 (2015) 714–719. [15] R.M. Reis, A.A.G.F. Beati, R.S. Rocha, M.H.M.T. Assumpção, M.C. Santos, R. Bertazzoli, M.R.V. Lanza, Use of gas diffusion electrode for the in situ generation of hydrogen peroxide in an electrochemical flow-by reactor, Ind. Eng. Chem. Res. 51 (2012) 649–654. [16] R.B. Valim, R.M. Reis, P.S. Castro, A.S. Lima, R.S. Rocha, M. Bertotti, M.R.V. Lanza, Electrogeneration of hydrogen peroxide in gas diffusion electrodes modified with tert-butyl-anthraquinone on carbon black support, Carbon 61 (2013) 236–244. [17] X. Yu, M. Zhou, G. Ren, L. Ma, A novel dual gas diffusion electrodes system for efficient hydrogen peroxide generation used in electro-Fenton, Chem. Eng. J. 263 (2015) 92–100. [18] P. Liang, M. Rivallin, S. Cerneaux, S. Lacour, E. Petit, M. Cretin, Coupling cathodic electro-fenton reaction to membrane filtration for AO7 dye degradation: a successful feasibility study, J. Membr. Sci. 510 (2016) 182–190. [19] F.V.E.D. Reis, V.S. Antonin, P. Hammer, M.C. Santos, P.H.C. Camargo, Carbonsupported TiO2-Au hybrids as catalysts for the electrogeneration of hydrogen peroxide: investigating the effect of TiO2 shape, J. Catal. 326 (2015) 100–106.

[20] J.M. Fontmorin, R.C. Burgos Castillo, W.Z. Tang, M. Sillanpää, Stability of 5,5dimethyl-1-pyrroline-N-oxide as a spin-trap for quantification of hydroxyl radicals in processes based on Fenton reaction, Water Res. 99 (2016) 24–32. [21] M. Pimentel, N. Oturan, M. Dezotti, M.A. Oturan, Phenol degradation by advanced electrochemical oxidation process electro-Fenton using a carbon felt cathode, Appl. Catal. B 83 (2008) 140–149. [22] H. Luo, C. Li, C. Wu, W. Zheng, X. Dong, Electrochemical degradation of phenol by in situ electro-generated and electro-activated hydrogen peroxide using an improved gas diffusion cathode, Electrochim. Acta 186 (2015) 486–493. [23] L. Zhou, M. Zhou, C. Zhang, Y. Jiang, Z. Bi, J. Yang, Electro-Fenton degradation of p-nitrophenol using the anodized graphite felts, Chem. Eng. J. 233 (2013) 185– 192. [24] A.J. Bard, R. Parsons, J. Jordan, Standard Potentials in Aqueous Solution, M. Dekker, 1985. [25] H. Wang, D.Z. Sun, Z.Y. Bian, Degradation mechanism of diethyl phthalate with electrogenerated hydroxyl radical on a Pd/C gas-diffusion electrode, J. Hazard. Mater. 180 (2010) 710–715. [26] S. Pan, J. Hu, S. Chen, N. Tang, Removal of Hg0 in wet FGD by transition metal/ H2O2, J. Chem. Phys. Res. 6 (2014) 517–523. [27] S.J. Freakley, M. Piccinini, J.K. Edwards, E.N. Ntainjua, J.A. Moulijn, G.J. Hutchings, Effect of reaction conditions on the direct synthesis of hydrogen peroxide with a AuPd/TiO2 catalyst in a flow reactor, ACS Catal. 3 (2013) 487– 501. [28] P. Jin, R. Chang, D. Liu, K. Zhao, L. Zhang, Y. Ouyang, Phenol degradation in an electrochemical system with TiO2/activated carbon fiber as electrode, J. Environ. Chem. Eng. 2 (2014) 1040–1047. [29] J. Chen, Y. Xia, Q. Dai, Electrochemical degradation of chloramphenicol with a novel Al doped PbO2 electrode: performance, kinetics and degradation mechanism, Electrochim. Acta 165 (2015) 277–287. [30] Y. Liu, Q. Wang, Removal of elemental mercury from flue gas by thermally activated ammonium persulfate in a bubble column reactor, Environ. Sci. Technol. 48 (2014) 12181–12189. [31] W. Zhang, G. Huang, W. Jia, H. Li, R. Zheng, Y. Zhou, Removal of phenol from synthetic waste water using Gemini micellar-enhanced ultrafiltration (GMEUF), J. Hazard. Mater. 235–236 (2012) 128–137. [32] K.J. Laidler, The development of the Arrhenius equation, J.Chem.Educ. 61 (1984) 494. [33] Z. Qiang, J.H. Chang, C.P. Huang, Electrochemical generation of hydrogen peroxide from dissolved oxygen in acidic solutions, Water Res. 36 (2002) 85–94.