key: cord-0827727-wrf62y23
authors: Liu, Jianwei; Kang, Xinyue; Liu, Xueli; Yue, Peng; Sun, Jianbin; Lu, Chen
title: Simultaneous removal of bioaerosols, odors and volatile organic compounds from a wastewater treatment plant by a full-scale integrated reactor
date: 2020-07-04
journal: Process Saf Environ Prot
DOI: 10.1016/j.psep.2020.07.003
sha: 5cda744b8bdda893a33ed8179a6d9e119b7671bd
doc_id: 827727
cord_uid: wrf62y23

Biological control of odors and bioaerosols in wastewater treatment plants (WWTPs) have gained more attention in recent years. The simultaneous removal of odors, volatile organic compounds (VOCs) and bioaerosols in each unit of a full-scale integrated-reactor (FIR) in a sludge dewatering room was investigated. The average removal efficiencies (REs) of odors, VOCs and bioaerosols were recorded as 98.5%, 25 94.7% and 86.4%, respectively, at an inlet flow rate of 5760 m(3)/h. The RE of each unit decreased, and the activated carbon adsorption zone (AZ) played a more important role as the inlet flow rate increased. The REs of hydrophilic compounds were higher than those of hydrophobic compounds. For bioaerosols, roughly 35% of airborne heterotrophic bacteria (HB) was removed in the low-pH zone (LPZ) while over 30% of total fungi (TF) was removed in the neutral-pH zone (NPZ). Most bioaerosols removed by the biofilter (BF) had a particle size larger than 4.7 μm while bioaerosols with small particle size were apt to be adsorbed by AZ. The microbial community in the BF changed significantly at different units. Health risks were found to be associated with H(2)S rather than with bioaerosols at the FIR outlet.

al., 2020; Liu et al., 2020; Patients et al., 2020; Santarpia et al., 2020; Tada et al., 2019; Yao et al., 2020) . A variety of bioaerosols, odors and VOCs volatilize from WWTPs during biological metabolism in wastewater treatment processes involving raw wastewater collection, aeration and sludge treatment facilities, such as pump houses, screen rooms, grit chambers, aeration tanks and sludge dewatering rooms. (Noh et al., 2019; Sánchez-Monedero et al., 2008) . The most significant air pollutants emitted by WWTPs are nitrogen compounds including ammonia (NH3); sulfur compounds including hydrogen sulfide (H2S) and mercaptans, bioaerosols, organic acids, chloride, alkanes and aromatics (Kasperczyk et al., 2019; . Generally, odorous compounds in WWTPs are mostly volatile, corrosive and irritating malodorous nuisances with very low odor thresholds (Alinezhad et al., 2019) . Therefore, as sites of pathogenic microorganism accumulation, WWTPs should not only meet conventional requirements for removing pollutants in wastewater, but must also be able to effectively eliminate air pollution for public health and environmental protection. Air pollutants produced by WWTPs have thus become a central public concern.

Various technologies have been proposed and applied for the treatment of odorous components and can be classified into physical/chemical systems (e.g., chemical scrubbers, incinerators, activated carbon adsorption, and ozone oxidation) and biological systems (e.g., biofiltration, biotrickling filtration, bioscrubbing, and activated sludge diffusion technologies) (Arellano-García et al., 2018; Gabriel et al., 2013; Qiu and Deshusses, 2017) . Each of these technologies have advantages and disadvantages and specific ranges of economic application (Mudliar et al., 2010) . Compared to physicochemical treatments, biological methods are less costly, no secondary waste generation, involve mild process conditions and are ecologically clean (Bindra et al., 2015; Rene et al., 2006) . Among them, BFs has proved to be some of the most promising tools for odorous compound treatment in WWTPs due to being compact and moderately priced and well suited to applications requiring operational flexibility. (Alinezhad et al., 2019; Lebrero et al., 2011; Lewkowska et al., 2016) . Moreover, BFs have been evaluated as means to control airborne microorganisms released in WWTPs (Sanchez-Monedero et al., 2003) , as when filled with high density surface area filler they may intercept microorganisms in air inlets. Sanchez-Monedero et al. (2003) found BFs to reduce concentrations of Aspergillus fumigatus and mesophilic bacteria by more than 90% and 39% J o u r n a l P r e -p r o o f on average, respectively. Due to its high purification efficiency for some recalcitrant substrates and provision of large surface areas for microorganisms to attach to, activated carbon adsorption is considered a reliable technology for air pollutant removal with several advantages for WWTP application (Bansode et al., 2003; Lillo-Ródenas et al., 2005) . Li et al. (2011) found that over 85% of airborne bacteria and fungi emitted from an oxidation ditch were taken up by activated carbon within 80 h of contact time.

However, while the removal performance of BFs is generally satisfactory under moderate conditions, it is reduced significantly when BFs are applied to undiluted recalcitrant waste gases at large quantities. The complex mixture of odorous compounds in WWTPs poses a special challenge to the use of BFs. The accumulation of sulfuric acid during the oxidation of H2S processes reduces both pH levels microbial activity and ultimately affects treatment efficiency. Therefore, it is difficult to remove pollutants with different properties by using a single conventional biological or physical/chemical approach. To overcome these shortcomings, new technologies such as two-stage BFs, integrated-bioreactors, and tricklebed bioreactors have been developed. Chung et al. (2007) reported that a two-stage BF can indeed improve NH3 removal from waste gases containing high H2S concentrations. As multiple pollutants are concurrently degraded in a BF, antagonistic, neutral, and synergistic interactions may occur between pollutants (Mohammad et al., 2017; Strauss et al., 2004) .

Two-phase or two-stage BFs are considered as means to enhance removal performance and alleviate antagonistic substrate interactions .

Despite numerous studies conducted on odors and VOCs, few studies have addressed the simultaneous removal of bioaerosols, odors and VOCs in contaminated gas streams. On this basis, a FIR is used in this study to simultaneously remove multiple pollutants from the sludge dewatering room of a WWTP in Beijing. The FIR includes a two-phase BF and an activated carbon AZ. The first BF phase is suspended phase (SP), which includes suspended microorganisms that remove some hydrophilic compounds and humidify the air stream. The second phase is the immobilized phase (IP), which is divided into the LPZ and NPZ, which are filled with polyurethane (PU) and volcanic rock forming a fixed biofilm, respectively. The LPZ was operated at pH 4-5 and dominated by acidophilic microorganisms mainly to remove sulfide. Different species of microbes live in suitable environments to degrade target J o u r n a l P r e -p r o o f 6 pollutants in each unit. Due to a large number of microorganisms attached to the packing medium, BF bioaerosols removal performance is affected by an increasing of air flow rate.

The activated carbon AZ is set after the BF for the thorough removal of bioaerosols and recalcitrant substrates and to maintain stable FIR removal. Based on emission and composition characteristics of exhaust in the WWTP, the treatment performance, long-term operational stability, and microbial community in the FIR for the simultaneous removal of multiple pollutants were investigated.

The FIR was constructed to treat sludge dewatering exhaust in a WWTP in Beijing ( Fig.   1 ), China. Nearly 2.0 × 10 5 m 3 /day of domestic wastewater was treated in the WWTP based on an anaerobic-anoxic-oxic process. The exhaust control system includes a gas collection unit, transmission pipeline and FIR. Sludge dewatering room exhaust was collected by pipes and introduced into the FIR treatment system through a blower. The two-phase BF composited a fiberglass cuboid with a height of 5.6 m, base area of 12 m, and total packed volume of 52.8 m 3 . The empty bed retention time (EBRT) was set to 24-48 s by changing the feeding air stream from 4320 to 8640 m 3 /h. Inlet air was humidified by passing through SP.

The two zones of the IP were filled with PU and volcanic rock medium, respectively. The AZ includes a pretreatment unit, adsorption unit and desorption unit. The adsorption unit was packed with modified activated carbon fiber (ACF) filler. The temperature and humidity of the system were measured using a Dewpoint Thermohygrometer (WD-35612, OAKTON, Germany). The FIR parameters are shown in Tab. 1.

[Insert table 1]

The FIR was operated for 307 d to treat sludge dewatering exhaust. The two-phase BF was initially inoculated with activated sludge supernatant obtained from the WWTP and with odor-degrading microorganisms from a BF in our laboratory for fast start-up. A mineral nutrient solution with 0.54 g/L KH2PO4, 1 g/L NaCl, 0.025 g/L MgSO4, 0.02 g/L CaCl2, J o u r n a l P r e -p r o o f 7 0.005 g/L FeSO4•4H2O, 0.000088 g/L MnSO4•H2O and 0.2 g/L glucose was sprayed onto the packing material at a flow rate of 5 m 3 /h 12 times a day for 1 min at a time to supply nutrients. The pH of the nutrient solution was adjusted via acid or alkali addition. Exhaust passes through the two-phase BF in an upflow direction through the air distribution device and then enters the AZ. The pretreatment unit of the AZ is mainly designed to reduce moisture in the air stream. Three sets of adsorption units form the main module of the AZ; two work online and the other operates in a regeneration state during operation. Steam was used in the desorption unit for desorption. Temperatures were kept at 150-180 ℃ by controlling the steam flow. Pollutants adsorbed in the ACF were desorbed and discharged with the steam. The desorption process was complete when temperatures fell below 60 ℃ and when the adsorption capacity of activated carbon fiber had recovered. 

The gases were sampled periodically at a sample port in 10 L Tedlar bags. For microbial enumeration, 1.0 mL of liquor in SP, 1.0 g of PU (cut into pieces) and 1.0 mL of leakage from the volcanic rock filter were taken from the sampling ports, mixed with 100 mL of sterile water and agitated for 10 min. The HB were incubated in nutrient agar (BR, Aoboxing Biotech, Co., China) at 37 °C for 48 h. TF were incubated in Martin Broth medium (BR, Aoboxing Biotech Co., China) at 28 °C for 5 days. Thiosulfate and modified Waksman media J o u r n a l P r e -p r o o f 8 were used to culture non-acidophilic Thiobacillus (NAT) and acidophilic Thiobacillus (AT), respectively (Cho et al., 1991) . Inoculation was conducted in triplicate, and the average value for each sample was calculated. The distinct colonies were sub-cultured and purified by streaking on fresh agar plates after incubation. The population sizes of ammonia oxidizing bacteria (AOB) and nitrite oxidizing bacteria (NOB) of packing mediums were determined by the MPN method. The basal medium and procedures used were selected based on a previous study (Yasuda et al., 2010) . Counts of AOB and NOB were recorded as MPN•g -1 of dry packing medium or MPN•mL -1 of liquor. USA) using nitrogen as a carrier gas was applied to measure H2S. Before sampling, all canisters were cleaned and then vacuumed using humid N2 pure gas (99.999%). Oven, injection and detector temperatures were set to 45 °C, 100 °C and 200 °C, respectively. A spectrophotometer (Xinyue-T6) was used to analyze NH3 in gas samples via Nessler reagent spectrophotometry and through the measurement of nitrate, nitrite, and sulfate concentrations in the liquid phase as described in our previous studies (Liu et al., 2017 (Liu et al., , 2011 . VOCs were analyzed using a gas chromatograph (GC, Agilent 6890N, USA) equipped with a DB-5MS column (60 m×0.32 mm×1.0 μm) and mass spectrometer (MS, Agilent 5973MSD). The United States Environmental Protection Agency's (US EPA's) Method TO-15 was adopted.

The sample air was first passed through a stainless steel cryotrap (1/8 in × 8 cm) packed with fine glass beads and cooled with liquid nitrogen at -170 °C. Several standard gases were used to calibrate VOCs, and Helium 5.0 was used as a carrier gas. The GC oven was first set to 35 °C for 5 min, to 150 °C at 5 °C /min, to 220 °C at 15 °C/min, and then kept at this temperature for 7 min. This process took 40 min, and the temperature of injection port was 100 °C. Each compound was identified based on its retention time and mass spectrum. 

Positive-hole correction was used to determine colony count concentrations. The results were calculated as the geometric mean of the replicates and expressed as colony-forming units per cubic meter of air (CFU/m 3 ).

N 1 +N 2 +N 3 +N 4 +N 5 +N 6 Q•t ×1000

where C denotes microbe concentrations in CFU/m 3 , N is the total number of microbes in each disk colony in CFU, t is sampling time in minutes, and Q is the gas flow rate during sampling in L/min.

The AERSCREEN model recommended by the US EPA was used to estimate maximum ground-level concentrations of air pollutants emitted from different sources under worst-case meteorological conditions (Asadi et al., 2014 ). The ARESCREEN model was applied to estimate concentrations of microorganism and gaseous compounds (NH3, benzene, H2S, ethylbenzene, and xylene) 50 m to 1000 m downwind the FIR outlet (Table S1 ). The FIR outlet was considered as a point source of emissions; software input parameters employed are shown in Table S2 .

The inhalation of microorganism and gaseous compounds was the main exposure pathway considered. Most microorganism and gaseous compounds released from the FIR pose noncarcinogenic risks. Exposure concentrations and noncarcinogenic risks of gaseous compounds were estimated using Eqs. (1) and (2), and those of bioaerosols were estimated using Eqs.

(3) and (4). employing commonly used culture dependent techniques for bioaerosol quantification. Thus, the value of 300 CFU/m 3 was employed in this study. When HQ < 1, noncarcinogenic risks can be disregarded. However, when HQ > 1, potential adverse health risks are of concern (Chen and Carter, 2017; Liu et al., 2018) .

During the sludge drying process, many odors, bioaerosols, and VOCs including sulfurcontaining compounds, nitrogen-containing compounds, aromatic compounds, and benzene series are generated and released (Li et al., 2011a; Lv et al., 2016; Weng et al., 2015) . A total of 18 pollutants, including 6 benzenes, 3 ketones, 2 inorganic odor compounds, 2 sulfur organic compounds, 2 esters, an alcohol and a halogenated compound were detected in this indicates that the two-phase BF gradually became rich in microorganisms, potentially degrading the target contaminants. Microorganisms in the two-phase BF gradually became adapted to highly odorous concentrations after the acclimatization period, and the REs of H2S and NH3 remained more than 95%. These results indicate that 37 days was sufficient for the FIR to successfully start and reach steady-state conditions. The complete removal of contaminants could be achieved under a steady state. We thus achieved superior odor treatment performance than previous works (Liu et al., 2017 (Liu et al., , 2008 with the same amount of EBRT, as the FIR was equipped with SP at the air intake port and AZ after the BF enhanced the treatment performance. Fluctuation of REs occurred in stage Ⅳ with a continued decrease in EBRT, which can be attributed to the increase in the inlet loading rate and less contact time between contaminants and microorganisms. Past works suggest that odor removal mechanisms are dependent on the sorption of odorous compounds into the biofilm layer on the media surface where biodegradation takes place, which relies on long residence times (Fletcher et al., 2014) . Overall, the REs of H2S and NH3 were stabilized at relatively high levels and maintained for nearly 9 months. The FIR's VOC removal performance is shown in Fig. 3 . When VOCs in the inlet flow through the support medium, these pollutants are absorbed by the biofilm and converted into innocuous products such as carbon dioxide, water, and cell mass without generating undesirable byproducts (Moe and Irvine, 2001 Hydrophilic substrates can be more effectively biodegraded in BFs because high gas-liquid transfer rates make target pollutants more available to microorganisms (Cheng et al., 2016b) .

However, the low mass-transfer rates are the main limitations to removal performance for 

Large quantities of bioaerosols containing pathogenic microbes are emitted from sludge dewatering rooms of WWTPs (Han et al., 2018; Li et al., 2011a; Szyłak-Szydłowski et al., 2016) . To remove microbial aerosols in an inlet air stream, it is necessary to reduce levels of air pollution generated by bioaerosols. Fig. 4 shows the REs and concentrations of airborne HB and TF in the FIR inlet and outlet. As outlets are located outside, the FIR both retains and emits bioaerosols. It acts as an emission source of outdoor air and a retainer of ambient air in a sludge dewatering room. The observed negative and positive values indicate that the studied biofilter acted as a bioaerosol retainer and emitter, respectively. According to previous research, a BF can act as a net emitter of bioaerosols (the absolute negative value is greater than the positive value), which is a major concern for regulators (Flores- Barbosa et al., 2020; Ibanga et al., 2018) . Overall, the FIR served as a retainer for both airborne HB and TF in this study. Bioaerosol capture mechanisms of a BF have been thought to include inertial deposition, diffusional or Brownian deposition and flow line interception (Ottengraf and Konings, 1991) , and these combine to affect bioaerosol impingement on solid media material such that as bioaerosol-laden air sweeps through the media bed, particles are deposited in the media (Ibanga et al., 2018) .

However, bioaerosol emissions become apparent in stages Ⅲ

and Ⅳ, which is related to an J o u r n a l P r e -p r o o f increase in the flow rate. A higher inlet load and thus higher biomass density in a biofilter seems to boost bioaerosol emissions (Chung, 2007; Vergara-Fernández et al., 2012b) . With an increase in biofilm thickness, the binding force between the biofilm and packing materials weakens, causing microorganisms to be carried by the airflow and emitted from the biofilter in the form of bioaerosols (Chung, 2007; Vergara-Fernández et al., 2012b; Wang et al., 2018) .

K. also found bacterial emissions of BF to increase from 449 ± 27 CFU/m 3 to 643 ± 46 CFU/m 3 as flow rates increased from 1500 m 3 /h to 3500 m 3 /h. However, increasing gas velocities first raise bioaerosol emissions and then decrease concentrations due to the shearing force and dilution effects (Hu et al., 2020) . Bioaerosols emitted from biofilters include a combination of nonimpacted microorganisms retained in treated process air and those blown off from the surfaces of media particles by the passing airstream, suggesting that the species composition of outlet air may be different from that of inlet air (Martens et al., 2001) . The contributions of different units to bioaerosol removal in a steady state are shown in Tab. 5.

With increasing gas flow rates, the REs of HB and TF decreased in the BF and AZ.

Microorganisms attached to biofilter packing material can be sheared off from the surface of the support media by the gas stream, promoting the emission of microbes in medium . Large amounts of airborne HB were removed in the LPZ (up to 38.62% in stage Ⅰ) and AZ while most TF removal occurred in the NPZ and AZ. The removal of HB was less effective than that of TF in SP. Factors such as gas flow rates, the moisture content of packing materials, EBRT, particle sizes and shapes, and operating conditions influence the removal performance of bioaerosols (Willeke et al., 1996; . Due to the different operating conditions of each unit, in this study the influence of microbial characteristics cannot be ignored. In the AZ, inlet concentrations of airborne HB and TF reached 414-2335

and 184-1011 CFU/m 3 , respectively. It should be noted that the AZ maintained a similar RE for airborne HB and TF and no selectivity was exhibited. The porous structure, large surface area, and hydrophobic surfaces of activated carbon favor the effective attachment of airborne microorganisms (Dunne, 2002; Li et al., 2011b; Lillo-Ródenas et al., 2005) . Hydrophobic interactions between cell and adsorbent surfaces enable cells to overcome repulsive forces J o u r n a l P r e -p r o o f active within a certain distance from adsorbent surfaces and irreversibly attach (Dunne, 2002; Li et al., 2011b) . Li et al. (2011b) obtained a 93.75% adsorption rate for culturable airborne bacteria and a rate of 100% for fungi over a longer contact period (8 h) than that used in this work. The AZ plays a crucial role in bioaerosol adsorption, maintaining total REs of bioaerosols at roughly 80%. Additionally, adopting short starvation periods, partial substrate limitation (with the elimination of some pollutants in the case of a gaseous mixture), high moisture levels and the mineral medium frequency of packing materials to avoid stressful conditions (limiting nutrients and drying) may help prevent the generation of bioaerosol emissions during BF operation (Flores-Barbosa et al., 2020; Vergara-Fernández et al., 2012a , 2012b Wang et al., 2018) .

The particle sizes of bioaerosols dictate where aerosolized pathogens deposit in the human respiratory tract, and thereafter the pathogens interact with host tissue, which can cause disease and a host of immunological responses (Thomas, 2013) . Larger particles (>8 mm) deposit in the upper respiratory tract in a size dependent manner from the nasal passage to larger bronchioles due to inertia, smaller particles (<1-3 mm) diffuse deep into the lung tissue, and fine particles (<2.5 μm) can be directly inhaled and adhere to the respiratory tract and alveolar region (Thomas, 2013; Wang et al., 2019 ). In addition, particle size plays a key role in aerosols residence times and transmission distances. Aerosols with small particle sizes have longer residence times and can spread further (Gong et al., 1997; Li et al., 2011a) .

Therefore, the risks associated with exposure to aerosols not only relate to their concentrations and species but also to their particle size distributions (Li et al., 2011b) . The particle size distributions of airborne HB and TF during stage Ⅲ in each sample port are shown in Fig. 6 . Tab. 4. This may have occurred because (a) HB and TF were removed in SP as incoming aerosol particle size distributions, leaving the particle size distribution unchanged and/or b) large amounts of HB and TF were absorbed while many microbes in SP were aerosolized due to mechanical disturbance while the particle size distribution of emitted bioaerosols was similar to that of incoming aerosols. Percentages of particles of individual size changed considerably after particles passed through the LPZ. The proportion of microorganisms with particle sizes of less than 4.7 μm increased rapidly in the LPZ and NPZ by roughly 70%, implying that the packing in IP better retains large particles. The main purpose of packing media in IP is to provide a surface for microorganism attachment and growth and to retain particles. Particle size distributions vary between IP inlet and outlet air with the outlet including larger proportions of smaller particles with 70% being < 4.7 μm. This may be a result of the filter bed preferentially trapping larger particles from the gas flow, and/or this may simply be the size range emitted by a BF (Sanchez-Monedero et al., 2003) . However, after passing through the AZ, bioaerosols with particle sizes of > 4.7 μm accounted for more than 80%, showing that HB and TF of large particle sizes were difficult to adsorb while small particles were readily captured by the activated carbon, corroborating previous studies (Li et al., 2011b) . Micrographs provided by Li et al. (2011b) demonstrate that activated carbon has a rough surface and includes various pore sizes on particle surfaces. Its small macrospore structures thus rendering it effective for the adsorption of small particles.

IP and the AZ better retain and capture airborne HB and TF with large and small particles, respectively. The FIR integrates the advantages of single conventional biological and adsorption processes. The combination of a two-phase BF and AZ can effectively remove bioaerosols with different particle sizes. Therefore, the FIR can reduce bioaerosol levels produced through sludge dewatering and thus reduce worker exposure risk. Performance of FIR mainly relies on the microorganisms, which composed of biodegradation by microorganisms, dissolution and absorption in the aqueous phase, and adsorption by the packing materials and microorganisms . In SP, as we know from Fig. 7 , the number of HB present was 2-3 orders of magnitude higher than the number of TF present with up to 2.43×107 CFU/mL found.

Microbes in the SP outlet sample port were characterized by aerosolized microbes in liquid phase under the aeration system and by an incomplete removal of inlet bioaerosols. As concentrations of HB are far higher than those of TF in SP, HB are more likely to be emitted during aeration. TF aerosols were absorbed by the liquid while fewer were aerosolized than HB, and thus fewer HB aerosols were removed in SP. A similar phenomenon occurred during the uptake of airborne HB in the LPZ. The numerous AOB and NOB found in SP is consistent with the removal performance of NH3. In stage Ⅲ

, nitrite, nitrate, and NH4 + concentrations reached 1.05 ± 0.19 mg/L, 6.28 ± 0.29 mg/L, and 11.24 ± 0.49 mg/L, respectively, showing that NH4 + was the main inorganic nitrogen-containing compound present in SP. As a less water-soluble compound, H2S has a poor removal effect of roughly a quarter of NH3 RE, though concentrations of NT remained high in the unit. The main products of sulfur-containing compound bio-oxidation are elemental sulfur and sulfate. S 2was also detected when H2S was dissolved in SP. Concentrations of SO4 2and S 2reached 1.08 ± 0.06 and 5.12 ± 0.23 mg/L in stage Ⅲ

. Most sulfur-containing compounds were present in the form of S 2in SP.

Contaminants in bulky gas are transferred from the gas phase to the water or biofilm phase by diffusion prior to biodegradation. The mass transfer rate of pollutants from air to water is related to hydrophobicity (Cheng et al., 2016b) . As aerial mycelia of fungi form a very large surface area and may come into direct contact with gas flows, fungal biofilters act as an alternative means to enhance hydrophobic compound transfer. Furthermore, fungi are more resistant to acidic and dry conditions than bacteria, which is a helpful property when BF J o u r n a l P r e -p r o o f 20 treatment processes produce acid compounds such as SO4 2- (Liu et al., 2013 further downwind from the FIR outlet. Due its strong capacity to remove multiple pollutants, the studied FIR is suitable for urban areas with high population densities. For the bioaerosols of sludge dewatering room control technologies, UV, H2O2, and O3 are applicable given their low energy consumption and high inactivation efficiency. Photocatalysis disinfection is also a very promising technology for controlling bioaerosols, as it generates less secondary pollution and given its high inactivation efficiency (Hu et al., 2020) . Proper ventilation and requiring operators to wear face masks are also strongly encouraged.

The average REs of odors, VOCs and bioaerosols were recorded as 98.5%, 94.7% and 86.4%, respectively, when the inlet air flow rate was 5760 m 3 /h. The simultaneous removal performance of each unit worsened with an increase in the inlet air flow rate. Microbial communities in BF changed significantly under different environmental conditions. Most hydrophobic compounds were removed in the LPZ and hydrophilic compounds were J o u r n a l P r e -p r o o f 22 degraded in SP and the NPZ. Approximately 35% of HB aerosols were removed in the LPZ while over 30% of TF aerosols were removed in the NPZ, which is related to the distribution of microorganisms in the two-phase BF. The combined use of a two-phase BF and AZ can remove bioaerosols with different particle sizes. The AZ can also thoroughly remove residual pollutants at the outlet of a two-phase BF and ensure stable removal capacities of the system.

In conclusion, the FIR can simultaneously remove bioaerosols, odors and VOCs effectively through the synergistic degradation of different units. The health risks were negligible 50 m downwind from the FIR outlet.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. J o u r n a l P r e -p r o o f 

Technical and economic investigation of chemical scrubber and bio-filtration in removal of H2S and NH3 from wastewater treatment plant

Simultaneous treatment of dimethyl disulfide and hydrogen sulfide in an alkaline biotrickling filter

Ammonia dispersion using experimental and modeling methods

Adsorption of volatile organic compounds by pecan shell-and almond shell-based granular activated carbons

Technological and life cycle assessment of organics processing odour control technologies

Compounds interaction on biodegradation of toluene and methyl ethyl ketone mixtures in a composite biofilter

Modeling potential occupational inhalation exposures and associated risks of toxic organics from chemical storage tanks used in hydraulic fracturing using AERMOD

Effects of anionic surfactant on n-hexane removal in biofilters

Challenges and solutions for biofiltration of hydrophobic volatile organic compounds

Removal characteristics of hydrogen sulphide and methanethiol by Thiobacillus sp. isolated from peat in biological deodorization

Evaluation of gas removal and bacterial community diversity in a biofilter developed to treat composting exhaust gases

Two-stage biofilter for effective NH3 Removal from Waste Gases Containing High Concentrations of H2S

Airborne route and bad use of ventilation systems as non-negligible factors in SARS-CoV-2 transmission

Bacterial Adhesion : Seen Any Good Biofilms Lately ?

Characterization of bioaerosol emissions from two biofilters during treatment of toluene vapours using epifluorescence microscopy

Understanding biofilter performance and determining emission concentrations under operational conditions

Evaluation of bioaerosols by flow cytometry and removal performance in a biofilter treating toluene/ethyl acetate vapors

Desulfurization of biogas in biotrickling filters

Modeling sea-salt aerosols in the atmosphere 1. Model development

Bacterial population and chemicals in bioaerosols from indoor environment: Sludge dewatering houses in nine municipal wastewater treatment plants

Bioaerosols emission and exposure risk of a wastewater treatment plant with A2O treatment process

A short review of bioaerosol emissions from gas bioreactors: Health threats, influencing factors and control technologies

Pilot-scale biofiltration at a materials recovery facility: The impact on bioaerosol control

Application of a compact trickle-bed bioreactor for the removal of odor and volatile organic compounds emitted from a wastewater treatment plant

Odor Assessment and Management in Wastewater Treatment Plants : A Review

Characteristics of odors emitted from municipal wastewater treatment plant and methods for their identification and deodorization techniques

Distribution characterization of microbial aerosols emitted from a wastewater treatment plant using the Orbal oxidation ditch process

wastewater treatment oxidation ditch by adsorption on activated carbon

Behaviour of activated carbons with different pore size distributions and surface oxygen groups for benzene and toluene adsorption at low concentrations

Removal of H2S by coimmobilized bacteria and fungi biocatalysts in a bio-trickling filter

Determination and risk assessment of airborne endotoxin concentrations in a university campus

A full-scale integrated-bioreactor with two zones treating odours from sludge thickening tank and dewatering house: performance and microbial characteristics

Removal of ammonia from waste gases by a biotrickling filter

Performance of two biofilters with neutral and low pH treating off-gases

Aerodynamic Characteristics and RNA Concentration of SARS-CoV-2 Aerosol in Wuhan Hospitals during COVID-19 Outbreak

Modelling the removal of volatile pollutants under transient conditions in a two-stage bioreactor using artificial neural networks

Effects of coprocessing sewage sludge in cement kiln on NOx

Reduction potential of microbial, odour and ammonia emissions from a pig facility by biofilters

Effect of nitrogen limitation on performance of toluene degrading biofilters

Performance of a thermophilic gas-phase biofilter treating high BTEX loads under steady-and transient-state operation

Bioreactors for treatment of VOCs and odours -A review

Reducing bacterial aerosol emissions from membrane bioreactors: The impact of SRT and the addition of PAC and calcium

Emission of microorganisms from biofilters

Aerosol and Surface Stability of SARS-CoV-2 as Compared with SARS-CoV-1

Odor and VOC treatment by biotrickling filters: pilot scale studies at the Hyperion Treatment Plant

Economical assessment of the design, construction and operation of open-bed biofilters for waste gas treatment

Performance of a monolith biotrickling filter treating high H2S from mimic biogas and elemental sulfur plugging control using pigging

Back-propagation neural network for performance prediction in trickling bed air biofilter

Effect of the aeration system on the levels of airborne microorganisms generated at wastewater treatment plants

Biofiltration at composting facilities: Effectiveness for bioaerosol control

Transmission Potential of SARS-CoV-2 in Viral Shedding Observed at the University of Nebraska Medical Center

Odour impact assessment by means of dynamic olfactometry, dispersion modelling and social participation

Investigation of removal capacities of biofilters for airborne viable micro-organisms

Mesophilic and thermophilic BTEX substrate interactions for a toluene-acclimatized biofilter

Underground sewage treatment plant: a summary and discussion on the current status and development prospects

Monogenic, polygenic, and oligogenic familial hypercholesterolemia

Particle size and pathogenicity in the respiratory tract

Removal of alpha-pinene from gases using biofilters containing fungi

Influence of the inlet load, EBRT and mineral medium addition on spore emission by Fusarium solani in the fungal biofiltration of hydrophobic VOCs

Temperature and moisture effect on spore emission in the fungal biofiltration of hydrophobic VOCs

Biofiltration of benzo[Α]pyrene, toluene and formaldehyde in air by a consortium of Rhodococcus erythropolis and Fusarium solani: Effect of inlet loads, gas flow and temperature

Dissimilar emission characteristics between bioaerosol and suspended particles from gaseous biofilters and bioaerosol health risk evaluation

Effects of aeration on microbes and intestinal bacteria in bioaerosols from the BRT of an indoor wastewater treatment facility

Release and control of hydrogen sulfide during sludge thermal drying

Penetration of airborne microorganisms through a surgical mask and a dust/mist respirator

Simultaneous Removal of Multicomponent VOCs in Biofilters

A Full-Scale Thermophilic Biofilter in the Treatment of Sludge Drying Exhaust: Performance, Microbial Characteristics and Bioaerosol Emission

Influence factors and health risk assessment of bioaerosols emitted from an industrial-scale thermophilic biofilter for off gas treatment

Strong influence of medium pH condition on gasphase biofilter ammonia removal , nitrous oxide generation and microbial communities

Aerosols from a wastewater treatment plant using oxidation ditch process: Characteristics, source apportionment, and exposure risks

Evaluation of a pilot-scale biotrickling filter as a VOCs control technology for the chemical fibre wastewater treatment plant

On airborne transmission and control of SARS-Cov-2

biofilter treating malodors from livestock manure composting

Odor composition analysis and odor indicator selection during sewage sludge composting

Profiles of inlet concentration (Cin), outlet concentration (Cout) and removal efficiencies (REs) for (a)

Profiles of inlet concentration (Cin), outlet concentration (Cout) and removal efficiency (RE) for VOCs

Profiles of inlet concentration (Cin), outlet concentration (Cout) and removal efficiencies (RE) for (a) HB, (b) TF

Bioaerosols emitted or retained during operation (a) HB, (b) TF

Microorganisms in different units (a) suspended phase (b) low-pH zone (c) neutral-pH zone

This work was supported by Beijing Science and Technology Commission Foundation (Z181100005518011).J o u r n a l P r e -p r o o f

2.40 × 10 -2 2.47 × 10 -2 1.04 × 10 -3 4.07 × 10 -2 50m 7.77 × 10 -3 2.40 × 10 -6 2.52 × 10 -6 1.05 × 10 -7 4.14 × 10 -6 100m 4.07 × 10 -3 1.21 × 10 -6 1.27 × 10 -6 5.34 × 10 -8 2.10 × 10 -6 EBZ: ethylbenzene J o u r n a l P r e -p r o o f