This is the accepted version of the journal article: Guimerà, Xavier; Mora Garrido, Mabel; Dorado, Antonio David; [et al.]. «Opti- mization of SO2 and NOx sequential wet absorption in a two-stage bioscrubber for elemental sulphur valorisation». Environmental Science and Pollution Research, Vol. 28, Issue 19 (May 2021), p. 24605-24617. DOI 10.1007/s11356-020-09607-1 This version is available at https://ddd.uab.cat/record/266907 under the terms of the license 1 Optimization of SO2 and NOx sequential wet absorption in a two-stage 2 bioscrubber for elemental sulphur valorisation 3 Xavier Guimeràa*, Mabel Morab, Antonio David Doradoa, Anna Bonsfillsa, David 4 Gabrielb, Xavier Gamisansa 5 a 6 Department of Mining Industrial and ICT Engineering, Universitat Politècnica de 7 Catalunya, Avinguda de les Bases de Manresa 61-73, 08240 Manresa, Spain. b 8 GENOCOV Research Group. Department of Chemical, Biological and 9 Environmental Engineering. Universitat Autònoma de Barcelona. Edifici Q. 08193 10 Bellaterra, Barcelona, Spain. * 11 Corresponding autor: Xavier Guimerà (xavier.guimera@upc.edu) 12 ABSTRACT 13 Flue gases contain SO2 and NOx that can be treated together for elemental sulfur 14 recovery in bioscrubbers, a technology that couples physical-chemical and biological 15 processes for gaseous emissions treatment in a more economic manner than classical 16 absorption. Sequential wet-absorption of SO2 and NOx from flue gas is thoroughly studied 17 in this work in a two-stage bioscrubber towards elemental sulphur valorisation pursuing 18 reuse of biological process effluents as absorbents. The optimal operating conditions 19 required for SO2 and NOx absorption in two consecutive spray-absorbers were defined 20 using NaOH-based absorbents. Overall, removal efficiencies of 98.9 and 55.9% for SO2 21 and NOx abatement were obtained in two in-series scrubbers operated under a gas contact 22 time of 1 and 100 s, and a liquid-to-gas ratio of 7.5 and 15 L·m-3, respectively. Higher 23 NOx removal efficiency to clean gas emission was obtained by oxidants dosing in the 24 absorber for NOx absorption. High NaHCO3 concentration in two-stage bioscrubber 25 effluent was exploited as alkaline absorbent for flue gas treatment. The performance of 26 scrubbers using an absorbent mimicking a reused effluent exhibited same removal 27 efficiencies than those observed using NaOH solutions. In addition, the reuse of 28 bioprocess effluent reduced reagents consumption by a 63.7%. Thus, the two-stage 29 bioscrubber proposed herein offers an environmentally friendly and economic alternative 30 for flue gas treatment. 31 KEYWORDS 32 Flue gas treatment, desulphurization, bioscrubber, SO2 valorisation, NOx 33 34 1. INTRODUCTION 35 Combustion processes give rise to one of the main air pollution sources from the 36 release of harmful chemicals such as CO2, SO2 and NOx to the atmosphere (Liu et al. 37 2010; Sun et al. 2016). Depending on the fuel and the characteristics of the combustion 38 facilities, concentrations of SO2 and NOx in off-gases range typically between 200 and 39 4000 ppmv, (Shen and Ayer 1975; Hu et al. 2000; Zevenhoven and Kilpinen 2001) and 40 200 and 1500 ppmv (Chen et al. 2007; Winnes and Fridell 2010; Tian et al. 2012), 41 respectively, which significantly exceeds the levels required by the global regulations. 42 Hence, reduction of flue gas emissions is in the public administration and industrial sector 43 agenda and their treatment has become critical to avoid health and welfare problems both 44 for living organisms and ecosystems (Aǧar et al. 2000; Bai and Meng 2005; Kan et al. 45 2010; Altuǧ et al. 2013; Mathieu et al. 2013). 46 Several technologies have been developed to remove SO2 and NOx from off-gases in 47 combustion processes. Wet and dry absorption processes are the traditional physical- 48 chemical desulphurization processes (Majeed et al. 1995; Wang et al. 2012), since SO2 49 presents a high solubility in aqueous solutions (11.29 g SO2·100 g-1 H2O (Perry and Green 50 1997)). On the other hand, NOx removal by scrubbing mechanisms is quite challenging 51 because NO, which is the major component of NOx comprising the 90-95% of the entire 52 emission (Adewuyi and Sakyi 2013), is sparingly soluble in aqueous solutions (0.00618 53 g NO·100 g-1 H2O (Perry and Green 1997)). For this reason, advanced treatment 54 technologies, such as selective catalytic reduction were developed for flue gas 55 denitrification (Zhu et al. 2004; Jin et al. 2005; Liu et al. 2010; Ding et al. 2014; Sun et 56 al. 2016). High removal efficiencies can be obtained for flue gas desulfurization and 57 denitrification combining these technologies (Adewuyi and Sakyi 2013; Ding et al. 2014; 58 Hao et al. 2017). However, the treatment of both acid gases through absorption and 59 catalytic technologies requires a high consumption of alkalis (Karatepe 2000) and 60 catalysts (Sun et al. 2016), which increases recurring operating costs and generates large 61 amounts of waste. Therefore, the development of novel low-cost, sustainable 62 technologies is essential. For this reason, flue gas treatment has been recently addressed 63 using biological techniques such as bioscrubbers (Lens et al. 2003; Lin et al. 2015; Sun 64 et al. 2018) and biotrickling filters (Philip and Deshusses 2003). Compared to physical- 65 chemical technologies, biological alternatives for flue gas treatment (bio-FGD) allowed 66 the recovery of pollutants as an innocuous or even a valuable product (Ottengraf and Diks 67 1992; An et al. 2010). 68 To this aim, a promising bioscrubbing process is presented herein to valorise pollutants 69 such as SO2 contained in flue gases as elemental sulfur. This novel bio-FGD alternative 70 is based on the sequential absorption of SO2 and NOx from flue gas in two consecutive 71 scrubbers, coupled to the biological treatment of sulphate/sulphite and nitrate/nitrite 72 contained in scrubbing liquors (Philip and Deshusses 2003; Wang et al. 2013; Sun et al. 73 2018). Biological treatment of the resulting sulphate/sulphite rich-streams is performed 74 through a two-steps bioprocess consisting of sulphate/sulphite biological reduction to H2S 75 followed by its partial oxidation to elemental sulphur under anoxic conditions using 76 nitrate/nitrite rich-stream as electron acceptor. The biological reduction of sulphate, 77 which is catalysed by sulphate reducing bacteria, has been studied over the past two 78 decades and have been developed using a different organics as carbon source (Rao et al. 79 2007; Mora et al. 2018) or an electron donor such as H2 (Van Houten et al. 1997; Lens et 80 al. 2003), and under a wide range of operating conditions. In addition, some alternatives 81 have coupled SO2 and NOx abatement from flue gases using absorbed NOx as electron 82 acceptor for autotrophic sulphide oxidation (Lee and Sublette 1991; Qian et al. 2015). 83 However, bio-FGD technologies still presents important limitations, mainly the large 84 consumption of alkali for SO2 absorption (Lens et al. 2003), but also the use of expensive 85 carbon sources for sulphate reduction (glucose, VFA or alcohols), low NOx removal due 86 to low NO solubility in water solutions (Adewuyi and Sakyi 2013) and reduced removal 87 efficiencies when treating SO2 loads above 0.6 kg S·m-3·d-1 (Wang et al. 2013; Lin et al. 88 2015). 89 Recently, crude glycerol has been proposed as a competitive substrate to reduce high 90 loads of sulphate to H2S (Mora et al. 2018). Crude glycerol is an organic waste produced 91 during biodiesel production, having the potential to improve the biological treatment of 92 flue gas scrubbing effluents. Fernández-Palacios et al. (2019) demonstrated that glycerol 93 can be successfully used for the long-term operation of a sulphidogenic, uplow anaerobic 94 sludge blanket (UASB) reactor, treating high sulphate loads (4.3 kg S·m-3·d-1). The 95 operation of this reactor was included in a potential alternative process to recover 96 elemental sulphur from flue gases called two-stage bioscrubber and consisting on a three 97 stages process: (1) SO2 and NOx sequential absorption, (2) sulphate reduction in the 98 UASB and (3) aerobic or anoxic sulphide oxidation in a continuous stirred tank reactor 99 (CSTR). 100 The sequential absorption of SO2 and NOx in two-stage bioscrubber is essential to 101 ensure optimal characteristics of bioreactor influents. To this aim, optimal conditions 102 required for the selective absorption of SO2 and NOx in two consecutive spray-absorbers 103 have been studied in this work. In addition, the reduction of reagents consumption in the 104 absorption unit has been assessed studying the alkaline character of the effluent of a 105 biological process for biosulfur recovery and its performance as specific absorbent for 106 flue gas treatment. The optimization of flue gas depuration was completed studying 107 oxidative reagents dosing in absorbents to increase NO oxidation and thus, overall NOx 108 removal efficiency. 109 110 2. EXPERIMENTAL METHODS 111 2.1.Process chemistry 112 2.1.1. Mechanisms of SO2 and NOx absorption 113 Based on the sulphur dioxide absorption mechanism (Eq. 1 to 6), flue gas 114 desulphurization through wet processes is generally performed using an alkaline solution 115 as absorbent for chemical absorption. 116 Sulphur dioxide diffuses from the gas phase to the gas-liquid interphase and it is 117 dissolved according to Henry's law (Eq. 1). In addition, a small amount of SO2 can be 118 dissociated following Eq. 2, a first order reaction with a reaction rate of 3.4·106 s-1 at 119 20ºC (Chang and Rochelle 1981). The dissolved SO2 reacts in the presence of hydroxide 120 ions through the second order reaction presented in Eq. 3 with a rate constant exceeding 121 109 L·mol-1·s-1 at 25ºC, while the generated hydrogen sulphite reacts instantaneously to 122 sulphite with additional hydroxide through proton transfer reaction defined in Eq. 4. 123 Caustic soda is usually used as absorbent since is a strong alkaline and thus a big a source 124 of hydroxide ions. In this sense, caustic soda will provide a surplus of hydroxide ions in 125 the aqueous phase and the overall reaction in the scrubber can be described according to 126 Eq. 5. On the other hand, under low pH conditions, the reaction of sulphite with dissolved 127 sulphur dioxide (Eq. 6) should be also considered. 𝑆𝑂 ( ) ↔ 𝑆𝑂 () (1) 𝑆𝑂 () + 𝐻 𝑂 ↔ 𝐻𝑆𝑂 ( ) + 𝐻( ) (2) 𝑆𝑂 () + 𝑂𝐻 ↔ 𝐻𝑆𝑂 ( ) (3) 𝐻𝑆𝑂 ( ) + 𝑂𝐻 ↔ 𝑆𝑂 ( ) +𝐻 𝑂 (4) 𝑆𝑂 ( ) + 2𝑂𝐻( ) → 𝑆𝑂 ( ) + 𝐻 𝑂( ) (5) 𝑆𝑂 ( ) + 𝑆𝑂 ( ) ↔ 2𝐻𝑆𝑂 ( ) (6) 128 129 The wet-type processes can be implemented using other alkaline reagents such as 130 calcium compounds (Sakai et al. 2002) or NaHCO3/Na2CO3 solutions (Ebrahimi et al. 131 2003). The absorption of SO2 into HCO3-/CO32- solutions and concomitant CO2 132 desorption is presented in Eq. 9. For this reason, it is important to also consider CO2 133 hydrolysis both at pH below 8 (Eq. 7) and at pH above 10 (Eq. 8). 𝐶𝑂 ( ) + 𝐻 𝑂 ↔ 𝐻( ) + 𝐻𝐶𝑂 ( ) (7) 𝐶𝑂 ( ) + 𝑂𝐻( ) ↔ 𝐻𝐶𝑂 ( ) (8) 𝑆𝑂 () + 𝐻𝐶𝑂 ( ) → 𝐶𝑂 ( ) + 𝐻𝑆𝑂 ( ) (9) 134 135 Gaseous NOx is a complex mixture of nitrogen species of different oxidation states. 136 Several reactions in gas phase can be considered in a NOx mixture (Newman and Carta 137 1988; Suchak et al. 1990; Thomas and Vanderschuren 1997). Reactions of NO and NO2 138 in presence of oxygen are shown in Eq. 10-12, and include NO oxidation to NO2 and the 139 instantaneous production of nitrogen species such as N2O3 and N2O4. 2𝑁𝑂( ) +𝑂 ( ) → 2𝑁𝑂 ( ) (10) 𝑁𝑂( ) + 𝑁𝑂 ( ) →𝑁 𝑂 ( ) (11) 2𝑁𝑂 ( ) →𝑁 𝑂 ( ) (12) 140 141 Absorption of NOx into liquid phase takes place both from dissolution and reaction of 142 the different species present in gas phase. The NO2, N2O3 and N2O4 are soluble gases 143 (Joshi et al. 1985) that are transferred to the liquid phase through instantaneous and 144 irreversible reaction with water to form nitrous and nitric acids as follows: 2𝑁𝑂 ( ) + 𝐻 𝑂( ) → 𝐻𝑁𝑂 ( ) + 𝑁𝑂 ( ) +𝐻 ( ) (13) 𝑁𝑂 ( ) + 𝐻 𝑂( ) → 2𝐻𝑁𝑂 ( ) (14) 𝑁𝑂 ( ) + 𝐻 𝑂( ) → 𝐻𝑁𝑂 ( ) + 𝑁𝑂 ( ) +𝐻 ( ) (15) 145 146 Nitrous oxide can also be generated from NO. Despite NO is sparingly insoluble and 147 non-reactive with water, under low concentrations reacts with NO2 (Joshi et al. 1985) 148 forming nitrous acid (Eq. 16). 𝑁𝑂( ) + 𝑁𝑂 ( ) + 𝐻 𝑂( ) → 2𝐻𝑁𝑂 ( ) (16) 149 150 In aqueous solution and especially under high nitric acid concentrations, nitrous acid 151 is very unstable and can be discomposed to NO (Eq. 17) that can be desorbed. In strong 152 alkaline solutions HNO2 decomposes through an instantaneous proton transfer reaction 153 (Eq. 18) preventing NO release. 3𝐻𝑁𝑂 ( ) → 𝐻𝑁𝑂 ( ) + 2𝑁𝑂( ) + 𝐻 𝑂( ) (17) 𝐻𝑁𝑂 ( ) + 𝑂𝐻 ( ) → 𝑁𝑂 ( ) + 𝐻 𝑂( ) (18) 154 155 2.1.2. Chemistry of NO oxidation 156 NO oxidation using oxidative reagents in the liquid phase takes place through 157 heterogeneous reactions between gaseous NO and aqueous species. Although several 158 reagents have been used for NO oxidation and absorption (Sun et al. 2016), the current 159 study is only focused on few of them. In the present work, the NO oxidation through 160 H2O2, NaClO, Na2S2O7 and KMnO4 solutions has been explored. Such oxidation 161 reactions are presented in Eq. 19-24. 𝑁𝑂( ) + 2𝐻 𝑂 () → 2𝑁𝑂 ( ) + 𝐻 𝑂( ) + 2𝐻 ( ) (19) 2𝑁𝑂( ) + 𝐶𝑙𝑂 ( ) + 𝐻 𝑂( ) → 2𝑁𝑂 ( ) + 𝐶𝑙 ( ) + 2𝐻 ( ) (20) 𝑁𝑂( ) +𝑆 𝑂 ( ) + 𝐻 𝑂( ) → 𝑁𝑂 ( ) + 2𝑆𝑂 ( ) + 2𝐻 ( ) (21) 2𝑁𝑂( ) +𝑆 𝑂 ( ) + 2𝐻 𝑂( ) (22) → 2𝑁𝑂 ( ) + 2𝑆𝑂 ( ) + 4𝐻 ( ) 2𝑁𝑂( ) + 3𝑆 𝑂 ( ) + 4𝐻 𝑂( ) (23) → 2𝑁𝑂 ( ) + 6𝑆𝑂 ( ) + 8𝐻 ( ) 3𝑁𝑂( ) + 𝑀𝑛𝑂 ( ) + 𝐻 𝑂( ) → 3𝑁𝑂 ( ) + 𝑀𝑛𝑂 ( ) + 2𝐻 ( ) (24) 162 163 2.2.Two-stage bioscrubber concept 164 2.2.1. Integrated flue gas bioscrubbing process 165 The two-stage bioscrubber integrates in a single process the removal of SO2 and NOx 166 from flue gases towards elemental sulphur valorisation. The conceptual process diagram 167 of the two-stage bioscrubber is detailed in Figure 1a and consisted of an absorption unit, 168 where sequential SO2 and NOx absorption is performed, and two biological reactors in 169 series for the treatment of resulting scrubbing liquors. To this aim, two consecutive spray- 170 scrubbers were designed and operated specifically for the absorption of SO2 and NOx, 171 respectively, obtaining separately two streams rich in sulphate/sulphite and nitrite/nitrate. 172 The first scrubber liquor is biologically treated in a heterotrophic UASB using crude 173 glycerol as carbon source. The sulphite and sulphate fed to the reactor are reduced to 174 sulphide, which is, in turn, oxidized to elemental sulphur in a consecutive CSTR operated 175 under anoxic conditions using nitrite and nitrate from the second scrubber effluent. The 176 biological sulphate reduction and the anoxic sulphide oxidation have been widely studied 177 (Jiang et al. 2013; Qian et al. 2015) and optimal conditions are described in detail in 178 Fernández-Palacios et al. (2019). 179 The absorption efficiency obtained for NOx treatment is generally low due to the scarce 180 solubility of NO in water. Nevertheless, the biological production of elemental sulphur is 181 performed using nitrite or nitrate as electron acceptor. For this reason, spray scrubbing 182 was also selected for NOx abatement in order to obtain a nitrogen laden aqueous stream. 183 The absorption stage was experimentally assessed herein through a lab-scale absorber. 184 2.2.2. Biosulphur recuperation 185 The technological feasibility of the biological process for biosulphur production 186 consisting of the sequential UASB-CSTR experimental setup was demonstrated in Mora 187 et al. (2020). The operation of the UASB reactor resulted in a sulphate removal efficiency 188 of 68.1% treating high sulphate concentrations (1000 mg S L-1) corresponding to a 189 maximum elimination capacity of 9.3 kg S m-3 d-1. The production of intermediate 190 compounds, such as organosulfur compounds, caused a decrease of elemental sulfur 191 productivity. These results showed that the compertitiviness of the process can be 192 improved enhancing the yield of sulphate reduction to sulfide. 193 Sulfide was not detected in CSTR effluent during elemental sulphur production. 194 Besides, sulfate concentration in the effluent of the CSTR was similar than sulfate 195 concentration in the influent which indicated the full capacity of the culture to oxidize 196 partially sulfide to sulfur. These results confirmed the proper design of the CSTR aeration 197 and fed. The implementation of a aeration/stirring control system based on redox potential 198 measurement ensured sulfide limiting conditions and avoided the accumulation of 199 dissolved sulfide in the CSTR and the consequent H2S release. The operation of CSTR 200 resulted in a sulfur productivitiy up to or 2.3 kg S m-3 d-1. Results also showed that residual 201 COD from the UASB outlet was partially adsorbed on on biosiulfur surface affecting its 202 structure and composition. 203 Elemental sulphur recovery from the CSTR effluent still requires to be optimized. 204 Nevertheless, a first approach showed that centrifugation is an effective biosulfur 205 recovery. The centrifugation allowed the separation of biomass layer deposited on 206 elemental sulfur. The obtained product exhibited a high purity (between 87.5% and 95%) 207 and a humidity content of 46%. However, costs associated to this specific separation 208 process are considerably high. For this reason, raw biosulphur sedimentation through the 209 addition of chemical agents (coagulants and flocculants) was assessed as an economical 210 alternative to centrifugation. Sedimentation was settled up through the addition of a 211 commercial cationic coagulant (FL4820) obtaining a product with a sulfur purity of 212 74.3% and 64.4% of sedimented solids. Further optimization through tailor-made 213 surfactants in to obtain higher biosulphur purity and solids recovery capacity. 214 Finally, dried biosulfur was obtained through a thermal treatment. From the 215 thermogravimetric results drying conditions biosulphur oxidation and melting were 216 established. Then the appropriate process to recover biosulfur from the CSTR was defined 217 as a sequence of settling after coagulant addition and air drying at 100ºC. 218 2.2.3. Experimental setup for flue gas absorption characterization 219 The schematic diagram of the spray-absorber used for the characterization of flue gas 220 treatment is detailed in Figure 1b. Absorption was characterized separately for each 221 pollutant generating synthetic gas by mixing metered amounts of SO2/NO and air. The 222 spray tower was fabricated in polyvinyl chloride (PVC) and consisted of a gas-liquid 223 contact section (5.9 cm of internal diameter and 30 cm high) and a liquid reservoir (11 224 cm of internal diameter and 20 cm height) with a working volume of 1800 mL. The 225 absorbent is injected into the top of the column using a peristaltic pump (7554-94 L/S 226 pump, Masterflex, USA) and is atomized in drops of 50-100 μm of diameter through a 227 spray nozzle (490.403.1Y.CA, Lechler, Germany). The pH and the ionic conductivity 228 were continuously measured at the reactor outlet using a multi-analyte probe (Multi 3420, 229 WTW, Germany). The synthetic flue gas was prepared using mass flow controllers (EL- 230 Flow Select, Bronkhorst, The Netherlands), and was fed counter-currently from the 231 bottom of the gas-liquid contactor. Continuous monitoring of SO2 or NOx (as a mixture 232 of NO and NO2) in the outlet gas phase was performed using an electrochemical SO2 233 sensor (SO2-B4, Alphasens, UK), an electrochemical NO sensor (4NO-2000, EuroGas, 234 UK) and an electrochemical NO2 sensor (4NO2-2000, EuroGas, UK), all of them with a 235 measuring range from 0 to 2000 ppmv. Measurement of higher pollutants concentrations 236 required flue gas dilution performed by two flowmeters (2100, Tecfluid, Spain). Residual 237 flue gas was bubbled through a NaOH filter before its emission to the atmosphere. The 238 selected fresh absorbent was continuously fed to ensure constant pH and composition at 239 reactor inlet. The compounds concentration, pH and ionic conductivity at liquid outlet 240 were continuously monitored until steady-state conditions were achieved. Experiments 241 were performed at constant temperature of 20±1 ºC and atmospheric pressure. Absorption 242 efficiency was evaluated for each pollutant by means of the following expression: Cin  Cout RE  100 (25) Cin 243 244 Where Cin is the inlet pollutant (SO2, NO or NO2) concentration and Cout is the outlet 245 pollutant (SO2, NO or NO2) concentration, both in ppmv. To ensure experiments accuracy, 246 three replicate measurements were made, obtaining relative errors below 5%. In this paper 247 only average values are shown. 248 2.3.Experimental design 249 The characterization of wet-absorption of SO2 and NOx was performed assessing the 250 influence of relevant operational parameters on process removal efficiency (RE). The 251 selected parameters presented in Table 1 were the gas contact time, the liquid-to-gas 252 flowrate (L/G) ratio, the sulphate and nitrate concentration in the phase, the absorbent pH 253 (adjusted using HCl 1M and NaOH 1M), SO2 and NOx concentration and the gas 254 temperature. The integrated process include the recovery of energy contained in flue gases 255 for elemental sulfur drying. For this reason, it was considered that flue gases temperature 256 at scrubber inlet would be always below 80ºC. The value of listed parameters in the 257 characterization assays is detailed in Table 1. 258 According to NOx reactions in gas phase presented in Eq. 10-12, NO contained in 259 synthetic flue gas stream was partially oxidized to NO2. For this reason, NOx absorption 260 was performed considering NOx concentration as the sum of NO and NO2 concentration 261 and assessing NOx RE considering both NO RE and NO2 RE. 262 The reuse of the alkalinity generated during sulphate reduction in the biological stage, 263 as absorbent in the flue gas scrubbing stage was evaluated by using different NaHCO3 264 concentrations. According to Fernández-Palacios et al. (2019) during the biological 265 sulphate reduction in an UASB using crude glycerol as electron donor, the dissolved CO2 266 concentration in the biological process effluent ranged between 0.05 and 0.20 g C- 267 HCO3·L-1. The suitability of reusing the biological process effluent for flue gas spray 268 scrubbing was assessed quantifying the SO2 RE using NaHCO3-based absorbents of 0.05 269 and 0.2 g C-HCO3·L-1. The UASB HCO3- concentration range was selected for the study 270 because no carbonate lost was detected during elemental sulphur production in CSTR. 271 Both NaHCO3-based were prepared by dissolving NaHCO3 in distillate water and 272 adjusting pH at the CSTR outlet pH (pH 7.5). The spray scrubbing using NaHCO3 273 solutions was studied considering the effect on RE of gas contact time (between 0.6 and 274 4 s), the L/G ratio (between 5 and 15 L·m-3) and SO2 concentration in flue gas (between 275 2000 and 4000 ppmv). 276 Finally, optimization of NOx absorption using specific absorbents was assessed 277 including different additives in alkaline absorbent prepared with NaOH solutions at pH 278 8. The additives were selected from their oxidative character (Thomas and Vanderschuren 279 1998; Zhao et al. 2011; Li et al. 2014; Han et al. 2015; Hao et al. 2017). To this aim, 280 alkaline absorbent including KMnO4 (0.3 M), NaClO (0.4 M), Na2S2O7 (0.2 M), H2O2 281 (0.15 M) were prepared and tested for NOx absorption. The potential of additives dosing 282 for NOx abatement was evaluated maintaining the same stoichiometric ratio between NO 283 and the different oxidative reagents was maintained for the preparation of absorbents. The 284 RE of prepared absorbent for NOx absorption was experimentally evaluated in a lab-scale 285 spray-scrubber fed with a gas stream of 500 ppmv of NO at a gas contact time of 10 s and 286 under a L/G ratio of 15 L·m-3. 287 288 3. RESULTS 289 3.1.Flue gas wet-absorption 290 3.1.1. Characterization of SO2 wet-absorption 291 Several sets of experiments were performed to characterize SO2 wet absorption. 292 Results shown in Figure 2 depict the effect of the tested operating variables on SO2 293 absorption efficiency. 294 Flue gas inlet velocity was increased resulting in a gas-liquid contact time decrease 295 from 10 to 0.6 s (Figure 2a) while the L/G ratio was maintained at 7.5 L·m-3. Results 296 showed a RE improvement of the 50%, from 20.6 to 68.4%. Keeping a constant L/G 297 within the scrubber ensures a constant gas-liquid contact-area. On the other hand, Wang 298 et al. (2015) showed that the resistance of the gas film declines due to the intensification 299 of the turbulences when the flues gas flowrate increases, enhancing SO2 mass transfer to 300 the liquid bulk. According to these results, SO2 absorption RE was not limited by the 301 shorter reaction times resulting from the decrease of the gas-liquid contact time 302 highlighting that absorption takes place through instantaneous irreversible reaction (Eq. 303 5), as is proposed in several studies (Hikita et al. 1977; Wang et al. 2012, 2015). 304 The L/G ratio refers to the volumetric ratio between absorbent and flue gas flowrates. 305 It is an important parameter for scrubbing performance since a higher L/G ratio results in 306 a higher SO2 RE but a higher energy consumption and operating costs. The L/G ratio in 307 spray absorbers for the removal of SO2 is typically set between 8 L·m-3 and 25 L·m-3. 308 Considering the high performance of NaOH solutions, the desulphurization efficiency 309 was evaluated between 5 L·m-3 and 15 L·m-3. To maintain the absorption characteristics, 310 experiments were performed at a gas-liquid contact time of 2 s. As showed in Figure 2b, 311 RE increased from 38.7% to 58.7% when L/G ratio rose from 5 L·m-3 to 12 L·m-3, and 312 from 68.7% to 79.8% when L/G ratio rose from 12 L·m-3 to 15 L·m-3 (until 70.8%). This 313 trend was in agreement with results found in literature (Gao et al. 2011; Wang et al. 2012, 314 2015). Increasing the L/G ratio at a constant value of flue gas flowrate rises absorbent 315 flowrate within the scrubber. Raising absorbent flowrate in the spray absorber means 316 increasing the interface area between droplets and gas, and more SO2 is absorbed (Zhao 317 et al. 2007). However, when the ratio was increased above 10 L·m-3, the droplets area 318 was large enough, and the increase in the ratio became negligible (Wang et al. 2015). 319 In Figure 2c, it can be observed an increase in RE from 50 to 60% when ionic sulphur 320 species concentration in absorbent increased to 10 g S·L-1. This results are in agreement 321 with Wang et al. (2012) that found stronger absorption capacities when Na2SO3 322 concentration in the absorbent exceeded 0.3 M. 323 The influence of NaOH concentration on scrubbing performance was evaluated 324 assessing the influence of absorbent pH on SO2 RE. The pH was ranged between 3 and 325 15. The operation at pH below 7 was also evaluated to study scrubbing performance under 326 NaOH depletion conditions. Results presented in Figure 2d show two different trends 327 along the studied pH range. In this sense, at pH below 12 RE decreased linearly from 328 60.8% to 29.6% while at pH above 12 a constant RE ranging between 59.5% and 60.8% 329 was observed. At low pH, SO2 absorption takes place through Eq. 2 which is much slower 330 than Eq. 3 and Eq. 4 (Hikita et al. 1977). This phenomenon indicated that mass transfer 331 of SO2 to the liquid phase is governed both by gas film and liquid film resistance (Wang 332 et al. 2015), leading to low efficiencies. Nevertheless, absorption took place through Eq.3 333 and Eq.4 when pH raised due to a higher hydroxide ions concentration, and liquid film 334 resistance fell quickly (Wang et al. 2012). Under this conditions, process is governed by 335 gas mass transfer and a higher and constant overall RE was obtained. 336 The SO2 concentration in flue gas is a parameter highly unstable depending on the fuel 337 composition or combustion facilities operation. For this reason, it is important to establish 338 the influence of SO2 inlet concentration on sulphur removal efficiency. Results presented 339 in Figure 2e showed that the SO2 inlet concentration increase from 1000 to 9000 ppmv 340 promoted mass transfer limitations. Since the L/G ratio was kept constant at 7.5 L·m-3, 341 the exhaustion of the absorbent took place at higher SO2 concentration. Under these 342 conditions, mass transfer limitations described above for low pH absorbents become 343 important resulting in a RE decrease from 70.9 to 35.1%. This results are in agreement 344 with results in literature (Gao et al. 2011; Wang et al. 2012, 2015; Rahmani et al. 2014), 345 and suggests that some actions (such as increasing L/G ratio or absorbent pH) should be 346 taken to increase RE at higher SO2 inlet concentrations. 347 The effect of temperature on SO2 absorption process is complex. Temperature rise can 348 reduce the solubility of SO2 but improve mass transfer coefficient and also accelerate 349 chemical reactions rates (Wang et al. 2012). For this reason, the flue gas influence on SO2 350 RE was evaluated. The liquid temperature at the beginning of the experiments was kept 351 at 25ºC. A flue gas temperature effect on RE was not observed within the tested range 352 (Figure 2f), since liquid phase temperature and thus SO2 solubility, kept steady during 353 absorption. 354 3.1.2. Characterization of NOx wet-absorption 355 The experimental results obtained during the characterization of NOx absorption are 356 presented in Figure 3. Results obtained during NOx wet-absorption characterization 357 showed RE below 50% both for NO and NOx abatement. These results highlighted strong 358 mass transfer limitations during NOx removal through absorption technologies due to low 359 solubility of NO in water (0.00618 g NO·100 g-1 H2O). Also, small differences were 360 observed between NO and NOx RE, since nitrogen species generated due to NO removal 361 presented a high and fast reactivity in water. 362 As observed in Figure 3a, an increase on gas retention time from 0.6 to 120 s resulted 363 in an improvement on NOx RE from 5.4 to 50.5% because the larger gas contact time 364 increased the time available for NO oxidation through Eq. 10-12. Higher solubility of 365 oxidized nitrogen species reduced gas side mass transfer limitations, and higher overall 366 NOx RE were obtained. Thus, the scrubbing performance for NOx abatement can be 367 enhanced by increasing retention time of flue gas in the scrubber. 368 However, the rise of L/G ratio from 5 to 15 L·m-3 (Figure 3b) only improved NOx RE 369 from 5.9 to 9.2%. These results evidenced that at the low contact time (2 s) maintained 370 during these assays, NO was not oxidized. Therefore, mass transfer rate of gaseous 371 species to the bulk liquid is controlled by gas film resistance. In this sense, the increase 372 of mass transfer interface at higher absorbent flowrate does not improve NOx transfer rate 373 resulting in low RE (below 10%). 374 Results presented in Figure 3c-d showed again that at a short contact time low NO 375 oxidation took place within the scrubber. NO is still the main component of NOx mixture 376 and under these conditions absorption is limited by Eq. 10 and gas film mass transfer 377 resistance. For this reason, an enhancement of NOx removal was not obtained increasing 378 reaction rate of aqueous species (Eq. 13-15). In this sense, no significant effects on NOx 379 RE were observed when liquid side mass transfer and bulk liquid kinetic characteristics 380 were modified both increasing nitrate absorbent concentration (Figure 3c) and absorbent 381 pH (Figure 3d). 382 The same phenomenon was observed when NO inlet concentration was increased. In 383 Figure 3e, a RE drop from 13.9% to 5.8% is observed when inlet NO concentration raised 384 from 500 to 2000 ppmv. These results highlighted the necessity of sharply reduce NO 385 mass transfer limitations, both increasing mass transfer rate or Eq. 10 reaction rate, in 386 order to obtain higher RE and to accomplish environmental requirements. 387 Similar to that observed in Figure 2, gas temperature had a no effect on NOx 388 absorption. The absorbent was not warmed by the hotter flue gas stream and no changes 389 on liquid temperature were observed during experimental assays. For this reason, a 390 constant RE between 6.8% and 7.3% was obtained when inlet gas temperature ranged 391 between 298 and 333 K. 392 3.1.3. Sequential SO2 and NOx wet-absorption 393 The segregation of sulphur and nitrogen ionic species in two separated liquid effluents 394 required for optimal performance of biological reactors was obtained through the 395 sequential absorption of SO2 and NOx from flue gas. To this aim, flue gas treatment unit 396 was designed as two scrubbers in series and selective absorption of both pollutants was 397 achieved exploiting different chemical properties of both pollutants. The first scrubber 398 was designed for SO2 absorption while the second one was designed for NOx absorption. 399 The operating conditions of scrubbers were defined from the experimental data obtained 400 in the characterization of flue gas absorption. The operating conditions of both scrubbers 401 established for sequential absorption of both pollutants in two consecutive units are 402 presented in Table 2. The removal efficiencies of both scrubbers for SO2 and NOx 403 abatement treating a gas stream with 2000 ppmv of SO2 and 1500 ppmv of NOx are 404 included in Table 2. 405 The difference in water solubility of both gaseous pollutants allowed sequential 406 absorption in two consecutive scrubbers. The gas contact time was set at 1 s in the first 407 scrubber. Under these operating conditions SO2 absorption was enhanced but NOx 408 absorption capacity was strongly limited. Removal efficiencies obtained in the first 409 scrubber for SO2 and NOx absorption are presented in Table 2. Higher SO2 solubility 410 resulted in a RE of 72.9% while RE for NOx absorption was kept at 5.4%. This spray- 411 scrubber was also operated under moderate L/G ratio and absorbent pH in order to 412 increase NOx gas-liquid mass transfer limitations and to reduce operating costs. 413 The second spray-absorber was operated at higher gas contact time to increase time 414 available for NO oxidation (Eq.10-12) and thus NOx absorption. The L/G ratio was also 415 increased to 15 L·m-3 to intensify NO oxidation products absorption from the gas to the 416 liquid phase. Under these operating conditions NO oxidation was increased and the NOx 417 generated were successfully absorbed, obtaining a RE for NOx removal of 50.5%. Results 418 showed that most of SO2 still remaining in flue gas was also absorbed. 419 Although overall removal efficiencies of 98.9% and 55.9% were obtained in two 420 consecutive spray-absorbers for SO2 and NOx, respectively, complete flue gas treatment 421 unit should be optimized to improve overall two-stage bioscrubber performance. In this 422 sense, it is necessary to explore the reduction of operating costs, to raise SO2 RE in the 423 first scrubber to increase sulphur recovery capacity of the process and to raise NOx RE to 424 ensure clean gas release. 425 3.2.Optimization of flue gas treatment 426 3.2.1. SO2 absorption using recovered NaHCO3/Na2CO3 427 Flue gas treatment through wet-absorption technology implies a high consumption in 428 alkali reagents and thus high recurring operating costs. To this aim, the two-stage 429 bioscrubber was designed including bioprocess effluent recirculation to the absorption 430 unit (Figure 1a) in order to recover the NaHCO3/Na2CO3 generated during heterotrophic 431 sulphate reduction due to glycerol oxidation. The optimization of flue gas absorption from 432 an economic point of view was performed evaluating the suitability of aqueous NaHCO3 433 solutions as absorbent for SO2 removal within the concentration range in biological 434 process effluent (Fernández-Palacios et al. 2019) in order to simulate the closed-loop 435 process. This study was performed quantifying the effect of the L/G ratio, the gas contact 436 time and the SO2 concentration on system RE. 437 Results obtained in the characterization of SO2 absorption using mimicking absorbents 438 of biological origin are presented in Fig. 4. NaHCO3 absorbent, both with low and with 439 high concentration, showed a RE for SO2 absorption above 80% and 40% in the most and 440 less favourable case, respectively. A strong effect of L/G ratio, gas contact time and SO2 441 concentration on absorption capacity was also observed. Experimental results obtained 442 working with an absorbent of 0.05 g C·L-1 concentration showed that a L/G ratio increase 443 from 5 to 15 L·m-3 produced an increase on SO2 RE from 55.6 to 76.3% and from 43.6 444 to 73.7% feeding a SO2 concentration of 2000 ppmv and 4000 ppmv, respectively. 445 Conversely, increasing gas contact time from 0.6 to 4 s, the RE for SO2 absorption 446 decreased from 81.6 to 59.7% and from 67.9 to 43.4% when lab-scale spray-scrubber was 447 fed with a 2000 ppmv and a 4000 ppmv SO2 concentration, respectively. 448 The comparison of NaOH-based and NaHCO3 absorbent performance for SO2 449 treatment is presented in Table 3. Experimental results presented in Figure 2a-b and 450 Figure 4 have been summarized in Table 3, and no differences were observed between 451 both absorbent. These results demonstrate that the use of a regenerated absorbent does 452 not introduce technical limitations in the absorption stage. In addition, according to results 453 obtained working with NaOH absorbent (in Figure 2a), higher RE were also observed at 454 a low flue gas contact time. These results proved that alkalinity release due to dissolved 455 CO2 stripping in spray-scrubber (Eq. 9) was more efficient when operating at high gas 456 flow rates. This is explained by turbulences in reactor under these operating conditions 457 that enhanced mass transfer of dissolved CO2. 458 Results obtained using NaHCO3 absorbent with higher concentration (Figure 4c-d) 459 showed the same trend. In this case, RE for SO2 removal, when L/G ratio increased from 460 5 to 15 L·m-3, rose from 62.1 to 81.8% and from 65.4 to 85.4% when a flue gas of 2000 461 and 4000 ppmv of SO2 was studied. On the other hand, an increase of gas contact time 462 from 0.6 s to 4 s also reduced RE for SO2 abatement. In this sense, RE decreased from 463 85.7% to 70.4% and from 84.8% to 69.3% was observed working at 2000 ppmv and 4000 464 ppmv, respectively. 465 These results showed that RE for SO2 absorption obtained using the absorbent with 466 the highest CO2 concentration are about a 10% higher. This trend confirmed that the 467 utilisation of NaHCO3/Na2CO3 solutions as flue gas absorbents was not limited by CO2 468 stripping in the spray scrubber since most of the alkalinity contained in these solutions 469 can be released during scrubbing process. These results also confirmed spray-scrubber as 470 a good alternative for flue gas treatment. 471 Therefore, it can be concluded that biological effluents can be a good alternative to 472 replace conventional NaOH-based absorbent in wet flue gas absorption Thereby, the 473 reduction of reagents consumption in absorption stage showed that two stage bioscrubber 474 reduced recurring operating costs related to NaOH consumption. 475 3.2.2. NOx oxidation 476 The reduction of mass transfer limitations for NOx absorption were evaluated dosing 477 oxidative reagents in the spray-scrubber in order to reduce the release of pollutants to the 478 atmosphere. NO oxidation enriches NOx composition in NO2 as well as in N2O3 and N2O4 479 (according to Eq. 10-12), which react irreversibly with water forming aqueous nitrogen 480 species. This alternative was evaluated assessing several oxidants for NOx absorption. 481 Results are presented in Figure 5, considering RE for NO oxidation and RE for the 482 absorption of oxidation products (RE-NOx). 483 Experimental results of NOx abatement including oxidizing agents in the scrubbing 484 solution showed higher RE for NO oxidation than for NOx absorption. Absorbents based 485 on H2O2, NaClO and Na2S2O7 presented similar oxidation efficiencies, ranging from 12 486 to 18%. However, Na2S2O7 is an acid salt and, therefore, presented low absorption 487 efficiencies. As can be seen in Eq. 13-15, despite NO oxidation products presented a 488 higher solubility, the alkaline character of the absorbent is key for NOx absorption. 489 Absorption based on NaOH solutions presented similar efficiencies both for NO 490 oxidation and NO2 absorption, close to the 30% and higher than the exhibited for most of 491 the tested oxidants. On the other hand, the KMnO4 solution stands out from the other 492 absorbents showing oxidation and absorption efficiencies close to 50 and 60%, 493 respectively. According to experimental results, the dosing of oxidative reagents in the 494 absorbent can be considered as a suitable alternative to increase RE for NOx scrubbing 495 and to ensure clean gas emissions. 496 3.3.Economic viability assessment 497 The improvement of the treatment of combustion gases through the two-stage 498 bioscrubber was quantified from an economic point of view by calculating the savings in 499 operating costs introduced by the proposed alternative. To this aim, the treatment of a flue 500 gas stream of 20000 Nm3·h-1 with a SO2 concentration of 3000 ppmv and a NOx 501 concentration of 500 ppmv in a spray-scrubber was defined as case study. In addition, a 502 RE of 90% was considered for both alternatives and a stripping efficiency of the 55% of 503 dissolved CO2 in the regenerated absorbent was supposed for the two-stage bioscrubber. 504 These conditions were selected according to typical emissions in combustion facilities 505 (Moore 1987; Karatepe 2000; Srivastava and Jozewicz 2001). According to process 506 chemistry defined in Eq. 1-18, operating costs for the treatment of the target emission in 507 a wet-absorption facility using NaOH-based absorbent and in the two-stage bioscrubber 508 considering costs-savings introduced by NaHCO3 rich stream reuse in absorption were 509 calculated (Table 4). 510 The treatment of flue gas stream of the case study through a conventional spray 511 scrubber using NaOH-based absorbents implies a reagents consumption of 1685 Tn 512 NaOH·year-1. On the other hand, the biological treatment of the 76.94 kg SO2·h-1 resulting 513 from flue gas treatment was assessed using experimental results presented in Fernández- 514 Palacios (Fernández-Palacios et al. 2019) for the characterization of biological reactors 515 performance. According to the authors, treatment of the scrubber effluent under optimal 516 operating conditions must be performed using crude glycerol as carbon source fed at a 517 carbon/sulphur ratio of 5 g O2·g-1 S. The operation of both biological reactors under these 518 conditions gives rise to a process effluent with an average dissolved CO2 concentration 519 of 0.10 g C·L-1, equivalent to an annual amount of 2558 Tn. The reuse of the NaHCO3 520 produced during the bioprocess allows annual NaOH savings of 1074 Tn, reducing the 521 operating costs of wet-absorption by a 63.7%. These results demonstrate the economic 522 competitiveness of the bio-FGD alternative proposed herein. 523 524 4. CONCLUSIONS 525 Experimental results obtained in this work enable to establish the optimal operating 526 conditions for the sequential wet-scrubbing of SO2 and NOx from flue gas. The SO2 and 527 NOx wet-absorption presented different performances caused by the difference in their 528 solubility. While high RE can be obtained for the treatment of SO2 working at low RT 529 (below 2 s) and at moderate L/G ratios (7.5 L·m-3), mass transport limitations for NOx 530 absorption resulted in RE below 10% under these operating conditions. The RT for NOx 531 abatement must be increased above 60 s in order to enable enough time for NO partial 532 oxidation and increase NOx absorption efficiency. Thus selective absorption of both 533 pollutants can be performed in two spray-scrubbers in series designed to work under 534 clearly differentiated RT. In addition, second scrubber can include the dosing of oxidative 535 reagents in order increase NO oxidation and capture NOx from flue gas. Thus, sequential 536 wet-scrubbing process allows to ensure clean gas emissions. The reuse of the two-stage 537 bioscrubber effluent as absorbent in the scrubbing unit showed the same RE than using a 538 NaOH-based absorbent. Economic comparison of proposed bio-FGD alternative against 539 spray-scrubber operated using NaOH-bases absorbents showed 63.74% savings on 540 operational costs due to the reduction in reagents consumption. 541 542 ACKNOLEDGEMENTS 543 Authors acknowledge the Spanish Government, through the project RTI2018-099362- 544 B-C21 MINECO/FEDER, EU for the financial support provided to perform this research. 545 546 REFERENCES 547 Adewuyi YG, Sakyi NY (2013) Simultaneous absorption and oxidation of nitric oxide 548 and sulfur dioxide by aqueous solutions of sodium persulfate activated by 549 temperature. Ind Eng Chem Res 52:11702–11711. 550 https://doi.org/10.1021/ie401649s 551 Aǧar A, Küçükatay V, Yargiçoǧlu P, et al (2000) The effect of sulfur dioxide inhalation 552 on visual evoked potentials, antioxidant status, and lipid peroxidation in alloxan- 553 induced diabetic rats. Arch Environ Contam Toxicol 39:257–264. 554 https://doi.org/10.1007/s002440010103 555 Altuǧ H, Gaga EO, Döǧeroǧlu T, et al (2013) Effects of air pollution on lung function 556 and symptoms of asthma, rhinitis and eczema in primary school children. Environ 557 Sci Pollut Res 20:6455–6467. https://doi.org/10.1007/s11356-013-1674-1 558 An T, Wan S, Li G, et al (2010) Comparison of the removal of ethanethiol in twin- 559 biotrickling filters inoculated with strain RG-1 and B350 mixed microorganisms. J 560 Hazard Mater 183:372–380. https://doi.org/10.1016/j.jhazmat.2010.07.035 561 Bai J, Meng Z (2005) Effects of sulfur dioxide on apoptosis-related gene expressions in 562 lungs from rats. Regul Toxicol Pharmacol 43:272–279. 563 https://doi.org/10.1016/j.yrtph.2005.09.002 564 Chang CS, Rochelle GT (1981) SO2 absorption into aqueous solutions. AIChE J 27:292– 565 298. https://doi.org/10.1002/aic.690270217 566 Chen JC, Liu ZS, Huang JS (2007) Emission characteristics of coal combustion in 567 different O2/N2, O2/CO2 and O2/RFG atmosphere. J Hazard Mater 142:266–271. 568 https://doi.org/10.1016/j.jhazmat.2006.08.021 569 Ding J, Zhong Q, Zhang S, et al (2014) Simultaneous removal of NOX and SO2 from 570 coal-fired flue gas by catalytic oxidation-removal process with H2O2. Chem Eng J 571 243:176–182. https://doi.org/10.1016/j.cej.2013.12.101 572 Ebrahimi S, Picioreanu C, Kleerebezem R, et al (2003) Rate-based modelling of SO2 573 absorption into aqueous NaHCO3/Na2CO3 solutions accompanied by the 574 desorption of CO2. Chem Eng Sci 58:3589–3600. https://doi.org/10.1016/S0009- 575 2509(03)00231-8 576 Fernández-Palacios E, Lafuente J, Mora M, Gabriel D (2019) Exploring the performance 577 limits of a sulfidogenic UASB during the long-term use of crude glycerol as electron 578 donor. Sci Total Environ 688:1184–1192. 579 https://doi.org/10.1016/j.scitotenv.2019.06.371 580 Gao H, Li C, Zeng G, et al (2011) Flue gas desulphurization based on limestone-gypsum 581 with a novel wet-type PCF device. Sep Purif Technol 76:253–260. 582 https://doi.org/10.1016/j.seppur.2010.10.013 583 Han Y, Li X, Fan M, et al (2015) Abatement of SO2–NOx binary gas mixtures using a 584 ferruginous active absorbent: Part I. Synergistic effects and mechanism. J Environ 585 Sci 30:55–64. https://doi.org/10.1016/j.jes.2014.10.012 586 Hao R, Yang S, Yuan B, Zhao Y (2017) Simultaneous desulfurization and denitrification 587 through an integrative process utilizing NaClO2/Na2S2O8. Fuel Process Technol 588 159:145–152. https://doi.org/10.1016/j.fuproc.2017.01.018 589 Hikita H, Asai S, Tsuji T (1977) Absorption of sulfur dioxide into aqueous sodium 590 hydroxide and sodium sulfite solutions. AIChE J 23:538–544. 591 https://doi.org/10.1002/aic.690230419 592 Hu Y, Naito S, Kobayashi N, Hasatani M (2000) CO 2 , NO x and SO 2 emissions from 593 the combustion of coal with high oxygen concentration gases. Fuel 79:1925–1932. 594 https://doi.org/http://dx.doi.org/10.1016/S0016-2361(00)00047-8 595 Jiang F, Zhang L, Peng G-L, et al (2013) A novel approach to realize SANI process in 596 freshwater sewage treatment - Use of wet flue gas desulfurization waste streams as 597 sulfur source. Water Res 47:5773–5782. 598 https://doi.org/10.1016/j.watres.2013.06.051 599 Jin Y, Veiga MC, Kennes C (2005) Bioprocesses for the removal of nitrogen oxides from 600 polluted air. J Chem Technol Biotechnol 80:483–494. 601 https://doi.org/10.1002/jctb.1260 602 Joshi JB, Mahajani VV, Juvekar VA (1985) Absorption of NOx Gases. Chem Eng 603 Commun 33:1–92. https://doi.org/10.1080/00986448508911161 604 Kan H, Wong C-M, Vichit-Vadakan N, et al (2010) Short-term association between sulfur 605 dioxide and daily mortality: The Public Health and Air Pollution in Asia (PAPA) 606 study. Environ Res 110:258–264. https://doi.org/10.1016/j.envres.2010.01.006 607 Karatepe N (2000) A comparison of flue gas desulfurization processes. Energy Sources 608 22:197–206. https://doi.org/10.1080/00908310050013983 609 Lee KH, Sublette KL (1991) Simultaneous combined microbial removal of sulfur dioxide 610 and nitric oxide from a gas stream. Appl Biochem Biotechnol 28–29:623–634. 611 https://doi.org/10.1007/BF02922637 612 Lens PNL, Gastesi R, Lettinga G (2003) Use of sulfate reducing cell suspension 613 bioreactors for the treatment of SO2 rich flue gases. Biodegradation 14:229–240. 614 https://doi.org/10.1023/A:1024222020924 615 Li Y, Zhong W, Ju J, et al (2014) Experiment on simultaneous absorption of NO and SO2 616 from sintering flue gas by oxidizing agents of KMnO4/NaClO. Int J Chem React 617 Eng 12:539–547. https://doi.org/10.1515/ijcre-2014-0066 618 Lin J, Li L, Ding W, et al (2015) Continuous desulfurization and bacterial community 619 structure of an integrated bioreactor developed to treat SO2 from a gas stream. J 620 Environ Sci (China) 37:130–138. https://doi.org/10.1016/j.jes.2015.05.029 621 Liu Y, Bisson TM, Yang H, Xu Z (2010) Recent developments in novel sorbents for flue 622 gas clean up. Fuel Process Technol 91:1175–1197. 623 https://doi.org/10.1016/j.fuproc.2010.04.015 624 Majeed JG, Korda B, Békássy-Molnár E (1995) Comparison of the efficiencies of sulfur 625 dioxide absorption using calcium carbonate slurry and sodium hydroxide solution in 626 an ALT reactor. Gas Sep Purif 9:111–120. https://doi.org/10.1016/0950- 627 4214(95)93948-J 628 Mathieu Y, Tzanis L, Soulard M, et al (2013) Adsorption of SOx by oxide materials: A 629 review. Fuel Process Technol 114:81–100. 630 https://doi.org/10.1016/j.fuproc.2013.03.019 631 Moore DJ (1987) Conditions in which power stations contribute to high ground level SO2 632 concentrations in the U.K. Atmos Environ 21:1849–1853. 633 https://doi.org/10.1016/0004-6981(87)90126-0 634 Mora M, Fernández-Palacios E, Guimerà X, et al (2020) Feasibility of S-rich streams 635 valorization through a two-step biosulfur production process. Chemosphere 253:1– 636 10. https://doi.org/10.1016/j.chemosphere.2020.126734 637 Mora M, Lafuente F javier, Gabriel D (2018) Screening of biological sulfate reduction 638 conditions for sulfidogenesis promotion using a methanogenic granular sludge. 639 Chemosphere 210:557–566. https://doi.org/10.1016/j.chemosphere.2018.07.025 640 Newman BL, Carta G (1988) Mass transfer in the absorption of nitrogen oxides in alkaline 641 solutions. AIChE J 34:1190–1199. https://doi.org/10.1002/aic.690340715 642 Ottengraf SPP, Diks RMM (1992) Process technology of biotechniques. Biotechnoques 643 Air Pollut Abat Odour Control Policies 51:17–31. https://doi.org/10.1016/S0166- 644 1116(08)70673-3 645 Perry RH, Green DW (1997) Perry’s chemical engineer’s handbook, 7 Edition. Mc Graw- 646 Hill 647 Philip L, Deshusses MA (2003) Sulfur dioxide treatment from flue gases using a 648 biotrickling filter - Bioreactor system. Environ Sci Technol 37:1978–1982. 649 https://doi.org/10.1021/es026009d 650 Qian J, Lu H, Jiang F, et al (2015) Beneficial co-treatment of simple wet flue gas 651 desulphurization wastes with freshwater sewage through development of mixed 652 denitrification-SANI process. Chem Eng J 262:109–118. 653 https://doi.org/10.1016/j.cej.2014.09.066 654 Rahmani F, Mowla D, Karimi G, et al (2014) SO2 removal from simulated flue gas using 655 various aqueous solutions: Absorption equilibria and operational data in a packed 656 column. Sep Purif Technol 153:162–169. 657 https://doi.org/10.1016/j.seppur.2014.10.028 658 Rao AG, Ravichandra P, Joseph J, et al (2007) Microbial conversion of sulfur dioxide in 659 flue gas to sulfide using bulk drug industry wastewater as an organic source by 660 mixed cultures of sulfate reducing bacteria. J Hazard Mater 147:718–725. 661 https://doi.org/10.1016/j.jhazmat.2007.01.070 662 Sakai M, Su C, Sasaoka E (2002) Simultaneous removal of SOx and NOx using slaked 663 lime at low temperature. Ind Eng Chem Res 41:5029–5033 664 Shen TT, Ayer LG (1975) Determination of S02 Emission from Oil Combustion Sources 665 by Fuel Analyses and S02 /C02 Relations. J Air Pollut Control Assoc 25:851–855. 666 https://doi.org/10.1080/00022470.1975.10470150 667 Srivastava RK, Jozewicz W (2001) Flue gas desulfurization: The state of the art. J Air 668 Waste Manag Assoc 51:1676–1688. 669 https://doi.org/10.1080/10473289.2001.10464387 670 Suchak NJ, Jethani KR, Joshi JB (1990) Absorption of Nitrogen Oxides in Alkaline 671 Solutions: Selective Manufacture of Sodium Nitrite. Ind Eng Chem Res 29:1492– 672 1502. https://doi.org/10.1021/ie00103a059 673 Sun J, Li L, Zhou G, et al (2018) Biological Sulfur Reduction to Generate H 2 S As a 674 Reducing Agent to Achieve Simultaneous Catalytic Removal of SO 2 and NO and 675 Sulfur Recovery from Flue Gas. Environ Sci Technol 52:4754–4762. 676 https://doi.org/10.1021/acs.est.7b06551 677 Sun Y, Zwolińska E, Chmielewski AG (2016) Abatement technologies for high 678 concentrations of NOx and SO2 removal from exhaust gases: A review. Crit Rev 679 Environ Sci Technol 46:119–142. https://doi.org/10.1080/10643389.2015.1063334 680 Thomas D, Vanderschuren J (1998) Removal of tetravalent NOx from flue gases using 681 solutions containing hydrogen peroxide. Chem Eng Technol 21:975–981. 682 https://doi.org/10.1002/(SICI)1521-4125(199812)21:12<975::AID- 683 CEAT975>3.0.CO;2-H 684 Thomas D, Vanderschuren J (1997) Modeling of NOx absorption into nitric acid solutions 685 containing hydrogen peroxide. Ind Eng Chem Res 36:3315–3322. 686 https://doi.org/10.1021/ie960436g 687 Tian H, Gao J, Lu L, et al (2012) Temporal trends and spatial variation characteristics of 688 hazardous air pollutant emission inventory from municipal solid waste incineration 689 in China. Environ Sci Technol 46:10364–71. https://doi.org/10.1021/es302343s 690 Van Houten RT, Yun SY, Lettinga G (1997) Thermophilic sulphate and sulphite 691 reduction in lab-scale gas-lift reactors using H2 and CO2 as energy and carbon 692 source. Biotechnol Bioeng 55:807–814. https://doi.org/10.1002/(SICI)1097- 693 0290(19970905)55:5<807::AID-BIT11>3.0.CO;2-8 694 Wang J, Cao Y, Zhong Q (2013) Formulation and optimization of biological removal of 695 flue gas pretreatment wastewater and sulfur recycling process by Box-Behnken 696 design. Water Sci Technol 67:2706–2711. https://doi.org/10.2166/wst.2013.175 697 Wang W, Yang C, Zhang J (2012) Absorption of Sulphur Dioxide from Flue Gas with 698 Sodium Alkali Solution in Packed Columns. Adv Mater Res 383–390:6409–6415. 699 https://doi.org/10.4028/www.scientific.net/AMR.383-390.6409 700 Wang Z, Peng Y, Ren X, et al (2015) Absorption of sulfur dioxide with sodium hydroxide 701 solution in spray columns. Ind Eng Chem Res 54:8670–8677. 702 https://doi.org/10.1021/acs.iecr.5b02146 703 Winnes H, Fridell E (2010) Emissions of NOX and particles from manoeuvring ships. 704 Transp Res Part D Transp Environ 15:204–211. 705 https://doi.org/10.1016/j.trd.2010.02.003 706 Zevenhoven, Kilpinen (2001) Chapter 2 Flue gases and fuel gases 2.1. Flue Gases Fuel 707 Gases 1–12 708 Zhao J, Jin B, Zhong Z (2007) The degree of desulphurization of a limestone/gypsum wet 709 FGD spray tower using response surface methodology. Chem Eng Technol 30:517– 710 522. https://doi.org/10.1002/ceat.200600347 711 Zhao Y, Han Y, Ma T, Guo T (2011) Simultaneous desulfurization and denitrification 712 from flue gas by ferrate(VI). Environ Sci Technol 45:4060–4065. 713 https://doi.org/10.1021/es103857g 714 Zhu Q, Van Teeffelen RM, Van Santen RA, Hensen EJM (2004) Effect of high- 715 temperature treatment on Fe/ZSM-5 prepared by chemical vapor deposition of FeCl3 716 II. Nitrous oxide decomposition, selective oxidation of benzene to phenol, and 717 selective reduction of nitric oxide by isobutane. J Catal 221:575–583. 718 https://doi.org/10.1016/j.jcat.2003.09.025 719