Simultaneous removal of SO2 and NO in a rotating drum biofilter coupled with complexing absorption by FeII(EDTA)

Biochemical Engineering Journal 114 (2016) 87–93

Contents lists available at ScienceDirect

Biochemical Engineering Journal journal homepage: www.elsevier.com/locate/bej

Regular article

Simultaneous removal of SO2 and NO in a rotating drum biofilter coupled with complexing absorption by FeII (EDTA) Jun Chen a , Siyang Gu b , Ji Zheng b , Jianmeng Chen a,∗ a Engineering Research Center of the Ministry of Education for Bioconversion and Biopurification, Zhejiang University of Technology, Hangzhou,310032, PR China b College of Environment, Zhejiang University of Technology, Hangzhou 310032, PR China

a r t i c l e

i n f o

Article history: Received 20 April 2016 Received in revised form 9 June 2016 Accepted 25 June 2016 Available online 25 June 2016 Keywords: Rotating drum biofilter Desulfurization Denitrification Complexing absorption

a b s t r a c t A promising process of microbial desulfurization and denitrification integrated with complexing absorption is under the development for simultaneous removal of SO2 and NO in a rotating drum biofilter (RDB). The process employs FeII (EDTA) to improve the NO mass transfer, the denitrifiers bacteria (NR) for the denitrification, and the sulfate reducing bacteria (SRB) for the desulfurization. Experimental results demonstrated that NO removal efficiency was significantly improved from 57.1% to 93.0% in the presence of 10 mM FeII (EDTA) in the 60-day operation. Meanwhile, the SO2 removal efficiency reached around 90%. Parametric tests showed the maximal removal efficiency of SO2 and NO could be achieved at 98.5% and 93%, respectively, in the conditions of 2000 mg/m3 SO2 , 800 mg/m3 NO, 1.8 min of empty bed residence time (EBRT), and 2 vol% oxygen. The microbial community analysis with the high-throughput sequencing indicated that the dominant denitrification bacteria with a maximal abundance of 35.7% were Pseudomonas which was mainly distributed in the external sphere of RDB and the dominant SRB was Desulfovibrio, which was mainly distributed in nutrient solution with a maximal abundance of 6.1%. © 2016 Elsevier B.V. All rights reserved.

1. Introduction The rapid industrial development in the recent years increases the emissions of sulfur dioxide (SO2 ) and nitrogen oxides (NOx ) which are the principal cause of the acid rain. In addition, NOx contribute to the photochemical smog, the depletion of the ozone layer, as well as global warming [1,2]. With the stringent emission standards of NOx and SO2 imposed by Gothenburg and Kyoto Protocols, the development of new technology and/or improvement of currently used methods are essential. Presently, chemical and physical technologies [3–6] such as oxidation absorption reduction [7], NO × SO process [8,9] and wet flue gas desulfurization coupled with selective catalytic reduction [10] are popular for the simultaneous NOx and SO2 removal. Although those conventional technologies play an important role in the NOx and SO2 removal, they are costly and generate secondary pollutants. Biotechnologies e.g., biofiltration have been regarded as a potential cost-effective alternative to conventional air pollution control technologies due to its low cost, easy operation and management, and low risk in secondary pollutant generation [11]. In the past

∗ Corresponding author. E-mail addresses: [email protected], [email protected] (J. Chen). http://dx.doi.org/10.1016/j.bej.2016.06.027 1369-703X/© 2016 Elsevier B.V. All rights reserved.

decades, some research on the simultaneous biological removal of SO2 and NOx has been reported [12,13]. Han et al. [14] has proved that under the anaerobic condition, the co-cultured bacteria had a better performance than the solely denitrifying microorganisms in the combined NO/SO2 biodegradation process. The research also showed that the simultaneous removal of NO and SO2 from flue gas by liquid-phase catalytic oxidation-biological purification performed well under certain conditions. Most of the above work presented a low NOx removal efficiency due to the limitation of the NO mass transfer, which retarded the industrial application of simultaneous removal of NO and SO2 by a biotrickling filters [15,16]. It has been confirmed that FeII (EDTA) can significantly enhance the NO mass transfer rate [17–21]. FeII (EDTA) can chelate NO into the solution to form FeII (EDTA)-NO (Eq.(1)). On the other hand, SO2 can dissolve in the solution and produces SO3 2− and SO4 2− . These compounds can be reduced to sulfide (S2− ) by sulfate reducing bacteria (SRB) in the anaerobic bioreactor (Eq.(2)) [22]. The produced sulfide can be used as electron donor by the autotrophic denitrifying bacteria to reduce FeII (EDTA)-NO, thus regenerating the absorbent FeII (EDTA) to sustain the continuous NO removal (Eqs.(3) and (4)) [23]. Wang et al. [24] has established a simultaneous absorption of NO and SO2 by FeII (EDTA) combined with Na2 SO3 solution without biofilter on the purpose of investigation

88

J. Chen et al. / Biochemical Engineering Journal 114 (2016) 87–93

on the reaction products and mechanism, the majority consumed SO2 were converted into SO4 2− and the NO was reduced to N2 O. Chen et al. [25] have found that based on NO and SO2 absorption into FeII (EDTA) solution in a scrubber combined with biological reduction, more than 87% FeIII (EDTA) and 98% FeII (EDTA)-NO could be reduced. NO(aq) + FeII (EDTA) ↔ FeII (EDTA) − NO water

SRB

(2)

−

4Fe (EDTA) − NO + HS → 4Fe

II

(EDTA) + 2N2 + SO42−

+H

1 Fe (EDTA) − NO + HS → Fe (EDTA) + N2 + S 0 + OH − 2 II

−

II

6

+

This work aimed at demonstrating the feasibility of the simultaneous removal of SO2 and NO with FeII (EDTA) as a solvent in a rotating drum biofilter (RDB). The parametric tests including the inlet SO2 and NO concentration, empty bed residence time (EBRT), and O2 concentration was also investigated to their impacts on the bioreactor performance. Additionally, the microbial community in the RDB was analyzed to figure out the dominant bacteria pertained to the biological denitrification and desulfurization. 2. Materials and methods 2.1. Chemicals NO (5% in N2 , v/v), N2 (99.999%), O2 (99.999%), and SO2 (0.5% in N2 , v/v) were obtained from Zhejiang Jingong Co, China. Disodium ethylenediaminetetraacetate (Na2 EDTA, Titriplex) FeCl2 ·4H2 O (99.5%), FeCl3 ·6H2 O (99.5%), d-glucose (99.5%) were obtained from Shanghai Chemical Reagent Co. (Shanghai, China). All other chemicals were analytical-grade reagents, commercially available and used without further purification. The FeII (EDTA) solution was prepared with equal concentration (40 mM) of Na2 EDTA and FeCl2 . The solution pH was adjusted with 0.1 mM phosphate buffer (pH 7.0). The FeII (EDTA)-NO complex solution was prepared by bubbling NO through the ferrous EDTA solution until full breakthrough of NO was observed (The inlet and outlet concentration of NO was measured to be equal with a NOx analyzer) in the effluent. The prepared solution was kept in glass serum vials under N2 positive pressure in order to avoid the oxidation of the ferrous EDTA in the solution. The FeII (EDTA)-NO saturated solution exhibited negligible break down over a period of one week under a blanket of inert gas. 2.2. Source of biomass and media composition A concentrated active sludge was collected from a secondary sedimentation tank in the Hangzhou Qige Wastewater Treatment Plant, China. After 2-week culture in 8 L liquid medium (containing 500 mg/L Na2 SO4 , 500 mg/L KNO3 , and 3500 mg/L C3 H5 O3 Na), the enriched sludge was inoculated to the RDB. 500 mL of nutrient medium was replaced manually once a day from the lumen of the rotating drum. The pH value of medium was maintained at 7.0 and temperature was 30 ± 5 ◦ C. The culture media includes 3500 mg/L C3 H5 O3 Na , 500 mg/L NaCl, 2000 mg/L K2 HPO4 , 600 mg/L MgSO4 ·7H2 O, 500 mg/L NaHCO3 and 1 mL/L of trace elements solution (CuSO4 , 1.0 g/L; FeSO4 , 1.0 g/L; MnSO4 , 5.0 g/L; Na2 MoO4 , 1.0 g/L; and ZnCl2 , 2.0 g/L). 2.3. Start-up of the RDB As shown in Fig. 1, the RDB treatment system was composed of gas supply section, an inspection section, and the main unit of

8

16 1 2 3

10 12

(3) (4)

13

7 4

(1)

SO2 → SO4 2− /SO3 2− →S 2− II

5

9

10

14

11 15

17

Fig. 1. Schematic representation of the laboratory-scale system. (1) N2 cylinder, (2) O2 cylinder, (3) NO cylinder, (4) digital mass flowmeter, (5) rotatmeter, (6) gas mixed container, (7) pressure meter, (8) dryer, (9) flue gas analyzer, (10) metering pump, (11) nutrient solution, (12) motor, (13) packing material, (14) tail gasabsorber, (15) nutrients reservoir, (16) SO2 cylinder, (17) pH control system.

the RDB. The start-up of the RDB was initiated with 2 L enriched sludge and 8 L culture media with 100–1000 mg/L sulfate and 100–1000 mg/L nitrate. The nutrient liquids promoted the growth of autotrophic bacteria and the rotating drum provided a sufficient contact between the packing layer and the nutrient liquids for the bacteria attachment. The effect of pH on the removal efficiency of NO and SO2 was investigated as show in Fig. S1, and the optimal pH was neutral or slightly alkaline for the growth of bacteria and enzyme activity. Hence, the pH value was adjusted to around 7.0, and the sample was taken to monitor the removal efficiency of SO4 2− and NO3 − . A certain concentration of SO2 (∼389 mg/m3 ) and NO (∼417 mg/m3 ) balanced by the N2 was supplied to the RDB instead of the SO4 2− and NO3 − . The NO and SO2 were absorbed by the solvent and the absorbed NO and SO2 were biologically reduced by the bacteria in the biofilm. During the RDB operation, the nutrient was cautiously added at a certain rate to provide the adequate nutrition for the growth of bacteria. 2.4. Experimental procedures After the start-up period, the performance of the simultaneous removal of SO2 and NO in the RDB was evaluated. In a typical test, the pH value was maintained at 7.0 and temperature was 30 ± 5 ◦ C. The inlet concentration of SO2 and NO was kept at around 2000 mg/m3 and 800 mg/m3 , respectively. In addition, the effects of SO2 concentration (∼2500 mg/m3 ), NO concentration (180–1200 mg/m3 ), EBRT (0.5–7 min), and O2 concentration (∼8 vol%) on the performance of the RDB were also investigated. Three samples named as S1, S2, and S3 were collected from the bottom of nutrient solution to investigate the evolution of microbial community after the start-up period, the SO2 inlet concentration investigation, and the completion of all the parametric tests, respectively. On top of that, after the completion of all the parametric tests, another three samples were collected along the radius of layers (external radius, median radius and inner radius named as R1, R2 and R3) to investigate the distribution of microbial community along the packing. 2.5. DNA extraction, amplification, and high-throughput sequencing Genomic DNA was extracted from the cells according to the instruction of the DNA isolation kit obtained from Shanghai Biotechnology Co., Ltd. Purified DNA was used as a template for PCR amplification with high-fidelity DNA polymerase. The 16 S rRNA genes were amplified using the universal primers 341F:

J. Chen et al. / Biochemical Engineering Journal 114 (2016) 87–93 110 100

3 200mg/m SO 2

2.6. Analytical methods

3 II 400mg/m SO +10mM Fe EDTA 2

3 400mg/m SO 2

The NO concentration was measured by a Model 42CHL NONO2 -NOx analyzer (0–5000 mg/m3 , Thermo Electron Co., USA). The SO2 concentration was measured by a flue gas analyzer (ecom-EN2, GER). NO3 − , SO4 2− and S2− concentrations were analyzed by ion chromatography (DX-500, Dionex Corporation) with a column of Dionexionpac AS14. The concentration of FeII (EDTA)-NO was measured by a model UV-2000 spectrophotometer (UNICO (Shanghai) Instruments Co., Ltd.). The concentration of FeII (EDTA) in solution was determined by a modified 1,10-phenanthroline colorimetric method at 510 nm. The FeIII (EDTA) concentration was calculated by the difference between total Fe and FeII . High-throughput sequencing technique was performed by MajorBio Pharmaceutical Technology Co., Ltd. (Shanghai, China). All the data shown in this paper were the mean values of duplicate or triplicate experiments.

The removal efficiency (%)

90 80

a

70 60 50

The NO removal efficiency The SO2 Removal efficiency

40 30 20 0

10

20

30

40

50

60

3. Results and discussion

Time (day)

3.1. The NO and SO2 simultaneous removal efficiency

4.5 4.0

b

4.0

3.5 3.0

3.0

The Fe EDTA-NO concentration The Fe EDTA concentration

2.5

2.5 2.0

The S concentration

2.0

1.5

1.5

1.0

2-

3.5

The concentration of S ( mM)

II

III

The concentration of Fe EDTA-NO, Fe EDTA ( mM)

4.5

0.5

1.0

0.0

0.5 30

35

40

45

50

55

60

89

-0.5

Time (day) Fig. 2. The SO2 and NO removal efficiency in the start-up period (a) and the change of FeII (EDTA)-NO, FeIII (EDTA) and S2− concentration in liquid phase (b). (T = 30 ± 5 ◦ C, [FeII (EDTA)-NO] = 0–10 mM, EBRT:1.8 min, [O2 ] = 2vol%, [SO2 ] = 200–400 mg/m3 , [NO] = 400 mg/m3 , Vgas :8 L/min, pH:7.0).

3 -CCCTACACGACGCTCTCCGATCTGCCTACGGGNGGCWGCAG-5 and 805R: 3 -GACTGGAGTTCCTTGGCACCCGAGAATTCCAGACTA CHVGGGTATCTAA TCC-5 . Both primers were modified to contain an Illumina adapter region (bold), and the forward primer was encoded with a 12 bp barcode for multiplex sequencing. The thermal PCR conditions were as follows: initial denaturation at 94 ◦ C for 5 min and 30 cycles of denaturation at 94 ◦ C for 1 min, annealing at 54 ◦ C for 1 min, and an extension at 72 ◦ C for 3 min, followed by a final extension at 72 ◦ C for 10 min. After amplification, PCR products were analyzed through agarose gel electrophoresis. The obtained data were optimized by removing low-quality sequences, unrecognized reverse primers, and any ambiguous base calls, with a length <200 bp. High-quality sequences were clustered into 97% similarity operational taxonomic units (OTUs) using UCLUST software. A representative sequence from each OTU was classified and phylogenetically assigned to a taxonomic identity (phylum and genus levels) using RDP classifier. Shannon diversity indices and species richness estimators were generated for each sample using Quantitative Insights Into Microbial Ecology (QIIME) pipeline version 1.3.0 (http://qiime.org) [26].

During the start-up period, a relatively low NO and SO2 inlet concentration of 200 mg/m3 to 400 mg/m3 was fed until steady-state operation was achieved. Fig. 2a showed that the after 30-day continuous operation, the NO removal efficiency stabilized at 57.1%, and the SO2 removal efficiency achieved at 90.3%, suggesting that the biofilm was formed in the RDB. After the addition of 10 mM FeII (EDTA), the removal efficiency of NO dramatically increased from 57.1% to 76.5%, and finally stabilized at around 93% in the following 30-day continuous operation. However, the removal efficiency of SO2 was not noticeably changed and maintained at around 90%. After the addition of 10 mM FeII (EDTA) (day 30–60), the change of FeII (EDTA)-NO, FeIII (EDTA), and S2− concentration in liquid phase were investigated to confirm the feasibility of simultaneous desulfurization and denitrification with the FeII (EDTA). As shown in Fig. 2b, the concentration of FeII (EDTA)-NO decreased in the first 5 days. After that, it gradually increased and finally reached at a stable level of ∼1.87 mM. Compared with the FeII (EDTA)-NO, an opposite tendency of the FeIII (EDTA) concentration was observed. It increased in the first 5 days and then gradually decreased to 0.63 mM. The concentration of S2− was kept at around 3.5 mM during the whole period. The addition of FeII (EDTA) facilitated the absorption of NO. However, part of the FeII (EDTA) would be oxidized to FeIII (EDTA) in the presence of oxygen. The absorbed NO was biologically reduced to harmless N2 . In the meantime, the absorbed SO2 was converted into SO3 2− and SO4 2− , and then was reduced into S2− by the SRB [27], which could help regenerate the FeII (EDTA) in the solution to sustain the continuous NO removal as shown in Eqs. (3) and (4). It has been reported that some autotrophic denitrification bacteria can use sulfide as electron donor to reduce FeII (EDTA)-NO, nitrite, and nitrate [23]. The reaction equations can be expressed as following [28–30]: 3HS − + 8NO2− + 5H + → 3SO42− + 4N2 + 4H2 O

(5)

3HS − + 2NO2− + 5H + → 3S + N2 + 4H2 O

(6)

In fact, dissimilatory SRB has been recognized to be able to enzymatically reduce ferric iron (FeIII ), especially water soluble ferric iron [31]. Chen et al. [25] has found that Desulfovibrio sp. CMX (SRB) can biologically reduce FeII (EDTA)-NO and FeIII (EDTA) simultaneously. Meanwhile, FeII (EDTA)-NO can be chemically reduced by sulfide with a molar ratio of 1:3.4-1:3.6 [32]. In this study, as shown in Fig. 2a, after adding FeII (EDTA) into the nutrient solution on the 30th day, the NO removal efficiency increased drastically

The removal efficiency of SO2 and NO (%)

activated sludge

activated sludge 28–33 min

108s

T. denitrificans, named D4 36s

SRB 76s

30.28s

EBRT

Anaerobic denitrifying bacteria

J. Chen et al. / Biochemical Engineering Journal 114 (2016) 87–93

Culture

90

95

90

85

The SO2 removal efficiency 80

88.9 99.9 90 95 78.3 99 80.85 100 93 98.5

RE(%)

75

The NO removal efficiency

0

500

1000

1500

2000

2500

3

The inlet SO2concentration (mg/m )

A rotating drum biofilter

A tandem twin-towersdesulfurization and denitrification process system

A continuous stirred tank reactor with a three-phase separator

bio-trickling filter

Bioreactor

bio-trickling filter

from 57.1% to 76.5% due to the enhancement in mass transfer of NO by FeII (EDTA). After about another 10 days’ operation, autotrophic denitrification bacteria were acclimated to use FeII (EDTA)-NO as electron acceptor for sulfide oxidation, and the removal efficiency became stable and finally achieved at 93%. In Fig. 2b, the concentration of FeII (EDTA)-NO, FeIII (EDTA) and S2− gradually reached a steadiness after the 50th day. This demonstrated that the reduction of SO2 into S2− by SRB and the reduction of FeII (EDTA)-NO by autotrophic denitrification bacteria using S2− as electron donor had generally balanced. Overall, 30-day continuous operation with relatively high level removal efficiency of SO2 and NO demonstrated the feasibility of simultaneous desulfurization and denitrification coupled with complexing absorption. The maximal removal efficiency of SO2 and NO could be achieved at 98.5% and 93%, respectively, in the conditions of 2000 mg/m3 SO2 , 800 mg/m3 NO, 1.8 min of EBRT, and 2 vol% oxygen. The maximum treating capacity for this system was 78.971 g/(m3 ·h) of SO2 and 29.825 g/(m3 h) of NO. A comparison between our work and other process for the simultaneous removal of NO and SO2 was shown in Table 1, which indicated that the proposed process in this study could achieve a higher removal efficiency of SO2 and NO compared to the reported data.

This work

[35]

[16]

NO 20 mg/m3 SO2 710 mg/m3 NO 2.01 g/m3 SO2 2.02 g/m3 NO 232 mg/L SO2 300 mg/L NO 2500 ∼ 3000 mg/m3 SO2 1500 ∼ 2000 mg/m3 NO 2000 mg/m3 SO2 800 mg/m3 [40]

Inlet gas concentration

[39]

3.2. The effect of inlet SO2 concentration

Reference

Table 1 Comparison of the NO and SO2 simultaneous removal efficiency of different bioreactors.

Fig. 3. The effect of initial SO2 concentration on the SO2 and NO removal efficiency. (T = 30 ± 5 ◦ C, [FeII (EDTA)-NO] = 10 mM, EBRT:1.8 min, [O2 ] = 2vol%, [SO2 ] = 0–2500 mg/m3 , [NO] = 800 mg/m3 , Vgas :8 L/min, pH:7.0).

Parametric tests were investigated after the start-up period. Fig. 3 showed that the removal efficiency of SO2 roughly decreased with the increase of inlet concentration of SO2 . When the inlet concentration of SO2 was below 400 mg/m3 , the removal efficiency was above 95%. The removal efficiency of SO2 dropped to 86.6% as the inlet concentration increased to 2475 mg/m3 . In general, the solubility of SO2 in water was high at room temperature. It was easy for microbes in RDB to remove SO2 at a relatively low concentration. As the SO2 concentration increased, the SO2 removal was limited by the activity of the SRB, resulting in the decrease of SO2 removal efficiency. When the SO2 inlet concentration was under 1231 mg/m3 , the removal efficiency of NO increased with the increase of SO2 inlet concentration. A certain increase of SO2 concentration led to the increase of S2− concentration in liquid phase, which could serve as electron donor for reducing FeII (EDTA)-NO and hence promoted the removal of NO. However, with the further accumulation of S2− , the activity of autotrophic denitrification bacteria was inhibited [23],

J. Chen et al. / Biochemical Engineering Journal 114 (2016) 87–93

100

The removal efficiency of SO2 and NO (%)

The removal efficiency of SO2 and NO (%)

100

91

95 90 85 80 75 70

The NO removal efficiency The SO2removal efficiency

65 60 0

200

400

600

800

1000

95 90

The SO2 removal efficiency 85

The NO removal efficiency

80 75 70 65

1200

0

1

2

3

4

5

6

7

8

Empty bed retention time (min)

3

Inlet concentration of NO(mg/m )

resulting in a decline of NO removal efficiency. Further increase in the inlet SO2 concentration resulted in a decrease of the NO removal efficiency. The highest removal efficiency of NO was achieved at 91.9% at the optimal SO2 inlet concentration of 1231 mg/m3 . 3.3. The effect of inlet NO concentration As shown in Fig. 4, the SO2 removal efficiency increased steadily from 78% to 98% with the increase of inlet NO concentration. When the inlet NO concentration was higher than 780 mg/m3 , the SO2 removal efficiency could sustain above 95%. The increase of inlet NO concentration led to the increase of FeII (EDTA)-NO concentration. FeII (EDTA)-NO can oxidize S2− in both chemical and biological ways [23,32], which benefits the reduction of dissolved SO2 by SRB. Therefore, as shown in Fig. 4, the removal efficiency of SO2 increased with the increase of inlet NO concentration. Conversely, the NO removal efficiency decreased dramatically when the inlet NO concentration was higher than 780 mg/m3 . It dropped to 62% at the inlet NO concentration of 1200 mg/m3 . That may be because high NO levels were associated with irreversible DNA fragmentation and cell death, and potentially impaired electron transport [33,34]. In the NO reduction system, the FeII (EDTA) was important to NO reduction, which was not only contribute to mass transferring of NO from the gas to liquid, but also was the direct electron donor for NO bioreduction. The quantitative FeII (EDTA) was the restrictive factor for the more NO reduction. 3.4. The effect of EBRT Fig. 5 showed that increasing EBRT could improve both SO2 and NO removal efficiency by prolonging contact and biological reduction time in the biofilm. The SO2 removal efficiency increased dramatically from 67.9% to 96.5% and the NO removal efficiency increased sharply from 73.1% to 92.7%, while the EBRT increased from 0.5 min to 1.8 min. With further increase in the EBRT from 1.8 to 7.2 min, the removal efficiency increased slowly and achieved at 99.5% and 97.5% for SO2 and NO, respectively. EBRT affects the contact time of gaseous pollutants with the RBD and thereby affects the biodegradation time. Theoretically, the increase in the EBRT can improve the removal efficiency of SO2 and NO. However, the longer EBRT means the larger engineering cost. Thus, an appropriate EBRT was always necessary in engineering application. As shown in Fig. 5,

Fig. 5. The effect of EBRT on the NO and SO2 removal efficiency. (T = 30 ± 5 ◦ C, [FeII (EDTA)-NO] = 10 mM, EBRT:0.5-7.2 min, [O2 ] = 2vol%, [SO2 ] = 2000 mg/m3 , [NO] = 800 mg/m3 , Vgas :1.67–24 L/min, pH:7.0).

The removal efficiency of SO2 and NO (%)

Fig. 4. The effect of initial NO concentration on the SO2 and NO removal efficiency. (T = 30 ± 5 ◦ C, [FeII (EDTA)-NO] = 10 mM, EBRT:1.8 min, [O2 ] = 2vol%, [SO2 ] = 2000 mg/m3 , [NO] = 180–1200 mg/m3 , Vgas :8 L/min, pH:7.0).

100

The SO2 removal efficiency The NO removal efficiency 90

80

70

60 0

2

4

6

8

The content of O2 (%) Fig. 6. The effect of O2 content on the NO and SO2 removal efficiency. (T = 30 ± 5 ◦ C, [FeII (EDTA)-NO] = 10 mM, EBRT:1.8 min, [O2 ] = 2-8vol%, [SO2 ] = 2000 mg/m3 , [NO] = 800 mg/m3 , Vgas :8L/min, pH:7.0).

the optimal EBRT of 1.8 min was achieved for further application of RDB. 3.5. The effect of O2 concentration Since flue gas may contain a certain concentration of O2 [35], the effect of O2 on microbial process for SO2 /NO removal is an important consideration which is another major factor affecting the metabolism of SRB and NR. A rise in O2 concentration within a certain range may inhibit the enzyme activity resulting in a decline of removal efficiency. Moreover, O2 can oxidize FeII (EDTA) into FeIII (EDTA) which may reduce the absorbent concentration and then affect the removal efficiency of NO. Fig. 6 showed that the removal efficiency of both SO2 and NO decreased with the increase of O2 concentration. As the O2 concentration was below 2 vol%, the removal efficiency of SO2 and NO sustained above 96.5% and 92.7%, respectively. The removal efficiency of SO2 dropped quickly to 59.7% and NO dropped to 70.1% when the O2 concentration reached at 8 vol%. The primary concern is the effect of O2 on SRB, which are sensitive to the redox potential of the medium. It was

92

J. Chen et al. / Biochemical Engineering Journal 114 (2016) 87–93

Fig. 7. The microbial community analysis in the RBD at genus level. (S1, S2, and S3 were samples after the start-up period, the SO2 inlet concentration investigation, and the completion of all the parametric tests, respectively. R1, R2 and R3 were samples from external radius, median radius and inner radius, respectively).

reported that Desulfovibrio desulfuricans were tolerant of up to 1.3 vol% O2 in the feed gas [23]. As shown in Fig. 6, at low concentration of 2 vol% O2 , the removal efficiency of SO2 remained at high level, which might attribute to the fact that the facultative anaerobic non-SRB heterotrophs in the co-cultivation system scavenge O2 kept the redox potential sufficiently low to favor the growth of SRB (and SO2 reduction). However, further increase in O2 concentration resulted in the inhibition of SO2 reduction. The influence of O2 on NO removal was mainly due to the oxidation of FeII (EDTA) to FeIII (EDTA). Additionally, the inhibition of SRB causing decrease of S2− concentration (electron donor) might affect the reduction of FeII (EDTA)-NO by autotrophic denitrification bacteria and thus retarding the NO removal.

4. Conclusion This work demonstrated the feasibility of the new process with FeII (EDTA) for the simultaneous removal of SO2 and NO in the RDB. The SO2 and NO removal efficiency could reach as high as 98.5% and 93%, respectively, in the conditions of 2000 mg/m3 SO2 , 800 mg/m3 NO, 1.8 min of EBRT, and 2 vol% oxygen. Preliminary microbial community analysis confirmed that Pseudomonas and Desulfovibrio were dominated and responsible for the biological denitrification and desulfurization, respectively, in the RDB. Acknowledgements The research was supported by the National Natural Science Foundation of China (No 21277125) and Program for Changjiang Scholars and Innovative Research Team in University. Appendix A. Supplementary data

3.6. Microbial community profile in the RDB Analysis of microbial community diversity was conducted by high-throughput sequencing technique. As shown in Fig. 7, the dominant genera in the RDB during the whole operating process were Pseudomonas, Spirochaeta, Synergistaceae uncultured, Clostridiales uncultured, Desulfovibrio, Halomonas, Bacillus, and Fastidiosipila for all the analyzed samples. Among them, the primary NR was Pseudomonas. Along the radius of layers from the external (R1) to inner (R3), Pseudomonas accounted for 35.7%, 27.2%, 23.1% respectively. Pseudomonas was reported as colorless sulfur bacteria, which could use reduced sulfur compounds (H2 S, HS− , S2− ) as electron donor and CO/CO2 as carbon source to reduce O2 and nitrate [36,37]. The dominant SRB was Desulfovibrio. The difference in the abundance of the Desulfovibrio was not noticeable between the samples after the start-up period (S1, 4.5%) and after the completion of the parametric tests (S3, 4.6%). The highest amount (6.1%) of the Desulfovibrio was achieved after the SO2 concentration investigation. Desulfovibrio is the primary SRB in wastewater treatment [38].

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.bej.2016.06.027. References [1] K. Skalska, J.S. Miller, S. Ledakowicz, Trends in NOx abatement: a review, Sci. Total Environ. 408 (2010) 3976–3989. [2] J.E. Ogden, P.K. Moore, Inhibition of nitric-oxide synthase − potential for a novel class of therapeutic agent, Trends Biotechnol. 13 (1995) 70–78. [3] A.A. Basfar, O.I. Fageeha, N. Kunnummal, S. Al-Ghamdi, A.G. Chmielewski, J. Licki, A. Pawelec, B. Tyminski, Z. Zimek, Electron beam flue gas treatment (EBFGT) technology for simultaneous removal of SO2 and NOx from combustion of liquid fuels, Fuel 87 (2008) 1446–1452. [4] E.S. Kikkinides, R.T. Yang, Simultaneous SO2 /NOx removal and SO2 recovery from flue-gas by pressure swing adsorption, Ind. Eng. Chem. Res. 30 (1991) 1981–1989. [5] Y.P. Mao, W. Bi, X.L. Long, W.D. Xiao, W. Li, W.K. Yuan, Kinetics for the simultaneous absorption of nitric oxide and sulfur dioxide with the hexamminecobalt solution, Sep. Purif. Technol. 62 (2008) 183–191. [6] W. Nimmo, A.A. Patsias, E. Hampartsoumian, B.M. Gibbs, P.T. Williams, Simultaneous reduction of NOx and SO2 emissions from coal combustion by calcium magnesium acetate, Fuel 83 (2004) 149–155.

J. Chen et al. / Biochemical Engineering Journal 114 (2016) 87–93 [7] J.J. Kaczur, Oxidation chemistry of chloric acid in NOx /SOx and air toxic metal removal from gas streams, Environ. Prog. 15 (1996) 245–254. [8] Z.C. Guo, Y.S. Xie, I.Y. Hong, J.Y. Kim, Catalytic oxidation of NO to NO2 on activated carbon, Energy Convers. Manage. 42 (2001) 2005–2018. [9] K.X. Li, L.C. Ling, C.X. Lu, W.M. Qiao, Z.Y. Liu, L. Liu, I. Mochida, Catalytic removal of SO2 over ammonia-activated carbon fibers, Carbon 39 (2001) 1803–1808. [10] Z.G. Huang, Z.P. Zhu, Z.Y. Liu, Q.Y. Liu, Formation and reaction of ammonium sulfate salts on V2 O5 /AC catalyst during selective catalytic reduction of nitric oxide by ammonia at low temperatures, J. Catal. 214 (2003) 213–219. [11] Y.M. Jin, M.C. Veiga, C. Kennes, Bioprocesses for the removal of nitrogen oxides from polluted air, J. Chem. Technol. Biotechnol. 80 (2005) 483–494. [12] B.N. Dasu, V. Deshmane, R. Shanmugasundram, C.M. Lee, K.L. Sublette, Microbial reduction of sulfur-dioxide and nitric-oxide, Fuel 72 (1993) 1705–1714. [13] K.H. Lee, K.L. Sublette, Simultaneous combined microbial removal of sulfur-dioxide and nitric-oxide from a gas-stream, Appl. Biochem. Biotechnol. 28-9 (1991) 623–634. [14] Y.Q. Han, W.J. Zhang, Simultaneous removal of NOx and SO2 in exhausted gas through landfill leachate, Indian. J. Exp. Biol. 48 (2010) 1237–1242. [15] A.J. Wang, D.Z. Du, N.Q. Ren, J.W. van Groenestijn, An innovative process of simultaneous desulfurization and denitrification by Thiobacillus denitrificans, J. Environ. Sci. Heal. A 40 (2005) 1939–1949. [16] X.C. Wang, X.Y. Bi, P.S. Sun, J.Q. Chen, P. Zou, X.M. Ma, J. Zhang, H.Y. Wang, X.Y. Xu, Effects of oxygen content on the simultaneous microbial removal of SO2 and NOx in biotrickling towers, Biotechnol. Bioprocess Eng. 20 (2015) 924–930. [17] P. van der Maas, P. van den Brink, S. Utomo, B. Klapwijk, P. Lens, NO removal in continuous BioDeNOx reactors: Fe(II)EDTA regeneration, biomass growth, and EDTA degradation, Biotechnol. Bioeng. 94 (2006) 575–584. [18] P. van der Maas, P. van den Bosch, B. Klapwijk, P. Lens, NO removal from flue gas by an integrated physicochemical absorption and biological denitrification process, Biotechnol. Bioeng. 90 (2005) 433–441. [19] J. Chen, Q.Z. Dai, H.F. Qian, Y.F. Jiang, J.M. Chen, Nitric oxide enhanced reduction in a rotating drum biofilter coupled with absorption by Fe(II)EDTA, J. Chem. Technol. Biotechnol. 88 (2013) 579–584. [20] S.H. Zhang, H. Chen, Y.F. Xia, N. Liu, B.H. Lu, W. Li, Current advances of integrated processes combining chemical absorption and biological reduction for NOx removal from flue gas, Appl. Microbiol. Biotechnol. 98 (2014) 8497–8512. [21] X.Y. Dong, Y. Zhang, J.T. Zhou, M.X. Chen, X.J. Wang, Z. Shi, Fe(II)EDTA-NO reduction coupled with Fe(II)EDTA oxidation by a nitrate- and Fe(III)-reducing bacterium, Bioresour. Technol. 138 (2013) 339–344. [22] I. Manconi, P. van der Maas, P.N.L. Lens, Effect of sulfur compounds on biological reduction of nitric oxide in aqueous Fe(II)EDTA2− solutions, Nitric Oxide-Biol Ch 15 (2006) 40–49. [23] C.W. Zielke, M.c. Wu, Method of Reducing Sulfur Oxide and Nitrogen Oxide Content of Flue Gas, Conoco Inc., Wilmington, Del US, 1986. [24] W. Li, W. Zhao, Z. Wu, Simultaneous absorption of NO and SO2 by Fe(II)EDTA combined with Na2 SO3 solution, Chem. Eng. J. 132 (2007) 227–232.

93

[25] M.X. Chen, J.T. Zhou, Y. Zhang, X.J. Wang, Z. Shi, X.W. Wang, Fe(III)EDTA and Fe(II)EDTA-NO reduction by a sulfate reducing bacterium in NO and SO2 scrubbing liquor, World J. Microb. Biotechnol. 31 (2015) 527–534. [26] Q. Wang, G.M. Garrity, J.M. Tiedje, J.R. Cole, Naive Bayesian classifier for rapid assignment of rRNA sequences into the new bacterial taxonomy, Appl. Environ. Microbiol. 73 (2007) 5261–5267. [27] L. Philip, M.A. Deshusses, Sulfur dioxide treatment from flue gases using a biotrickling filter − Bioreactor system, Environ. Sci. Technol. 37 (2003) 1978–1982. [28] C. Jing, Z. Ping, Q. Mahmood, Simultaneous sulfide and nitrate removal in anaerobic reactor under shock loading, Bioresour. Technol. 100 (2009) 3010–3014. [29] Q. Mahmood, P. Zheng, Y. Hayat, E. Islam, D. Wu, J. Ren-Cun, Effect of pH on anoxic sulfide oxidizing reactor performance, Bioresour. Technol. 99 (2008) 3291–3296. [30] Q. Mahmood, P. Zheng, J. Cai, D.L. Wu, B.L. Hu, J.Y. Li, Anoxic sulfide biooxidation using nitrite as electron acceptor, J. Hazard. Mater. 147 (2007) 249–256. [31] Y.L. Li, H. Vali, J. Yang, T.J. Phelps, C.L. Zhang, Reduction of iron oxides enhanced by a sulfate-reducing bacterium and biogenic H2 S, Geomicrobiol. J. 23 (2006) 103–117. [32] X.J. Wang, Y. Zhang, X.Y. Dong, M.X. Chen, Z. Shi, J.T. Zhou, Fe(II)EDTA-NO reduction by sulfide in the anaerobic aqueous phase: stoichiometry and kinetics, Energy Fuel 27 (2013) 6024–6030. [33] M.C. Pedroso, J.R. Magalhaes, D. Durzan, Nitric oxide induces cell death in Taxus cells, Plant Sci. 157 (2000) 173–180. [34] K.S. Gould, O. Lamotte, A. Klinguer, A. Pugin, D. Wendehenne, Nitric oxide production in tobacco leaf cells: a generalized stress response, Plant Cell Environ. 26 (2003) 1851–1862. [35] X.C. Wang, X.Y. Bi, P.S. Sun, J.Q. Chen, P. Zou, X.M. Ma, J. Zhang, H.Y. Wang, X.Y. Xu, Effects of oxygen content on the simultaneous microbial removal of SO2 and NOx in biotrickling towers, Biotechnol. Bioprocess Eng. 20 (2015) 924–930. [36] B. De Gusseme, P. De Schryver, M. De Cooman, K. Verbeken, P. Boeckx, W. Verstraete, N. Boon, Nitrate-reducing, sulfide-oxidizing bacteria as microbial oxidants for rapid biological sulfide removal, FEMS Microbiol. Ecol. 67 (2009) 151–161. [37] E.Y. Chernousova, E.V. Belousova, E.Y. Gavrish, G.A. Dubinina, T.P. Tourova, M.Y. Grabovich, Molecular phylogeny and taxonomy of colorless, filamentous sulfur bacteria of the genus Thiothrix, Microbiology 81 (2012) 332–341. [38] H. Yu, C. Chen, J.C. Ma, X.J. Xu, R.G. Fan, A.J. Wang, Microbial community functional structure in response to micro-aerobic conditions in sulfate-reducing sulfur-producing bioreactor, J. Environ. Sci. China 26 (2014) 1099–1107. [39] Z. Xie, Simultaneous desulfurization and denitrification by a biological method, J. Environ. Eng. China (2009). [40] Y.Q. Han, W.J. Zhang, J. Xu, A performance study of simultaneous microbial removal of NO and SO2 in a biotrickling-filter under anaerobic condition, Braz. J. Chem. Eng. 28 (2011) 189–196.