Simultaneous removal of NO and SO2 from flue gas

Chemosphere 193 (2018) 1216e1225

Contents lists available at ScienceDirect

Chemosphere journal homepage: www.elsevier.com/locate/chemosphere

Simultaneous removal of NO and SO2 from flue gas by combined heat and Fe2þ activated aqueous persulfate solutions Yusuf G. Adewuyi*, Nana Y. Sakyi, M. Arif Khan Chemical, Biological and Bio Engineering Department, North Carolina Agricultural and Technical State University, Greensboro, NC, 27411, USA

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

Advanced oxidation process for the simultaneous removal of SO2 and NO was developed. The process involved combined temperature and Fe2þ activation of Na2S2O8. SO2 removal was almost 100% even at the low temperature of 30 C. NO removal was 79% at 70 C, in presence of both Fe2þ and SO2. Chemistry and reaction pathways for the combined NO/SO2 removal are proposed.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 9 May 2017 Received in revised form 13 November 2017 Accepted 17 November 2017 Available online 17 November 2017

The use of advanced oxidation processes (AOPs) to integrate flue gas treatments for SO2, NOx and Hg0 into a single process unit is rapidly gaining research attention. AOPs are processes that rely on the generation of mainly the hydroxyl radical. This work evaluates the effectiveness of the simultaneous removal of NO and SO2 from flue gas utilizing AOP induced by the combined heat and Fe2þ activation of aqueous persulfate, and elucidates the reaction pathways. The results indicated that both SO2 in the flue gas and Fe2þ in solution improved NO removal, while the SO2 is almost completely removed. Increased temperature led to increase in NO removal in the absence and presence of both Fe2þ and SO2, and in the absence of either SO2 or Fe2þ, but the enhanced NO removal due to the presence of SO2 alone dominated at all temperatures. The removal of NO increased from 77.5% at 30 C to 80.5% and 82.3% at 50 C and 70 C in the presence of SO2 alone, and from 35.3% to 62.7% and 81.2%, respectively, in the presence of Fe2þ alone. However, in the presence of both SO2 and Fe2þ, NO conversion is 46.2% at 30 C, increased only slightly to 48.2% at 50 C; but sharply increased to 78.7% at 70 C compared to 63.9% for persulfateonly activation. Results suggest NO removal in the presence of SO2 is equally effective by heat-only or heat-Fe2þ activation as the temperature increases. The results should be useful for future developments of advanced oxidation processes for flue gas treatments. © 2017 Elsevier Ltd. All rights reserved.

Handling Editor: X. Cao Keywords: Nitric oxide Sulfur dioxide Oxidation Activated persulfate Temperature Fe2þ

1. Introduction * Corresponding author. E-mail address: [email protected] (Y.G. Adewuyi). https://doi.org/10.1016/j.chemosphere.2017.11.086 0045-6535/© 2017 Elsevier Ltd. All rights reserved.

Nitrogen oxides (NOx, mainly NO2 and NO) and sulfur oxides (mainly SO2) released from the burning of fossil fuel cause

Y.G. Adewuyi et al. / Chemosphere 193 (2018) 1216e1225

considerable environmental and health problems (Adewuyi et al., 1999). Stringent environmental laws and regulations have driven recent advances in air emission control technologies (Baranski and Underwood, 2014). Wet flue gas desulfurization (WFGD) and selective catalytic reduction (SCR) or selective non-catalytic reduction (SNCR) are commercially used for the abatement of SO2 and NOx emissions, respectively (Sweeney and Liu, 2001). However, separate control of SO2 and NOx has resulted in high capital and operational costs, large space requirement and environmental footprint (Rezaei et al., 2015). From a system design and cost perspective, it is desirable to integrate air pollution control system for SO2, NOx and Hg0 into a single operational unit (Hutson et al., 2008; Zheng et al., 2014; Chiu et al., 2015; Liu and Adewuyi, 2016). Therefore, the use of wet scrubbing agents utilizing electrolyzed seawater (Yang et al., 2016; Han et al., 2017), cost-effective chemical oxidants and/or metal chelates (Adewuyi and Khan, 2015, 2016), and advanced oxidation processes involving photochemical methods (Jethani et al., 1990; Adewuyi et al., 1999; Adewuyi and Owusu, 2003; Wu et al., 2008; Liu and Zhang, 2011; Liu et al., 2012; Shahrestani et al., 2017), are emerging for the simultaneous multicomponent treatment of flue gases. Adewuyi et al. reported the use of aqueous sodium persulfate (Na2S2O8) activated by heat for the NO only and simultaneous removal of NO and SO2; and Na2S2O8 activated by combined heat and Fe2þ for NO removal (Khan and Adewuyi, 2010; Adewuyi and Sakyi, 2013a, b; Adewuyi et al., 2014). The persulfate (PS) or peroxydisulfate anion (S2O2 8 ) is a strong oxidant, kinetically slow at ordinary conditions but highly reactive when activated by low pH (10). Wang et al. (2016) also studied NO removal using a dual oxidant (H2 O2 =S2 O2 8 ) at 50 C in a solution of 0.1 M Na2S2O8, 1% H2O2 and pH of 11. When PS is activated by heat, light, ultrasound, transition metal ions or alkaline pH, sulfate radicals ( ) are generated, which are subsequently responsible for the production of hydroxyl ( ) radicals as in Eqs. (1)e(4), resulting in faster kinetic rates of PS with substrates (House, 1962; Wardman, 1989; Adewuyi, 2005a, b).

o 2 S2 O2 8 þ 2e /2SO4 ; E ¼ 2:01 V

(1)

(2) (3) (4) For pH < 3, the increased rate of PS disappearance is by a slightly different overall reactions as in Eq. (5) and is due to acid-catalyzed asymmetric fission of PS (Eq. (6)) followed by hydrolysis of the radical (Eq. (7)), leading to the formation of Caro's acid (H2SO5) (Johnson et al., 2008). 2 S2 O2 8 þ H2 O /H2 SO5 þ SO4

1217

generally induced by alkaline conditions as in Eqs. (8)e(10) (Furman et al., 2010): OH

2 þ S2 O2 ! HO 8 þ 2H2 O 2 þ 2SO4 þ 3H

(8)

(9) (10)

The PS anion, OH and serve as oxidants for NOx and S(IV). In addition, other side reactions between the bisulfite ion (HSO 3 ) and NO2(aq), nitrite (NO2 ) or nitrous acid (HNO2) in the pH range of 3.0e8.0, result in the formation of N-S intermediates such as hydroxylaminedisulfonate (HON(SO3)2 2 or HADS) and nitrososulfonic acid (ONSO 3 or NSS) as shown in Eqs. (11)e(14), which lead to enhanced NO removal (Littlejohn et al., 1986; Siddiqi et al., 2001; Ajdari et al., 2015, 2016). 2 þ 2 NO2ðaqÞ þ HSO 3 þ H2 O/2 NO2 þ SO4 þ 3 H

(11)

2 þ NO 2 þ H þ 2 HSO3 4HONðSO3 Þ2 þ H2 O

(12)

þ NO 2 þ H þ HSO3 4ONSO3 þ H2 O

(13)

2 ONSO 3 þ HSO3 /HONðSO3 Þ2

(14)

In 2013, Adewuyi and Sakyi investigated the removal of NO by temperature-only activated aqueous PS and showed that NO removal was greatly enhanced by SO2, while the SO2 itself was completely removed (Adewuyi and Sakyi, 2013b). They also studied the removal of NO in the presence of Fe2þ but absence of SO2 using Na2S2O8 (0.1 M) solution simultaneously activated by temperature and 0.01 M Fe2þ and observed further increase in NO removal by almost 10% at all temperatures (23e90 C) (Adewuyi and Sakyi, 2013a; Adewuyi et al., 2014), compared to a prior study involving temperature-only activation at the same conditions (Khan and Adewuyi, 2010). It is also well-known that Fe2þ ions promotes interactions between NOx and S(IV) in aqueous solutions that result in the formation of nitrosyl complexes (Chang, 1986; Tursi c et al., 2001). However, to the best of our knowledge, the simultaneous removal of NO and SO2 from flue gas by PS activated by combined temperature and Fe2þ has never been specifically studied. The goal of this study was to evaluate the simultaneous NO and SO2 removal efficiencies, and elucidate the chemistry and reaction pathways of 2þ the NOx-S(IV)-S2O2system. 8 -Fe

2. Experimental 2.1. Materials

(5)

(6) (7) On the other hand, in basic solutions with 1.0e3.0 M NaOH, complex activation reactions have been reported, leading to the formation of transient oxygen species, hydroperoxide (HO 2 ) and ) (Buxton et al., 1988); in addition to the intersuperoxide ( conversion of to resulting from the narrow gap in redox potential (2.6 vs 2.7) between these radicals, a condition

Sodium persulfate (powder, >98%), iron (II) sulfate heptahydrate (FeSO4.7H2O, >99%), concentrated sulfuric acid (95e98%) and 1.0 N sodium hydroxide solution were obtained from Acros Organics, Morris Plains, NJ, USA. Ultrahigh purity, 5.0-grade nitrogen (N2) and NO/SO2 mixtures in ultrapure (4.8-grade) N2 were obtained from Airgas National Welders, Charlotte, NC, USA. Deionized water was obtained from a Milli-Q Advantage A 10 purifier with Elix 5 system from Millipore Corporation, Bedford, MA, USA. The water had a resistivity of at least 18.2 MU.cm, and the total organic contents, silicates and heavy metals contents were reduced to very low parts per billion (ppb) levels.

1218


2.2. Methods The schematic diagram for the NO absorption consisted of a thermally jacketed bubble column reactor made of Pyrex glass (5.1 cm i.d. 61-cm length; designed in-house and custom-built by Ace Glass, Inc., Vineland, NJ, USA), a flue gas blending system consisting of a Dynablender mass flow controller (Matheson Trigas, Montgomeryville, PA, USA) with two flow transducer calibrated to allow a maximum flow of 5 standard liters per minute (SLPM) gas, and an analytical train of Fourier Transform Infrared (FTIR) spectrometer (Tensor 27; Bruker Optics, Billerica, MA, USA). The scrubber was operated in a semibatch mode with the simulated gas flowing upward continuously at a rate of 0.1 SLPM. The jacketed column (with its cooling/heating temperature maintained by circulating water controlled by refrigerated bath, Neslab RTE7D1, Thermo Scientific, Newington, NH, USA) was initially filled with 750 ml of water and N2 was passed through it for 15 min to purge it of dissolved oxygen. Freshly prepared Fe2þ and PS solutions of appropriate quantity to make 1.0 ± 0.05 l total volume of solution (corresponding to a liquid height of ~0.5 m) of desired concentration at the corresponding temperature were added to the water consecutively at the top of the column (previously determined to be the best injection method from a previous study (Adewuyi and Sakyi, 2013a)), and gas bubbling started immediately. The experimental setup is shown in Fig. 1 and is discussed in detail elsewhere (Adewuyi and Owusu, 2003; Khan and Adewuyi, 2010; Adewuyi and Sakyi, 2013a; Adewuyi and Khan, 2016). NO, NO2 and other NOx species’ and SO2 concentrations in the inlet gas were initially measured and the outlet gas continuously monitored and recorded by proprietary software, Enformatic FTIR Collection Manager (EFCM) from FTIR.com. The solution pH was monitored continuously by a pH semi micro tip probe (Orion 8103 BN) mounted to the side of the column, and the initial and final pH were also recorded using Accumet pH meter 50 (Thermo Scientific). Fe2þ species is readily soluble in wide ranges of pH (2.0e9.0) while Fe3þ precipitates in the form of ferric oxyhydroxides at higher pH (4.0), and therefore, acidification to pH of about 3.0 improves iron availability and activation. For Fe2þ activation experiments, the deionized water in the column was first adjusted to pH between 3.0 and 3.5 (±0.1) with 5.0 N sulfuric acid. The volume of the auxiliary chemical added for pH adjustment was small in comparison to the total volume of reaction mixture but was enough to maintain both the initial and final reaction pH in the range of 2.5e3.0 and minimized the possibility of Na2S2O8 oxidation of Fe2þ to Fe3þ. The concentrations of Fe2þ, total Fe (i.e., Fe2þ þ Fe3þ), and Fe3þ (by difference) were determined using a modified 1,10-phenanthroline colorimetric method, using Beckman DU-7500 spectrophotometer (Beckman Coulter Inc., Fullerton, CA, USA) (Adewuyi et al., 2014). The details of the analytical procedures, including iron speciation (Fe2þ and Fe3þ), analysis of gas-phase NOx and anions (showing oxidation products of mainly NO 3 and not NO2 ), and material balances in solution, are reported in previous studies (Khan and Adewuyi, 2010, 2011; Adewuyi et al., 2014).

respectively. The NO concentration-time profiles show the usual characteristic involving initial sharp drop in NO exit concentration due to absorption in unsaturated solutions, and mixing and dilution effects of the purge gas, previously observed and explained (Adewuyi et al., 1999; Adewuyi and Owusu, 2003; Khan and Adewuyi, 2010). [NO]out was calculated by averaging the values of the outlet concentration after 3000 s. 3. Results and discussion 3.1. Proposed reaction pathways The use of advanced oxidation processes (AOPs) involving aqueous-phase activated persulfate for the removal of SO2, NOx and Hg0 from flue gas is rapidly gaining attention because the PS anion, OH and serve as effective oxidants for these species (Liu and Adewuyi, 2016). Incidentally, it is well known that the effective oxidation of SO2/NOx to H2SO4/HNO3 in the atmosphere both also involve the hydroxyl radical OH, Eo ¼ 2.8 V) (Buxton et al., 1988; Seinfeld and Pandis, 2006). The hydroxyl radical has also been shown to be the main oxidant in the sonochemical oxidation of SO2/NOx in aqueous solutions (Adewuyi and Owusu, 2006; Owusu and Adewuyi, 2006; Adewuyi and Khan, 2012). The thermal and Fe2þ ion activation , which is responsible for of S2O2 8 leads to the production of the production of OH that acts as the main oxidant for the conversion of both NO and SO2 (Adewuyi and Sakyi, 2013b; Adewuyi et al., 2014). Hence, the simultaneous removal of the NO and SO2 in aqueous solutions of PS is also thought to be dependent on the consumption of the dissolved NO and SO2 by the various reactive species generated by PS activation in the liquid phase, which maintains the driving force needed for the absorption process. The detailed proposed pathways for the 2þ system are depicted in a reaction scheme NOx-S(IV)-S2O28 -Fe illustrated in Fig. 2. As shown in Fig. 2, the chemistry of simultaneous consumption of NO and SO2 by PS jointly activated by temperature and Fe2þ is complex, since it takes place through direct electron transfer from the PS anion, free radical oxidation reactions via and , and the direct reaction of the NOx species with S(IV) and Fe2þ species. The main reactions involved in the simultaneous oxidation of NOx and SO2 are:

(16) (17) (18) (19) (20)

2.3. Data processing method

(21)

The inlet and outlet NO concentrations and time profiles were obtained from the FTIR's EFCM recorded data, and used to calculate the fractional percent conversion from,

XNO ¼

½NOin ½NOout 100% ½NOin

(22) (23)

(15)

where [NO]in and [NO]out are the steady state concentration in parts per million (ppm) of NO at the inlet and outlet of the reactor,

K24

þ SO2 þ H2 O ! HSO 3 þ H

(24)


1219

Fig. 1. Schematic diagram of the experimental set-up.

Reagents (From Top Of The Column)

[ Radicals [ Inter/reCombination

S(IV) Species and Interaction

[

S-species Reaction Paths N-species Reaction Paths Fe-species Reaction Paths [ Reaction Paths

Recycling

[

[ Direct Absorption by

Acidic Condition [ [ O

Activation

Generation

[

Heat

Alkaline Condition [

[

[ (Very High pH)

[ Absorption Absorption Inlet Flue Gas Fig. 2. Schematic diagram of the NOx-S(IV)-PS-Fe2þ reacting system.

End Products

1220


K25

þ HSO ! SO2 3 3 þ H

(25)

mixture of NO (1000 ppm) and SO2 (1550 ppm) in aqueous solutions of 0.1 M Na2S2O8 and 0.01 M Fe2þ are discussed hereafter.

(26) (27) (28) (29) (30) Also, the Caro's acid formed via reactions (5) and (7) with the peroxymonosulfate (PMS or oxone) anion (HSO 5 ) as the oxidizing species can generate H2O2 according to the reaction (31) (Adewuyi and Owusu, 2003; Wu et al., 2016).

H2 SO5 þ H2 O / H2 O2 þ H2 SO4

(31)

Although free radicals are not produced according to these reactions at the low pH, the oxidation ability of the solution could increase due to increased presence of the oxidative species, HSO 5 and H2 O2 , and promote both denitrification and desulfurization reactions (Khan and Adewuyi, 2010; Wu et al., 2016). In a prior work, we demonstrated that HSO 5 is a powerful oxidant for both NOx and S(IV), especially in the pH range of 6.5e8.5 via the overall reaction in Eq. (32) (Adewuyi and Owusu, 2003). 2NO þ 4HSO 5 þ HSO3 þ H2 O/2HNO3 þ 5HSO4

(32)

3.2. Simultaneous removal of NO and SO2 by Fe2þ activated Na2S2O8 The oxidation of NOx and S(IV) occurs mainly via the radicals generated following the decomposition of PS (Eqs. (16)e(18)). The and NO (Eq. (19)) is at almost diffusion reaction between

controlled rate (k19 ¼ 2 1010 M1 s1 ) and a factor of 100 or with Fe2þ faster than the reaction between 8 in Eq. (29) or Eq. (30) (k29 ¼ 3:2 10 M1 s1 or k30 ¼ 3:0 108 M 1 s1 ). The possibility of reaction 29 and 30 at the beginning of the experiments is also negligible as the gas was started bubbling immediately after the addition of Fe2þ and and PS solution at the respective temperatures. Therefore, generated are more likely to react directly with NOx (Eqs. (19)e(23)) and S(IV) (Eqs. (26)e(28)) species (Adewuyi and Sakyi, 2013b; Adewuyi et al., 2014). As shown in Fig. 3 using a gas mixture of about NO (500 ppm) and SO2 (1800 ppm) in a solution of 0.1 M PS and 0.01 M Fe2þ, SO2 is almost completely removed even at the low temperature of 30 C (even with this feed ratio of over 3:1 for SO2 and NO), while 40.7% NO is simultaneously removed at the same conditions. It should be noted that, although all the NOx species were continuously monitored, only presence of NO was noticeable in the effluent consistent with our prior observations (Khan and Adewuyi, 2010; Adewuyi and Sakyi, 2013b). In our previous work using heat-only activated 0.1 M PS for the removal of NO at various temperatures, NO conversion difference was marginal between using NO feed of 500 and 1000 ppm (Khan and Adewuyi, 2010). Therefore, the results of the absorption-oxidation of NO in the presence of SO2 induced by the temperature and/or Fe2þ activation of PS in the temperature range of 30e70 C, using a gas

3.3. Effect of temperature on NO removal in the presence of Fe2þ and/or SO2 The effects of temperature on NO concentration profiles using 0.1 M PS only and in the presence and absence of 0.01 M Fe2þ and/ or SO2 are illustrated in Fig. 4aec. Fig. 4a shows the NO exit gas concentration as a function of time at 30 C. As shown in this figure, NO conversion in the presence of SO2 gas using PS solution alone is 77.5%, but decreases significantly to 46.2% with Fe2þ in solution with PS, while in the absence of SO2, the NO conversion is only 35.3% with both PS and Fe2þ in solution but only 19.0% in presence of PS alone. It should be noted that, compared with the results in Fig. 3, NO removal of 46.2% with Fe2þ in solution is higher compared to 40.7%. This is likely due to the higher competition of the SO2 for available oxidants due to the higher SO2/NO feed ratio as noted previously. As shown in Fig. 4b, with temperature at 50 C, it can be observed that NO conversion in the presence of SO2 gas using aqueous solution of PS alone is 80.5%, but decreases drastically to 48.2% with Fe2þ in solution with PS, while in the absence of SO2, the NO conversion is enhanced to 62.7% with both PS and Fe2þ in solution and 53.5% in presence of PS alone. The results at 70 C shown in Fig. 4c follow similar trends to those at 50 C except the changes with temperature are not as significant. The NO conversion in the presence of SO2 gas using aqueous solution of PS alone is 82.7%, but decreases to 78.7% with Fe2þ in solution with PS, while in the absence of SO2, the NO conversion is enhanced to 81.2% with both PS and Fe2þ in solution and 63.9% in presence of PS alone. It appears that at the higher temperature, the difference between the presence of SO2 gas on NO conversion due to temperature activated PS is marginal compared with the NO conversion in the absence of SO2 resulting from combined temperature and Fe2þ activation (82.7% vs. 81.2%). The effect of temperature on the steady state conversion of NO (at time 3000 s) using 0.1 M PS only and in the presence and absence of 0.01 M Fe2þ and/or SO2 is illustrated in Fig. 5, and can be summarized as follows. In the presence of SO2 alone, the steadystate NO conversion increased from 77.5% at 30 C to 80.5% at 50 C and 82.7% at 70 C, while in the presence of Fe2þ alone, the results for the respective temperatures were 35.3%, 62.7% and 81.2%, respectively. However, in the presence of both SO2 and Fe2þ, NO conversion is 46.2% at 30 C, increased only slightly to 48.2% at 50 C, which is lower than the result of PS-only activation of 53.5%, but increased sharply to 78.7% at 70 C, compared to 63.9% for PSonly activation. In general, the results indicate that both SO2 in the flue gas and Fe2þ in solution improve NO removal, while the SO2 is completely removed. Increased temperature led to increased conversions of NO in the presence and absence of either SO2 or Fe2þ alone or in combination but the enhanced NO removal due to the presence of SO2 alone dominated at all temperatures. Also, the results suggest that NO removal in the presence of SO2 is equally effective by temperature-only or the combined temperature-Fe2þ activation as the temperature increases. As observed in Fig. 4a, in the absence of SO2 and Fe2þ in solution, and at lower temperature of 30 C, the initial drop in NO concentration (i.e. a high absorption rate) could not be maintained and the concentration returns to a higher value and remained steady for the rest of the experimental duration. At the lower temperatures, the PS activation alone is not as pronounced enough to provide adequate oxidizing radicals and in the absence of SO2, the enhanced NO absorption effect due to the side reactions of SO2 with the NOx


Fig. 3. Concentration profile for the simultaneous removal of NO and SO2 at 30 C for 0.1 M Na2S2O8 solution in the presence of 0.01 M Fe2þ.

species is missing. The stoichiometry and the rates for the reactions 2 between S(IV) ions (HSO 3 and SO3 ) and NO2 in aqueous solutions have been studied by a number of investigators (Littlejohn et al., 1986; Clifton et al., 1988). The rate constant at ambient temperature is significant and increases from about 1.2 107 M1s1 near pH 5 (in 1.0 mM phosphate buffer) to 2.9 107 M1s1 at pH 13 (in 0.1 M KOH) (Clifton et al., 1988). Also, the presence of SO2 gas induces faster NO gas absorption (Adewuyi and Sakyi, 2013b). These could be due to the fact that, in the presence of SO2, there are also important side reactions between bisulfite ion (HSO 3 ) and NO2(aq), nitrite (NO 2 ) or nitrous acid (HNO2) (Eqs. (11)e(14)) to produce N-S intermediates such as hydroxylamine disulfonate (HON(SO3)2 2 or HADS) and nitrososulfonic acid (ONSO 3 or NSS), which have been identified by a number of investigators using various analytical methods including UV spectrophotometry, ion chromatograph, Raman spectroscopy and electrophoresis (capillary ion analysis) (Littlejohn and Chang, 1984, 1986, 1994). As observed by Littlejohn et al. (1986), the main loss mechanism for nitrite ion in flue gas desulfurization systems is the reaction with bisulfite to form N-S compounds. Also, Fe2þ ions and its complexes promote the inter2 action between NO or HNO2 and HSO 3 /SO3 via nitrosyl complexes, leading to the formation of HADS, sulfamate (H2N(SO3)-), N2O and 2þ HSO to Fe3þ. 4 , and is accompanied by the oxidation of Fe Completion of the catalytic cycle occurs spontaneously in the 2 presence of HSO sic et al., 2001). N2O usually forms in 3 /SO3 (Tur strong acid solutions, but which in O2-deficient solutions reacts with H2O to form N2 and HNO2 (that is subsequently oxidized to HNO3) (Chang et al., 1983). The triple negative effects of the absence of SO2, lack of additional Fe2þ activation of PS and low temperature are likely to result in the inability of the PS scrubbing solution to sustain higher absorption capacities for long and the low NO steady state conversion of 19.0% observed at 30 C. However, in the presence of Fe2þ, the NO conversion increases to 35.3% possibly due to the combined temperature/Fe2þ activation of PS. It can also be observed from Fig. 5 that at 30 C, the presence of SO2 enhances NO absorption in the presence or absence of Fe2þ, but that the SO2 enhancement effect in the absence of Fe2þ is more significant at the lower temperature. However, as seen in Fig. 4b, the dips representing initial absorption rates, obvious in Fig. 4a, become less noticeable with an increase in the temperature to 50 C and disappear at the higher temperature of 70 C as shown in Fig. 4c, indicating that increasing the temperature of the system at the same PS and Fe2þ concentration levels led to an increased conversion of NO. The rate constants of PS decomposition have been reported to be 1.0 107 s1 and 5.7 105 s1 (measured for pH 1.3) at 25 C and 70 C, respectively, suggesting that the rate constant at 70 C is 570 times greater than that obtained at 25 C (House, 1962). The activation energy for the PS thermolysis has been reported to be

1221

33.5 kcal mol1 (18 kcal mol1, in presence of Fe2þ), indicating that decomposition of PS is much faster at high temperature than at low temperature (Kolthoff and Miller, 1951; Aher et al., 2017). This means that the thermal activation simply enhances the rate of generation of reactive radicals at the same PS concentration. Therefore, it is expected that elevated temperature could promote the decomposition of PS to form more sulfate radicals, resulting in an increased removal of NO. It should be noted that high temperature also reduces the physical solubility of NO, which is unfavorable for gas-liquid mass transfer. On the other hand, the mass transfer coefficient (KLa) increases with temperature due to the acceleration of molecular diffusion, and the driving force provided by chemical reaction in a reactive gas-liquid process. In a recent study, Adewuyi et al. (2014) determined that chemical reaction plays the dominant role in the temperature range of this study. As can be observed from Fig. 4c, it appears the enhancement effects of the presence of SO2 and simultaneous activation effect of Fe2þ on NO conversion are about the same, resulting in 5e9% over the case of the presence of PS alone at the highest temperature, 70 C of this study. 3.4. Effects of excess iron and discussion on the presence of O2 and CO2 It has also been reported that NO bounds to FeSO4 in solution according to reaction (33) with an equilibrium or a stability constant, K33 ¼ k1/k-1 ¼ 450 M1, where the forward (k1) and backward (k-1) rate constants are 6.2 105 M1s1 and 1.4 103 s1, respectively at 25 C (Kustin et al., 1966). K33

FeSO4 þ NO ! FeðNOÞSO4

(33)

It is conceivable that the effect of this reaction could be significant to the extent it ties up Fe2þ in solution needed for the activation of PS. However, in Fe2þ-activated PS system, it is expected that all the Fe2þ reacts according to reaction (17) and remains virtually non-existent in the solution and the possibility of reaction (33) is negligible to nil. In order to test this possibility, and the Fe2þ concentration to minimize reaction (33), we performed material balance for iron for different initial concentration of Fe2þ (0.005e0.02 M) with initial PS concentration of 0.1 M at 50 C. The results are shown in Fig. 6aed and Table 1, depicting the various iron species in solution. The total Fe is the sum of Fe2þ, Fe3þ and non-labile Fe. The non-labile Fe is the difference between initial iron concentration and total measured iron, and consists of iron hydroxide and other non-labile iron species present in very small amounts. Fig. 6a shows Fe2þ concentration with time, and with the same figure enlarged (Fig. 6b), Fe2þ remain close to zero throughout the experiments. As indicated in Table 1, steady state average Fe2þ concentrations are about 1.0 105 M, 1 104 M and 1.7 104 respectively, for initial Fe2þ concentrations of 0.005, 0.01 and 0.02 M. For 0.005 M initial Fe2þ concentration, almost 100% of the Fe2þ seemed to be converted at the beginning of the experiment resulting from PS activation according to reaction (17). For 0.01 M initial Fe2þ, 99% of the Fe2þ is converted to Fe3þ (92%) and other non-labile species’ (7%). When initial Fe2þ increased from 0.005 to 0.01 M, non-labile formation does not seem to increase significantly, but after 0.01 M, it is more than doubled in quantity (4.4 104 M to 1.1 103 M), suggesting the detrimental effect of excessive Fe2þ on PS activation (Adewuyi and Sakyi, 2013a). Therefore, with 0.01 Fe2þ, reaction (33) does not appear to be significant. In addition to SO2, the effects of other flue gas constituents such as O2 and CO2, which constitute 2e15% and 10e12% of the fuel gas system respectively, may be important. However, using aqueous

1222


NO Concentra on (ppm)

1200

Only Persulfate, only NO

At 30 °C (a)

1000 800

18.96%

600

35.33% 46.21%

400 200

77.54%

0 0


1200

1000

2000 Time (s)

At 50 °C (b)

3000

4000

Only Persulfate, only NO

1000 800 600

48.19% 53.46% 62.67%

400 200

80.50%

0 0

1200

1000

2000 Time (s)

3000

4000

At 70 °C (c)


1000 Only Persulfate, only NO 800 600 400

63.86%

200

78.71% 81.19% 82.67%

0 0

1000

2000 Time (s)

3000

4000

Fig. 4. NO concentration profiles for 0.1 M Na2S2O8 solution in the presence of 0.01 M Fe2þ with and without SO2 in the flue gas at (a) 30 C, (b) 50 C and (c) 70 C.

PMS for treating a flue gas containing NO and SO2, we previously showed that the presence of 3e10% O2 increased NO conversion by only 3% (maximum) compared with the base case without O2, and 5e10% CO2 had no significant effects on NO removal (Adewuyi and

Owusu, 2003). Also, a recent study of the NO removal from flue gas in the presence of O2 using heat-activated 0.1 M aqueous ammonium persulfate solutions at 50 C and pH ¼ 2.3, Wu et al. (2016) found that NO conversion was nearly stable with O2


1223

90 80 NO Conversion (%)

70 60 50 40 Only Persulfate, only NO

30 20 10 0 20

30

40

50 60 Temperature (°C)

70

80

Fig. 5. Dependence of NO fractional conversion on temperature for 0.1 M Na2S2O8 solution in the presence of 0.01 M Fe2þ with and without SO2 in the flue gas.

concentration in the range of 2e7%, i.e., an increase of only 2% (conversion of 39e41%) with O2 of 2e5%, and remaining fairly constant thereafter (up to 7% O2). They speculated that the role of

5E-04

(a)

(b)

4E-04

1.5E-02

Fe2+ Conc. (M)

Fe2+ Conc. (M)

2.0E-02

the O2 was similar to that of inert gas N2 under the given acidic condition. Therefore, we deemed it unnecessary to evaluate the effects of both O2 and CO2 in this study.

0.02 M Initial Fe 1.0E-02 0.01 M Initial Fe 0.005 M Initial Fe

5.0E-03

0.02 M Initial Fe

3E-04

0.01 M Initial Fe 0.005 M Initial Fe

2E-04 1E-04 0E+00

0.0E+00 0

1000

2000

3000

0

4000

1000

(c)

Non-labile Fe Conc. (M)

Fe3+ Conc. (M)


1.0E-02

5.0E-03

4000

(d)

3E-03

0.02 M Initial Fe

1.5E-02

3000

Time (s)

Time (s)

2.0E-02

2000

0.02 M Initial Fe 2E-03


1E-03

0E+00

0.0E+00 0

1000

2000

Time (s)

3000

4000

0

1000

2000

3000

4000

Time (s)

Fig. 6. Speciation plots for different iron components for different initial Fe2þ concentration (0.005e0.02 M) with initial persulfate concentration of 0.1 M at 50 C: (a) Fe2þ, (b) Enlarged Fe2þ, (c) Fe3þ, (d) Non-labile Fe profiles.

1224


Table 1 Concentration of different iron species’ for initial Fe2þ concentration of 0.005e0.02 M with initial persulfate concentration 0.1 M at 50 C. Iron Species'

Concentration (M) Initial Fe2þ Conc.

Fe

2þ

Fe3þ

Total Measured Fe

Non-labile Fe

Maximum Minimum Steady State Maximum Minimum Steady State Maximum Minimum Steady State Maximum Minimum Steady State

Avg.

Avg.

Avg.

Avg.

4. Conclusion The use of advanced oxidation processes (AOPs) for multi-pollutants’ reductions from flue gas is rapidly gaining worldwide attention as alternatives to using a different technology for each specific pollutant (e.g., NOx, SO2 or Hg). An AOP involving the combined heat and Fe2þ activation of Na2S2O8 for the simultaneous removal of NO and SO2 has been studied. The results indicated that both SO2 in the flue gas and Fe2þ in solution improved NO removal, while the SO2 is almost completely removed. Increased temperature led to increased conversions of NO (percent inlet NO removed) in the presence and absence of either SO2 or Fe2þ alone or in combination but the enhanced NO removal due to the presence of SO2 alone dominated at all temperatures. In the presence of SO2 alone, the steady-state NO conversion increased from 77.5% at 30 C to 80.5% at 50 C and 82.3% at 70 C, while in the presence of Fe2þ alone, the results in the respective temperatures were 35.3%, 62.7% and 81.2%. However, in the presence of both SO2 and Fe2þ, NO conversion is 46.2% at 30 C, increased only slightly to 48.2 at 50 C, which is lower even compared to PS alone activation (53.5% at 50 C), but increased sharply to 81.2% at 70 C, compared to 63.9% for PS activation alone. The results suggest that simultaneous removal of NO and SO2 by aqueous PS is equally effective by combinative temperature and Fe2þ activation or temperature alone activation as the temperature increases. However, the current study also indicates a number of areas in which additional research is needed, such as the determination of intermediates/products for comprehensive material balance, which is beyond the scope of current investigation but the subject of a future study. Acknowledgments The authors wish to acknowledge the contribution of National Science Foundation (NSF) for the funding received via Grant CBET0651811. References Adewuyi, Y., He, X., Shaw, H., Lolertpihop, W., 1999. Simultaneous absorption and oxidation of NO and SO2 by aqueous solutions of sodium chlorite. Chem. Eng. Commun. 174, 21e51. Adewuyi, Y.G., 2005a. Sonochemistry in environmental remediation. 1. Combinative and hybrid sonophotochemical oxidation processes for the treatment of pollutants in water. Environ. Sci. Technol. 39, 3409e3420. Adewuyi, Y.G., 2005b. Sonochemistry in environmental remediation. 2. Heterogeneous sonophotocatalytic oxidation processes for the treatment of pollutants in water. Environ. Sci. Technol. 39, 8557e8570. Adewuyi, Y.G., Khan, M.A., 2015. Nitric oxide removal by combined persulfate and ferrouseEDTA reaction systems. Chem. Eng. J. 281, 575e587. Adewuyi, Y.G., Khan, M.A., 2016. Nitric oxide removal from flue gas by combined persulfate and ferrouseEDTA solutions: effects of persulfate and EDTA

5.00 103

1.00 102

5

4

1.34 0 8.27 4.94 4.57 4.62 4.95 4.58 4.63 4.17 5.10 3.69

10

106 103 103 103 103 103 103 104 105 104

1.39 8.27 1.07 9.75 9.29 9.45 9.89 9.37 9.56 6.27 1.13 4.41

10 105 104 103 103 103 103 103 103 104 104 104

2.00 102 2.71 1.17 1.73 1.91 1.84 1.87 1.94 1.85 1.89 1.50 1.17 1.10

104 104 104 102 102 102 102 102 102 103 104 103

concentrations, temperature, pH and SO2. Chem. Eng. J. 304, 793e807. Adewuyi, Y.G., Khan, M.A., Sakyi, N.Y., 2014. Kinetics and modeling of the removal of nitric oxide by aqueous sodium persulfate simultaneously activated by temperature and Fe2þ. Ind. Eng. Chem. Res. 53, 828e839. Adewuyi, Y.G., Khan, N.E., 2012. Modeling the ultrasonic cavitation-enhanced removal of nitrogen oxide in a bubble column reactor. AlChE J. 58, 2397e2411. Adewuyi, Y.G., Owusu, S.O., 2003. Aqueous absorption and oxidation of nitric oxide with oxone for the treatment of tail gases: process feasibility, stoichiometry, reaction pathways, and absorption rate. Ind. Eng. Chem. Res. 42, 4084e4100. Adewuyi, Y.G., Owusu, S.O., 2006. Ultrasound-induced aqueous removal of nitric oxide from flue gases: effects of sulfur dioxide, chloride, and chemical oxidant. J. Phys. Chem. A 110, 11098e11107. Adewuyi, Y.G., Sakyi, N.Y., 2013a. Removal of nitric oxide by aqueous sodium persulfate simultaneously activated by temperature and Fe2þ in a lab-scale bubble reactor. Ind. Eng. Chem. Res. 52, 14687e14697. Adewuyi, Y.G., Sakyi, N.Y., 2013b. Simultaneous absorption and oxidation of nitric oxide and sulfur dioxide by aqueous solutions of sodium persulfate activated by temperature. Ind. Eng. Chem. Res. 52, 11702e11711. Aher, A., Papp, J., Colburn, A., Wan, H., Hatakeyama, E., Prakash, P., Weaver, B., Bhattacharyya, D., 2017. Naphthenic acids removal from high TDS produced water by persulfate mediated iron oxide functionalized catalytic membrane, and by nanofiltration. Chem. Eng. J. 327, 573e583. Ajdari, S., Normann, F., Andersson, K., Johnsson, F., 2015. Modeling the nitrogen and sulfur chemistry in pressurized flue gas systems. Ind. Eng. Chem. Res. 54, 1216e1227. Ajdari, S., Normann, F., Andersson, K., Johnsson, F., 2016. Reduced mechanism for nitrogen and sulfur chemistry in pressurized flue gas systems. Ind. Eng. Chem. Res. 55, 5514e5525. Baranski, K., Underwood, B., 2014. Mitigate air pollution with catalytic technology. Chem. Eng. Prog. 110, 30e34. Buxton, G.V., Greenstock, C.L., Helman, W.P., Ross, A.B., 1988. Critical review of rate constants for reactions of hydrated electrons, hydrogen atoms and hydroxyl radicals (,OH/,O) in aqueous solution. J. Phys. Chem. Ref. Data 17, 513e886. Chang, S.G., 1986. Technical analyses of a wet process for flue gas simultaneous desulfurization and denitrification. In: Markuszewski, R., Blaustein, B.D. (Eds.), Fossil Fuels Utilization. American Chemical Society, pp. 159e175. Chang, S.G., Littlejohn, D., Lynn, S., 1983. Effects of metal chelates on wet flue gas scrubbing chemistry. Environ. Sci. Technol. 17, 649e653. Chiu, C.-H., Hsi, H.-C., Lin, H.-P., 2015. Multipollutant control of Hg/SO2/NO from coal-combustion flue gases using transition metal oxide-impregnated SCR catalysts. Catal. Today 245, 2e9. Clifton, C.L., Altstein, N., Huie, R.E., 1988. Rate constant for the reaction of nitrogen dioxide with sulfur(IV) over the pH range 5.3-13. Environ. Sci. Technol. 22, 586e589. Furman, O.S., Teel, A.L., Watts, R.J., 2010. Mechanism of base activation of persulfate. Environ. Sci. Technol. 44, 6423e6428. Han, Z., Yang, S., Zhao, D., Liu, B., Pan, X., Yan, Z., 2017. An investigation of mass transfer-reaction kinetics of NO absorption by wet scrubbing using an electrolyzed seawater solution. RSC Adv. 7, 18821e18829. House, D.A., 1962. Kinetics and mechanism of oxidations by peroxydisulfate. Chem. Rev. 62, 185e203. Hutson, N.D., Krzyzynska, R., Srivastava, R.K., 2008. Simultaneous removal of SO2, NOX, and Hg from coal flue gas using a NaClO2-enhanced wet scrubber. Ind. Eng. Chem. Res. 47, 5825e5831. Jethani, K.R., Suchak, N.J., Joshi, J.B., 1990. Selection of reactive solvent for pollution abatement of NOx. Gas. Sep. Purif. 4, 8e28. Johnson, R.L., Tratnyek, P.G., Johnson, R.O.B., 2008. Persulfate persistence under thermal activation conditions. Environ. Sci. Technol. 42, 9350e9356. Khan, N.E., Adewuyi, Y.G., 2010. Absorption and oxidation of nitric oxide (NO) by aqueous solutions of sodium persulfate in a bubble column reactor. Ind. Eng. Chem. Res. 49, 8749e8760. Khan, N.E., Adewuyi, Y.G., 2011. A new method of analysis of peroxydisulfate using ion chromatography and its application to the simultaneous determination of

Y.G. Adewuyi et al. / Chemosphere 193 (2018) 1216e1225 peroxydisulfate and other common inorganic ions in a peroxydisulfate matrix. J. Chromatogr. A 1218, 392e397. Kolthoff, I.M., Miller, I.K., 1951. The chemistry of persulfate. I. The kinetics and mechanism of the decomposition of the persulfate ion in aqueous medium. J. Am. Chem. Soc. 73, 3055e3059. Kustin, K., Taub, I.A., Weinstock, E., 1966. A kinetic study of the formation of the ferrous-nitric oxide complex. Inorg. Chem. 5, 1079e1082. Littlejohn, D., Chang, S.G., 1984. Identification of species in a wet flue gas desulfurization and denitrification system by laser Raman spectroscopy. Environ. Sci. Technol. 18, 305e310. Littlejohn, D., Chang, S.G., 1986. Determination of nitrogen-sulfur compounds by ion chromatography. Anal. Chem. 58, 158e160. Littlejohn, D., Chang, S.G., 1994. Oxidative decomposition of nitrogen-sulfur oxides. Ind. Eng. Chem. Res. 33, 515e518. Littlejohn, D., Hu, K.Y., Chang, S.G., 1986. Kinetics of the reaction of nitric oxide with sulfite and bisulfite ions in aqueous solution. Inorg. Chem. 25, 3131e3135. Liu, Y., Adewuyi, Y.G., 2016. A review on removal of elemental mercury from flue gas using advanced oxidation process: chemistry and process. Chem. Eng. Res. Des. 112, 199e250. Liu, Y., Zhang, J., 2011. Photochemical oxidation removal of NO and SO2 from simulated flue gas of coal-fired power plants by wet scrubbing using UV/H2O2 advanced oxidation process. Ind. Eng. Chem. Res. 50, 3836e3841. Liu, Y., Zhang, J., Pan, J., Tang, A., 2012. Investigation on the removal of NO from SO2containing simulated flue gas by an ultraviolet/fenton-like reaction. Energy Fuels. 26, 5430e5436. Owusu, S.O., Adewuyi, Y.G., 2006. Sonochemical removal of nitric oxide from flue gases. Ind. Eng. Chem. Res. 45, 4475e4485. Rezaei, F., Rownaghi, A.A., Monjezi, S., Lively, R.P., Jones, C.W., 2015. SOx/NOx removal from flue gas streams by solid adsorbents: a review of current

1225

challenges and future directions. Energy Fuels. 29, 5467e5486. Seinfeld, J.H., Pandis, S.N., 2006. Atmospheric Chemistry and Physics: from Air Pollution to Climate Change, Second Ed. Wiley, New Jersey. Shahrestani, M.M., Rahimi, A., Momeni, M., 2017. Experimental study and mathematical modeling of NO removal using the UV/H2O2 advanced oxidation process. Chem. Eng. Technol. 40, 1149e1157. Siddiqi, M.A., Petersen, J., Lucas, K., 2001. A study of the effect of nitrogen dioxide on the absorption of sulfur dioxide in wet flue gas cleaning processes. Ind. Eng. Chem. Res. 40, 2116e2127. Sweeney, A.J., Liu, Y.A., 2001. Use of simulation to optimize NOx abatement by absorption and selective catalytic reduction. Ind. Eng. Chem. Res. 40, 2618e2627. Tursi c, J., Grgi c, I., Bizjak, M., 2001. Influence of NO2 and dissolved iron on the S(IV) oxidation in synthetic aqueous solution. Atmos. Environ. 35, 97e104. Wang, Z., Wang, Z., Ye, Y., Chen, N., Li, H., 2016. Study on the removal of nitric oxide (NO) by dual oxidant (H2O2/S2O2 8 ) system. Chem. Eng. Sci. 145, 133e140. Wardman, P., 1989. Reduction potentials of one-electron couples involving freeradicals in aqueous solution. J. Phys. Chem. Ref. Data 18, 1637e1755. Wu, B., Xiong, Y., Ru, J., Feng, H., 2016. Removal of NO from flue gas using heatactivated ammonium persulfate aqueous solution in a bubbling reactor. RSC Adv. 6, 33919e33930. Wu, Z.B., Wang, L., Zhao, W.R., 2008. Kinetic study on regeneration of FeIIEDTA in the wet process of NO removal. Chem. Eng. J. 140, 130e135. Yang, S., Han, Z., Pan, X., Yan, Z., Yu, J., 2016. Nitrogen oxide removal using seawater electrolysis in an undivided cell for ocean-going vessels. RSC Adv. 6, 114623e114631. Zheng, C., Xu, C., Zhang, Y., Zhang, J., Gao, X., Luo, Z., Cen, K., 2014. Nitrogen oxide absorption and nitrite/nitrate formation in limestone slurry for WFGD system. Appl. Energy 129, 187e194.