Abstract

Antibiotic resistance genes (ARGs), as a new type of environmental pollutant that threaten human health, have been detected in the effluent of sewage treatment systems. In this study, the removal from water of ARGs, 16S rRNA, class 1 integron (intI1), and dissolved organic carbon (DOC) were investigated using processes combining nano-iron (nFe), ultrasound (US), activated persulfate (PS) and ultrafiltration (UF). The oxidation mechanism was also studied. The results showed that both nFe and US activation could improve the oxidative effect of PS, and the effect of nFe was better than that of US. Compared with PS-UF, nFe/PS-UF and US/PS-UF significantly enhanced the removal of various ARGs and DOC. nFe/PS-UF was the most effective treatment, reducing cell-associated and cell-free ARGs by 1.74–3.14-log and 1.00–2.61-log, respectively, while removing 30% of DOC. Pre-oxidation methods using PS, nFe/PS, and US/PS significantly enhanced the efficacy of UF for removing DOC with molecular weights above 50 kDa and below 10 kDa, but the removal of DOC between 10 and 50 kDa decreased. The free radicals SO4·− and ·OH were shown to participate in the process of ARGs oxidation.

HIGHLIGHTS

  • Persulfate oxidation pretreatment significantly improved the removal effect of ultrafiltration on ARGs due to release of SO4·−.

  • Nano-iron forms a loose and porous oxide shell on the surface of the particles, which can adsorb ARGs.

  • Cavitation during ultrasound treatment can lyse the cell structure, transforming cell-associated ARGs into the free state

Graphical Abstract

Graphical Abstract
Graphical Abstract

INTRODUCTION

In 2006, Pruden et al. proposed antibiotic resistance genes (ARGs) as a new type of environmental pollutant (Pruden et al. 2006). ARGs have physical and chemical properties that cause them to persist in the environment and biological properties that enable them to be reproduced and disseminated, even if the cells carrying them are inactivated. Consequently, the DNA of ARGs can survive in the environment and resist degradation (Wen et al. 2015). Studies have shown that microorganisms can be used as expression sites for ARGs, and the presence of microorganisms will lead to the transfer and spread of ARGs. In addition, integron (int), as a mobile genetic factor of bacteria, can also carry one or more genes related to ARGs for horizontal transfer between bacteria (Chen 2019). At present, there are five types associated with the ARGs integron; among them, Class I integron (intI1) is the most common type. The removal of ARGs in water is closely related to intI1 and microorganisms, and one of the reasons for the resistance of microorganisms is the acquisition of exogenous ARGs and genetic recombination (Enne et al. 2001). The detection of 16S rRNA can directly reflect the microbial abundance in water samples (Li 2015), so the concentration change of 16S rRNA is closely related to the removal of ARGs. Wastewater treatment plants are being scrutinized as potential hot spots for the spread of antibiotic resistance in the environment because they provide convenient conditions for the proliferation of ARG-carrying bacteria and the horizontal transfer of ARGs between different microorganisms (Zhang et al. 2009). In addition, since ARGs and organic matter in water can be adsorbed together through interaction, organic matter provide a good protection place for ARGs after adsorption (Cai et al. 2006; Nguyen et al. 2010), so the removal of organic matter in water can promote the reduction of ARGs concentration. However, traditional wastewater treatment methods do not significantly reduce ARGs, so high levels of ARGs can still be detected in secondary effluent (Jiao 2017).

In recent years, membrane technology has been used in the field of reclaimed water treatment because of its high efficiency, low cost, and convenient management. Studies have shown that ultrafiltration (UF) technology is effective in removing ARGs (Nunes & Peinemann 1992). Currently, to improve the efficacy of membrane technology, various pretreatment methods are employed. Common pretreatment methods include coagulation, oxidation, and adsorption (Fan et al. 2014). In addition, advanced oxidation technology has been widely used for water treatment. Research has shown that advanced oxidation combined with UF processes can effectively remove organic pollutants from water (Gong 2019). The removal of ARGs by advanced oxidation technology mainly results from the oxidative destruction of the ARGs by free radicals generated within the system (Chen & George 2018). Persulfate (PS) oxidation is an emerging oxidation technology due to its longer bond length and lower bond dissociation energy (Kolthoff & Miller 1951), which can produce more efficient and stable sulfate radicals (SO4·) that are less affected by the environment and hydroxyl radicals (·OH) (He 2017). PS exhibits strong oxidizing ability against pollutants including bacterial cells and DNA (Zhang et al. 2019a), so it can effectively remove ARGs. However, the oxidation capacity of PS is limited and activation is required to improve its efficacy (Xiao 2016). At present, commonly used activation techniques include thermal, ultrasonic, electrode, and metal activation, among other methods (Zhang et al. 2012; Shi et al. 2017; Li & Zhu 2019). In recent years, nano-iron (nFe) and ultrasound (US) have been used to activate PS, generating reactive oxidants for the treatment of organic pollutants, but there have been few studies of the treatment of ARGs in secondary effluent.

In this study, different combinations of nFe and US with UF were used to activate PS to explore the efficacy of various processes for the removal of ARGs and dissolved organic carbon (DOC) in secondary effluent from a wastewater treatment plant, analyze the relationship between intI1, 16S rRNA and ARGs removal. The study aims to improve the safety of reclaimed water.

MATERIALS AND METHODS

Materials

Water samples were collected from the secondary sedimentation tank of Gaobeidian Sewage Treatment Plant, Beijing, China. Gaobeidian Sewage Treatment Plant has a capacity of 1.5 million m3/d, it collects most of the domestic sewage in the southern part of Beijing, the eastern suburb industrial zone, the embassy area and the whole sewage of the chemical road. This plant adopts the traditional activated sludge process. The first stage uses the pre-anoxic stage propulsive activated sludge process. The second stage is divided into two series, one is the anoxic activated sludge process and the other is the anoxic and aerobic denitrification activated sludge process. The water sample was collected from the outlet of the secondary sedimentation tank, and transported back to the laboratory after sampling.

All water samples were stored at 4 °C and basic tests of raw water quality were carried out as soon as possible, shaking thoroughly before testing. Raw water quality data are shown in Table 1.

Table 1

Raw water quality indexes

Average valueAverage value
pH 7.3 ± 0.1 DOC (mg/L) 7.86 ± 0.50 
Temperature(°C) 24.9 ± 1.0 TP (mg/L) 1.56 ± 0.50 
Average valueAverage value
pH 7.3 ± 0.1 DOC (mg/L) 7.86 ± 0.50 
Temperature(°C) 24.9 ± 1.0 TP (mg/L) 1.56 ± 0.50 

Polyethersulfone (PES) plate UF membrane with molecular weight cutoff of 100 kDa was obtained from Millipore Co. (USA), and the UF membrane diameter was 76 mm, the membrane effective area was 41.8 cm2. Sodium PS from Fuchen Chemical Reagent Co. (Tianjin, China), and nFe from Guangzhou Jie Chuang Trading Co. (Guangdong, China). The KQ5200DE ultrasonic cleaner (Kunshan Ultrasonic Instrument Co., China), was operated to a maximum frequency of 40 kHz. TP was detected by U2001 UV-Vis Spectrophotometer (HITACHI, Japan) potassium persulfate digestion-molybdenum antimony anti-spectrophotometry.

Experimental setup and methods

The oxidation pretreatment test was carried out at room temperature (23 ± 2 °C) with the following steps: for the PS oxidation test, various concentrations of PS (2, 4, 6, 8, and 10 mM) were added to raw water and stirred with a constant temperature magnetic stirrer (B11-2, Shanghai Sile Instruments Co., Ltd, China) at 200 rpm for 30 min. A maximum stirring capacity of 5,000 mL for the magnetic stirrer. For the nFe-activated PS oxidation test, various concentrations of nFe (0.5, 1, 2, and 4 mM) were added to raw water containing 4 mM PS and stirred with the constant temperature magnetic stirrer at 200 rpm for 30 min. For the US-activated PS oxidation test, raw water containing 4 mM PS was subjected to US (16, 24, 32, and 40 kHz) activation and stirred with the constant temperature magnetic stirrer at 200 rpm for 30 min. The UF test adopted model 8,400 UF cup (Millipore, USA), in which the UF mode was dead end filtration, and the filtration pressure (0.10 MPa) was provided by nitrogen.

Detection of ARGs

DNA extraction

The water sample was filtered by 1 kDa UF membrane, the membrane was shredded into the DNA extraction kit (MP Biomedicals, USA), and stored in −20 °C refrigerator before extraction and analysis. The content and purity of the extracted DNA were determined by ultramicro spectrophotometer (Thermo Co., USA).

PCR primer synthesis

Six genes were detected, including 2 tetracycline resistance genes (tetA, tetC), 2 sulfanilamide resistance genes (sulI, sulII), intI1 and 16S rRNA. The primers for the detection genes were synthesized by Qingke Biological Co., Ltd. The primer related information is shown in Table 2.

Table 2

Primer sequences of ARGs

Name of ARGsPrimer F gene sequencePrimer R gene sequenceAnnealing temperature (°C)Product length (bp)
tetA GCTACATCCTGCTTGCCTTC CATAGATCGCCGTGAAGAGG 60 210 
tetC CTTGAGAGCCTTCAACCCAG ATGGTCGTCATCTACCTGCC 68 418 
sulI CGCACCGGAAACATCGCTGCAC TGAAGTTCCGCCGCAAGGCTCG 57 162 
sulII TCCGGTGGAGGCCGGTATCTGG CGGGAATGCCATCTGCCTTGAG 60 191 
intICCTCCCGCACGATGATC TCCACGCATCGTCAGGC 55 280 
16S rRNA CCTACGGGAGGCAGCAG TTACCGCGGCTGCTGGCAC 55 193 
Name of ARGsPrimer F gene sequencePrimer R gene sequenceAnnealing temperature (°C)Product length (bp)
tetA GCTACATCCTGCTTGCCTTC CATAGATCGCCGTGAAGAGG 60 210 
tetC CTTGAGAGCCTTCAACCCAG ATGGTCGTCATCTACCTGCC 68 418 
sulI CGCACCGGAAACATCGCTGCAC TGAAGTTCCGCCGCAAGGCTCG 57 162 
sulII TCCGGTGGAGGCCGGTATCTGG CGGGAATGCCATCTGCCTTGAG 60 191 
intICCTCCCGCACGATGATC TCCACGCATCGTCAGGC 55 280 
16S rRNA CCTACGGGAGGCAGCAG TTACCGCGGCTGCTGGCAC 55 193 

Quantitation of ARGs

SYBR-Green real-time quantitative PCR (qPCR) was used to detect tetA, tetC, sulI, sulII, intI1 and 16S rRNA genes. Fluorescent qPCR was conducted using a PRISM® 7,900HT instrument (Applied Biosystems, USA). The qPCR products were cloned and their DNA sequenced. Plasmids were extracted from the positive clones using a plasmid extraction kit (Tiangen, China). Optical density values of plasmids were determined and used to prepare standard curves.

The plasmid copy number conversion formula was (Luo et al. 2019):
formula
(1)

After qPCR assay, gene quantity was evaluated in absolute abundance. Absolute abundance was obtained by normalizing gene copies with the volume of extracted sample to evaluate the ARGs amount in unit volume of water samples.

The removal of ARGs is recorded as the difference between the logarithmic value of ARGs concentration in raw water at the base of 10 and the logarithmic value in process effluent at the base of 10.

Quantitative fluorescence PCR 20.0 μL system: 10.0 μL Master Mix(2×), 0.4 μL PCR Forward Primer (10 μmol/L), 0.4 μL PCR Reverse Primer (10 μmol/L), 8.2 μL ddH2O (sterilized distilled water), 1.0 μL Template DNA. The fluorescence quantitative PCR reaction procedure was 95 °C 3 min, 95 °C 30 s, 56 °C 30 s for 40 cycles、72 °C 30 s, Last 95 °C 15 s, 60 °C 15 s and 95 °C 15 s. Melt-curve analysis was 60 °C to 95 °C and the fluorescence was collected every 0.2 °C to form dissolution curve. The annealing temperature and reaction time were adjusted according to the primers.

Existence morphology detection of ARGs

Nucleases are a class of enzymes that can degrade DNA or RNA. The first nuclease discovered is deoxyribonuclease (DNase I) (Dirk & Hans 2007). Many studies have shown that the environment contains persistent free-state DNA. These extracellular DNA can be adsorbed on the complex particles in the sediment and soil, reducing contact with deoxyribonuclease and avoiding degradation (Niemeyer & Gessler 2002). Studies have shown that extracellular DNA molecules can also serve as carriers of resistance genes, which can gain resistance through horizontal gene transfer through contact with bacteria in the environment. Therefore, the use of DNase I to degrade free extracellular DNA in the water environment is helpful to distinguish intracellular versus extracellular ARGs.

After mixed the test water sample, it was packed into a sterile bottle of up to 1 L volume. Each group was divided into two samples, one for direct extraction DNA tested and the test result is total ARGs content. The cell-free ARGs was degraded by adding 500 U of deoxyribonuclease (Dnase I) (about 250 μg DNA, in excess of that required for the DNA content in the water sample) to another water sample. The SHA-B water bath thermostatic oscillator (Shanghai Hechen Energy Technology Co., Ltd, China) was used for oscillate reaction in 15 °C environment for 2 h then the reaction was stopped after holding for 10 min at 80 °C, and the result of DNA detection in the extracted water sample is the content of cell-associated ARGs. The content of cell-free ARGs is the difference between total ARGs content and the content of cell-associated ARGs.
formula
(2)

Free radical analyses

Free radical quenching test

The species of free radicals involved in the reactions were identified using the ·OH and SO4· inhibitors tert-butanol and methanol. Tert-butanol and methanol were added to water before PS addition at a concentration ratio of inhibitor to PS of 50:1 and 100:1, respectively. After a reaction time of 30 min, water samples were stored at 4 °C before analysis.

Free radical capture and recognition

Qualitative detection of free radicals in the reaction system used an E500-9.5/12 electronic paramagnetic resonance (EPR) spectrometer. The free radical trapping agent was 5,5-dimethyl-1-pyrroline N-oxide (DMPO).

Quantitation of free radicals

Free radicals were quantified by determining the content of SO4· produced in the system using p-hydroxybenzoic acid (HBA) and the SO4· oxidation product benzoquinone (ρ-BQ). HBA (200 mg) was added to 1 mL of ethanol and diluted to 50 mg/L with pure water. PS was added to 500 mL of this solution for 30 min, before a quencher was added and the solution sampled. An LC-20A high performance liquid chromatograph was used to determine ρ-BQ concentrations.

Other analytical methods

A TOC-VCPH total organic carbon analyzer (Shimadzu Co., Japan) was used to determine DOC in water. The molecular weight distribution of organic matter in water was established by gel chromatography. Determination of water samples was by PL-GPC 50 gel permeation chromatograph (Agilent Technologies Co., USA). A total of three Ultrahydrogel water soluble columns, column length 300 mm, were used. The mobile phase flow rate is 0.3 mL/min, the flow phase parameter is: 0.002 M (Na2HPO4), 0.1 M (NaCl), 0.002 M (KH2PO4). Water samples were filtered through a 0.45 μm membrane before analysis.

RESULTS AND DISCUSSION

Removal of ARGs by PS and UF

Comparison of PS activation methods

Oxidation treatment of secondary effluent was carried out by means of PS only, nFe/PS, and US/PS. The efficiency of removal of different types of ARGs, intI1 and 16S rRNA was initially assessed after oxidation under various concentrations of PS alone. When the optimum PS concentration was determined, PS was activated by nFe and US. The oxidation effects on ARGs, intI1 and 16S rRNA by different nano-iron concentrations and ultrasonic frequencies were compared to determine the optimum activation conditions of PS by nFe and US (Figure 1).

Figure 1

Removal effect of PS and different activation methods on ARGs in secondary effluent (a) PS; (b) nFe/PS; (c) US/PS. Note: Quantitative fluorescence PCR was used to detect tetA, tetC, sulI, sulII, intI1 and 16S rRNA genes concentration, and the error bars signify technical triplicate qPCR reactions.

Figure 1

Removal effect of PS and different activation methods on ARGs in secondary effluent (a) PS; (b) nFe/PS; (c) US/PS. Note: Quantitative fluorescence PCR was used to detect tetA, tetC, sulI, sulII, intI1 and 16S rRNA genes concentration, and the error bars signify technical triplicate qPCR reactions.

It can be seen from Figure 1 that the concentration of ARGs in water decreased significantly after PS was added. At the optimum PS concentration of 4 mM, the concentrations of tet A, tet C, sul I, sul II, int I 1, and 16S rRNA in raw water decreased, respectively, from 1.26 × 105 copies/mL, 1.38 × 106 copies/mL, 2.95 × 106 copies/mL, 1.45 × 106 copies/mL, 3.90 × 107 copies/mL, and 1.74 × 108 copies/mL, to 2.40 × 104 copies/mL, 5.16 × 104 copies/mL, 8.95 × 105 copies/mL, 4.27 × 105 copies/mL, 1.51 × 106 copies/mL, and 1.64 × 107 copies/mL. After PS was activated by nFe and US, the removal of ARGs was significantly enhanced. Activation of PS was greatest at an nFe dose of 2 mM. The removal of ARGs increased by 49.0%, 46.4%, 46.1%, 48.0%, 23.1%, and 24.1%, respectively, under nFe/PS compared with PS alone. The activation of PS was greatest when the US frequency was 40 kHz. US/PS also increased ARGs removal by 49.1%, 46.3%, 30.1%, 26.8%, 18.5%, and 22.5%, respectively, compared with PS alone. This can be explained by, firstly, after addition of PS, SO4· is formed that can oxidize ARGs, while the addition of excessive PS will inhibit the oxidative effect of SO4· (Zhang 2018; Zhang et al. 2019b). Secondly, after addition of nFe, Fe0 activates PS and more SO4· is produced during the transition of the three ions Fe0, Fe2+, and Fe3+. At the same time, due to the presence of Fe ions, ·OH is also generated and cooperates with SO4· to remove ARGs. When nFe reaches a certain concentration, SO4· will react with Fe2+ formed by the reaction, thereby inhibiting the oxidative effect (Long et al. 2019). Under the action of US, sonochemical effects (cavitation, high temperature and high pressure) occur in the water, resulting in the formation of more SO4· (Lu 2018), which enhances the removal of ARGs by PS.

Removal efficiency of ARGs by combined processes

This study combines different persulfate activation methods and UF to investigate the optimum activation conditions (PS = 4 mM, nFe = 2 mM, US frequency = 40 kHz) for a combined process to remove ARGs from water (Figure 2 and Table 3).

Table 3

Removal of ARGs in secondary effluent by different combined processes(-log)

tet Atet Csul Isul IIint I 116S rRNA
UF 0.73 0.41 0.38 0.57 0.41 0.44 
PS-UF 2.03 3.11 3.02 2.90 2.22 2.39 
nFe/PS-UF 2.51 3.20 3.52 3.24 2.61 2.51 
US/PS-UF 2.06 3.11 3.39 2.98 1.59 2.44 
tet Atet Csul Isul IIint I 116S rRNA
UF 0.73 0.41 0.38 0.57 0.41 0.44 
PS-UF 2.03 3.11 3.02 2.90 2.22 2.39 
nFe/PS-UF 2.51 3.20 3.52 3.24 2.61 2.51 
US/PS-UF 2.06 3.11 3.39 2.98 1.59 2.44 

Note: PS = 4 mM, nFe = 2 mM, US frequency = 40 kHz.

Figure 2

Removal effect of different combined processes on ARGs in water (PS = 4 mM, nFe = 2 mM, US frequency = 40 kHz). Note: Quantitative fluorescence PCR was used to detect tetA, tetC, sulI, sulII, intI1 and 16S rRNA genes concentration, and the error bars signify technical triplicate qPCR reactions.

Figure 2

Removal effect of different combined processes on ARGs in water (PS = 4 mM, nFe = 2 mM, US frequency = 40 kHz). Note: Quantitative fluorescence PCR was used to detect tetA, tetC, sulI, sulII, intI1 and 16S rRNA genes concentration, and the error bars signify technical triplicate qPCR reactions.

Figure 2 and Table 3 illustrate that after PS was activated by nFe and US, the combined process had a significantly higher ARGs removal efficiency than UF alone, and that the activation effect of nFe was greater than with US. The amount of the ARGs tet A, tet C, sulI, and sulII removed by UF alone was limited: 0.73-log, 0.41-log, 0.38-log, and 0.57-log, respectively. Using a combination of PS and UF, the removal of the above ARGs significantly increased, reaching 2.03-log, 3.11-log, 3.02-log, and 2.90-log, respectively. The removal of ARGs by the nFe/PS-UF combined process was 2.51-log, 3.20-log, 3.52-log, and 3.24-log, respectively. Under the US/PS-UF combined process, ARGs removal was 2.06-log, 3.11-log, 1.39-log, and 2.98-log, respectively.

In summary, PS oxidation pretreatment significantly improved the removal effect of UF on ARGs due to release of SO4·. However, due to the limited oxidation capacity of PS, removal of ARGs was further improved by activation of PS by nFe and US. This is due to more SO4·− being produced after activation, ·OH being produced via the change of Fe valence states, and the cavitation caused by US. Furthermore, since nFe particles form a loose and porous oxide layer on the surface of the filter membrane during the oxidation process, the nFe/PS-UF combined process can adsorb and remove more ARGs from water.

Removal of cell-associated and cell-free ARGs from membrane influent and effluent

Through the elimination of free state ARGs by DNA degradation enzyme, the existence and identity of various ARGs in UF membrane influent and effluent were analyzed following the three combined processes (Figure 3).

Figure 3

Cell-associated and cell-free ARGs concentrations in membrane influent and effluent in different combined processes (a) direct UF; (b) PS-UF; (c) nFe/PS-UF; (d) US/PS-UF. Note: Quantitative fluorescence PCR was used to detect ARGs concentration, and the error bars signify technical triplicate qPCR reactions.

Figure 3

Cell-associated and cell-free ARGs concentrations in membrane influent and effluent in different combined processes (a) direct UF; (b) PS-UF; (c) nFe/PS-UF; (d) US/PS-UF. Note: Quantitative fluorescence PCR was used to detect ARGs concentration, and the error bars signify technical triplicate qPCR reactions.

It can be seen from Figure 3 that the concentrations of cell-associated tet A, tet C, sulI and sulII in raw water were 4.73-log, 5.48-log, 6.01-log, and 5.72-log, respectively. The concentrations of cell-free ARGs were higher, specifically 4.86-log, 6.03-log, 6.29-log, and 5.96-log, respectively. After UF, the concentrations of ARGs in both types of water sample were reduced and the removal of cell-associated ARGs was enhanced. The reductions of cell-associated tet A, tet C, sulI and sulII were 1.05-log, 0.60-log, 0.81-log, and 0.86-log, respectively. The reductions of the ARGs in the free state were 0.59-log, 0.37-log, 0.25-log, and 0.46-log, respectively. This may be due to the membrane pore size: UF retaining more of the cell-associated ARGs. After PS, nFe/PS, and US/PS pre-oxidation, the concentrations of ARGs in the cell-associated and cell-free states in secondary effluent water were lowered. The concentrations of cell-associated and cell-free ARGs in membrane influent and effluent were lowest following the nFe/PS-UF combined process. Reductions of the above cell-associated ARGs by the nFe/PS-UF combined process were 1.84-log, 1.74-log, 3.14-log, and 2.80-log, while cell-free ARGs reductions were 1.27-log, 1.00-log, 2.61-log, and 2.30-log, respectively.

This can be explained by, firstly, SO4· released by PS destroying the cell membrane structure of microorganisms. When the cell-associated ARGs are released into water, entering a free state, they undergo oxidation by SO4· and ·OH and physical retention by UF. Secondly, after PS is activated by nFe and US, the concentration of SO4· and ·OH in the water increases, oxidation is significantly improved, and the concentration of cell-associated and cell-free ARGs is reduced (Yan et al. 2010). Thirdly, during the activation process, nFe forms a loose and porous oxide shell on the surface of the particles (Denis et al. 2013), which can adsorb ARGs. Furthermore, cavitation during US treatment can lyse the cell structure, transforming cell-associated ARGs into the free state, thereby reducing the concentration of cell-associated ARGs (Liu 2018; Wang 2018). nFe/PS and US/PS pre-oxidation processes improve the removal of cell-associated and cell-free ARGs by UF.

Removal of organic matter by combined processes

DOC removal efficiency

During the test, DOC concentrations were determined for UF membrane influent and effluent under the optimal parameters for the combined processes PS-UF, nFe/PS-UF, and US/PS-UF. Results are shown in Figure 4.

Figure 4

Removal of DOC in secondary effluent by different combined processes (PS = 4 mM, nFe = 2 mM, US frequency = 40 kHz). Note: A TOC-VCPH total organic carbon analyzer was used to determine DOC concentration in water, and the error bars signify technical triplicate DOC detection.

Figure 4

Removal of DOC in secondary effluent by different combined processes (PS = 4 mM, nFe = 2 mM, US frequency = 40 kHz). Note: A TOC-VCPH total organic carbon analyzer was used to determine DOC concentration in water, and the error bars signify technical triplicate DOC detection.

The concentration of DOC in the secondary effluent was 7.86 mg/L. After UF treatment, DOC was reduced by 12.0% to 6.92 mg/L. Under the combination processes PS-UF, nFe/PS-UF, and US/PS-UF, membrane influent DOC concentrations were 6.98, 6.51, and 6.66 mg/L, respectively, while concentrations in membrane effluent were 6.00, 5.51, and 5.74 mg/L. Thus, the rate of DOC removal increased to 23.8, 30.0, and 27.1%, respectively, showing that all three pre-oxidation processes can effectively improve the efficacy of UF. This is due to the SO4· generated in the pre-oxidation processes oxidizing and degrading the organic matter. Following activation by nFe and US, PS can release more SO4· and ·OH is generated. The synergistic effect of the two free radicals enhances the removal of DOC.

Efficiency of combined processes for removal of varying molecular weight organic matter

Raw water and membrane effluent from UF, PS-UF, nFe/PS-UF, and US/PS-UF treatments, under optimum conditions (PS = 4 mM, nFe = 2 mM, US frequency = 40 kHz), were subjected to gel chromatography to determine the efficiency of removal of organic matter with different molecular weights. Results are shown in Figure 5 and Table 4.

Table 4

Peak area and removal efficiency of different molecular weight intervals in water samples after different combined processes for secondary effluent treatment

Peak area (>100 K Da)Removal ratePeak area (50 K–100 K Da)Removal ratePeak area (10 K–50 K Da)Removal ratePeak area (<10 K Da)Removal rate
Raw water 28.37  38.01  18.38  11.89  
UF membrane effluent 15.19 46.4% 37.92 0.2% 17.96 2.3% 10.75 9.6% 
PS-UF membrane effluent 12.71 55.2% 30.79 19.0% 19.26 −4.8% 5.91 50.3% 
nFe/PS-UF membrane effluent 6.05 78.7% 28.91 23.9% 19.97 −8.6% 6.59 44.6% 
US/PS-UF membrane effluent 8.50 70.0% 30.78 19.0% 19.77 −7.6% 6.56 44.8% 
Peak area (>100 K Da)Removal ratePeak area (50 K–100 K Da)Removal ratePeak area (10 K–50 K Da)Removal ratePeak area (<10 K Da)Removal rate
Raw water 28.37  38.01  18.38  11.89  
UF membrane effluent 15.19 46.4% 37.92 0.2% 17.96 2.3% 10.75 9.6% 
PS-UF membrane effluent 12.71 55.2% 30.79 19.0% 19.26 −4.8% 5.91 50.3% 
nFe/PS-UF membrane effluent 6.05 78.7% 28.91 23.9% 19.97 −8.6% 6.59 44.6% 
US/PS-UF membrane effluent 8.50 70.0% 30.78 19.0% 19.77 −7.6% 6.56 44.8% 
Figure 5

Molecular weight distribution of organic matters in secondary effluent and membrane effluent from different combined processes. Note: The molecular weight distribution of organic matter in water was established by gel chromatography.

Figure 5

Molecular weight distribution of organic matters in secondary effluent and membrane effluent from different combined processes. Note: The molecular weight distribution of organic matter in water was established by gel chromatography.

Organic matter in secondary effluent was found in various molecular weight ranges (>100 kDa, 50–100 kDa, 10–50 kDa, and <10 kDa). The removal rates of organic matter in these ranges by UF were 46.4, 0.2, 2.3, and 9.6%, respectively. Removal of organic matter in the ranges >100 kDa, 50–100 kDa, and <10 kDa was improved under the combined processes PS-UF, nFe/PS-UF, and US/PS-UF, with the largest increase being seen in the >100 kDa range. However, the removal of organic matter in the range 10–50 kDa decreased by 4.8, 8.6 and 7.6%, respectively. This is because the SO4· and ·OH released by PS during the pre-oxidation process can oxidize and degrade large molecular weight compounds (>100 kDa) into medium and small molecular weight compounds. They also oxidize and degrade organic matter below 10 kDa. As the reaction progresses, the concentrations of SO4· and ·OH in the water decrease. After organic matter in the ranges >100 kDa and 50–100 kDa is oxidized to 10–50 kDa, oxidation and degradation cannot continue, resulting in an increase in the concentration of organic matter in this range (Liu 2016; Su 2018).

Correlation analysis between ARGs and DOC

In order to better explore the influence of the removal of organic matter in the secondary effluent on the removal of ARGs, a linear fitting analysis of the concentration of DOC and ARGs was carried out during the test, and the combined processes of PS-UF, nFe/PS-UF and US/PS-UF were studied. Under different test conditions, the correlation between the DOC concentration in the membrane effluent and the content of different types of ARGs (significant correlation basis is p < 0.05). Results are shown in Figure 6.

Figure 6

Correlation between DOC concentration and ARGs concentration in different combined processes (a) PS-UF; (b) nFe/PS-UF; (c) US/PS-UF. Note: Statistical Product and Service Solutions (SPSS) was used to analyze the significance of correlation.

Figure 6

Correlation between DOC concentration and ARGs concentration in different combined processes (a) PS-UF; (b) nFe/PS-UF; (c) US/PS-UF. Note: Statistical Product and Service Solutions (SPSS) was used to analyze the significance of correlation.

It can be seen from Figure 6 that in the PS-UF, nFe/PS-UF, and US/PS-UF combined processes, there is a significant correlation between the concentration of DOC and the four ARGs, tetA, tetC, sulI, and sulII (p < 0.05). The PS-UF fitting results R2 were 0.956, 0.911, 0.906, 0.870, respectively. The nFe/PS-UF fitting results R2 were 0.892, 0.826, 0.894, 0.911, respectively. The US/PS-UF fitting results R2 were 0.993, 0.974, 0.993, 0.997, respectively. The results show that the removal of DOC in water by three different combination processes of PS-UF, nFe/PS-UF, and US/PS-UF could promote the reduction of ARGs concentration. This is because the pre-oxidation process of PS, nFe/PS, US/PS alone could oxidize organic matter by generating SO4·, while nFe particles are activated, they will also remove organic matter through their own adsorption; at the same time, organic matter in the water will also be degraded under the action of ultrasound. Therefore, the removal of organic matter in water is conducive to the reduction of ARGs concentration.

Mechanism of action of free radicals in combined processes

Free radical identification in combined processes

To determine the free radicals involved in oxidation during the different processes, tert-butanol and methanol were used as free radical inhibitors in quenching experiments. ARGs concentration changes in water after pre-oxidation by PS, nFe/PS, and US/PS were measured after adding various concentrations of inhibitors. Results are shown in Figure 7.

Figure 7

Effect of tert-butanol and methanol on removal of ARGs by different pre-oxidation processes (a) PS; (b) nFe/PS; (c) US/PS. Note: Quantitative fluorescence PCR was used to detect ARGs concentration, and the error bars signify technical triplicate qPCR reactions.

Figure 7

Effect of tert-butanol and methanol on removal of ARGs by different pre-oxidation processes (a) PS; (b) nFe/PS; (c) US/PS. Note: Quantitative fluorescence PCR was used to detect ARGs concentration, and the error bars signify technical triplicate qPCR reactions.

When 0.1 M methanol was added to water, the concentrations of tetA, tetC, sulI, and sulII ARGs increased to 4.94-log, 6.01-log, 6.34-log, and 6.03-log, respectively, following PS pre-oxidation. When 0.2 M methanol was added, concentrations increased further to 5.02-log, 5.96-log, 6.38-log, and 6.09-log, respectively. When 0.1 M tert-butanol was added, ARGs concentrations after pre-oxidation were 4.49-log, 4.90-log, 6.02-log, and 5.66-log, respectively. With 0.2 M tert-butanol, concentrations rose to 4.56-log, 5.05-log, 6.08-log, and 5.72-log, respectively. Results of nFe/PS and US/PS pre-oxidation followed a similar pattern, indicating that both tert-butanol and methanol inhibit the removal of ARGs from secondary effluent under all three pre-oxidation treatments. The larger the dose, the greater the inhibitory effect. Methanol inhibited the removal of all four ARGs better than tert-butanol. Tert-butanol consumes ·OH in the water, inhibiting the removal of ARGs. However, addition of methanol inhibits the removal of ARGs further, indicating that in addition to ·OH, SO4· participated in the oxidation reaction.

Free radical capture and identification by DMPO and EPR

To further identify the free radicals reacting involved in the reaction in the three oxidation processes of PS, nFe/PS and US/PS, DMPO was used to capture the free radicals and water samples were scanned by EPR spectroscopy. An EPR spectrum is shown in Figure 8.

Figure 8

EPR spectra of radical capture by different pre-oxidation processes. Note: Qualitative detection of free radicals in the reaction system used an E500-9.5/12 electronic paramagnetic resonance (EPR) spectrometer.

Figure 8

EPR spectra of radical capture by different pre-oxidation processes. Note: Qualitative detection of free radicals in the reaction system used an E500-9.5/12 electronic paramagnetic resonance (EPR) spectrometer.

Under the three pre-oxidation processes PS, nFe/PS, and US/PS, characteristic peaks for DMPO-SO4 (intensity ratio 1:1:1:1:1:1, green triangle) and DMPO-OH (intensity ratio 1:2:2:1, pink circle) were evident (Xia et al. 2017; Wu et al. 2019). This shows that in all three pre-oxidation processes both free radicals SO4· and ·OH participate in the reaction. When the PS was pre-oxidized, the characteristic peaks were smaller, indicating that the concentration of SO4· and ·OH was lowered by the oxidation process. This explains why the PS-UF combined process demonstrated improved ARGs removal capacity compared with UF alone, but that this removal capacity was limited. After PS was activated by US, the intensities of the characteristic peaks of SO4· and ·OH were significantly higher, demonstrating the activation effect of US. When PS was activated by nFe, the SO4· and ·OH peaks were also higher, showing the activation effect of nFe on PS. The highest concentrations of SO4· and ·OH and the largest ARGs removal capacity were seen following the nFe/PS-UF combined process.

Concentration of SO4· generated by the pre-oxidation process

When PS and its activation products were added to water, SO4· was produced and further converted to ·OH under different oxidation processes. The concentration of SO4· radicals produced by the three pre-oxidation processes (PS, nFe/PS, and US/PS) under the optimal dosages of PS, nFe and US are shown in Table 5.

Table 5

ρ-BQ and SO4· concentration in different pre-oxidation processes

Pre-oxidation methodPeak height (mAU)ρ-BQ concentration (mg/L)SO4·concentration (μM)
PS 1860.6 17.5 161.8 
nFe/PS 2444.3 22.9 212.3 
US/PS 2166.2 20.3 188.2 
Pre-oxidation methodPeak height (mAU)ρ-BQ concentration (mg/L)SO4·concentration (μM)
PS 1860.6 17.5 161.8 
nFe/PS 2444.3 22.9 212.3 
US/PS 2166.2 20.3 188.2 

Note: PS = 4 mM, nFe = 2 mM, US frequency = 40 kHz.

Each oxidation process produced a different SO4· concentration in water. Pre-oxidation with PS alone produced a SO4· concentration of 161.8 μM. When PS was activated by nFe and US, SO4· concentration increased to 212.3 and 188.2 μM, respectively. Considering the three combined processes for removing ARGs from secondary effluent, it is clear that the higher the SO4· concentration in the water, the better the removal of ARGs. This results from PS reacting with water to generate SO4· radicals, which then further react with water to generate ·OH, and the two free radicals jointly oxidize the ARGs. When nFe is added, Fe0 reacts with PS to form Fe2+, which then reacts with PS to form SO4·, which proceed to oxidatively degrade the ARGs. Fe2+ is also converted to Fe3+, which can react with the Fe0, generating Fe2+ and continuing to activate PS. This continuous cycle of conversion of Fe2+ and Fe3+ is the main reason why the activation effect of nFe is greater than that of US. US activation is caused by cavitation, which provides thermodynamic energy to promote production of SO4· from PS (Shen 2014).

CONCLUSIONS

  • (1)

    The pre-oxidation processes PS, nFe/PS, and US/PS can effectively oxidize and remove ARGs from secondary effluent. The optimum concentration of PS was 4 mM, nFe was 2 mM, and the best US frequency was 40 kHz. At this time, the ARGs tet A, tet C, sul I, and sul II were reduced in secondary effluent after the three combined processes by 0.66–2.06-log, 2.51–3.52-log, and 1.39–3.11-log, respectively. In all cases, the removal of free ARGs was significantly higher than when UF alone was used. The greatest reduction of cell-associated and cell-free ARGs was seen following nFe/PS-UF treatment, being 1.74–3.14-log and 1.00–2.61-log, respectively.

  • (2)

    The removal efficiency of DOC by the three combined processes was significantly better than that of UF alone, with the highest removal rate (30.0%) following nFe/PS-UF treatment. All three pre-oxidation methods significantly enhanced the capacity of UF to remove organic matter in the molecular weight ranges >100 kDa, 50–100 kDa, and <10 kDa. The range >100 kDa was the most improved. However, removal of DOC in the 10–50 kDa range decreased under the combined processes.

  • (3)

    Free radical quench testing and EPR spectrometry showed that the free radicals SO4· and ·OH were involved in the pre-oxidation processes and they synergistically removed ARGs from water. When oxidant concentrations were optimized, SO4· concentrations in the water after the oxidation processes (PS, nFe/PS, and US/PS) were 161.8 μM, 212.3 μM, and 188.2 μM, respectively. nFe had the better activation effect on PS, leading to the highest SO4· concentration in water.

ACKNOWLEDGEMENTS

The research was supported by the National Natural Science Foundation of China (Grant No. 52070011, No.51678027 and No. 51678026).

DATA AVAILABILITY STATEMENT

All relevant data are included in the paper or its Supplementary Information.

REFERENCES

REFERENCES
Cai
P.
Huang
Q.
Zhang
X.
2006
Interactions of DNA with clay minerals and soil colloidal particles and protection against degradation by DNase
.
Environmental Science & Technology
40
(
9
),
2971
2976
.
Chen
S. J.
2019
A Study on the Distribution of Integrons and Drug Resistance Gene Islands in Proteus Mirabilis
.
Guangdong Pharmaceutical University
,
Guangdong
,
China
.
Chen
L.
George
Z. Z.
2018
Removal techniques of antibiotic resistant bacteria and resistant genes in sewage
.
Ecology and Environmental Sciences
27
(
11
),
2163
2169
.
Denis
O. C.
Brent
S.
Magdalena
K.
Hardiljeet
B.
Christopher
K.
2013
Nanoscale zero valent iron and bimetallic particles for contaminated site remediation
.
Advances in Water Resources
51
,
104
122
.
Enne
V. I.
Livermore
D. M.
Stephens
P.
Hall
L. M. C.
2001
Persistence of sulphonamide resistance in Escherichia coliin the UK despite national prescribing restriction
.
The Lancet
357
(
9265
),
1325
1328
.
Fan
Q.
Gu
P.
Yuan
Y. L.
Zhang
G. H.
2014
Research progress in the regeneration technologies of powdered activated carbon
.
Industrial Water Treatment
34
(
04
),
1
4
.
Gong
M. Y.
2019
Efficency of Mn(III) Pre-Oxidation Enhancing Gravity Driven Membrane to Remove Atrazine Water
.
Harbin Institute of Technology
,
Harbin
,
China
.
He
G. R.
2017
Study on Removal of Amoxicillin and Carbamazepine by Sono-Fe0-Activated Persulfate System
.
Huazhong University of Science and Technology
,
Hubei
,
China
.
Jiao
Y. N.
2017
The Distribution Characteristic and Mechanism of Antibiotic Resistome Between Dyeing and Domestic Wastewater
.
Zhejiang University
,
Zhejiang
,
China
.
Kolthoff
I. M.
Miller
I. K.
1951
The chemistry of persulfate. I. The kinetics and mechanism of the decomposition of the persulfate ion in aqueous medium1
.
Journal of the American Chemical Society
73
(
7
),
3055
3059
.
Li
J. F.
2015
Microbial Diversity Research Based on High-Throughput Sequencing Data of 16S rRNA and Metagenome
.
Tsinghua University
,
Beijing
,
China
.
Li
B. Z.
Zhu
J.
2019
Oxidative degradation of perfluorooctane sulfonate by a sono-activated persulfate system in an aqueous solution
.
Journal of Agro-Environment Science
38
(
04
),
912
920
.
Liu
Q. L.
2016
Efficiency and Mechanism of Microcystis Aeruginosa Removal by Persulfate
.
Harbin Institute of Technology
,
Harbin
,
China
.
Liu
Z. F.
2018
Removal of Resistant Bacteria from Wastewater and its Mechanism of Resistance Reduction
.
Nanchang Hangkong University
,
Jiangxi
,
China
.
Long
H. B.
Gao
H. F.
Li
C.
2019
Review on remediation of petroleum hydrocarbon contaminated soil with iron activated persulfate
.
Environmental Protection of Chemical Industry
39
(
03
),
241
246
.
Lu
X. H.
2018
Ultrasound and Heat Activated Persulfate for the Degradation of Bisphenols in Water
.
Southwest Jiaotong University
,
Sichuan
,
China
.
Luo
X.
Yuan
L. X.
Zhang
W. L.
Zhong
W. Z.
Jiang
Y. F.
Zhang
Y.
Xu
D. S.
2019
Correlation study between resistance genes and microbial communities in pharmaceutical wastewater treatment plants
.
China Environmental Science
39
(
2
),
831
838
.
Niemeyer
J.
Gessler
F.
2002
Determnination of free DNA in soils
.
Journal of Plant Nutrition And Soil Science
165
(
2
),
121
124
.
Nunes
S. P.
Peinemann
K. V.
1992
Ultrafiltration membranes from PVDF/PMMA blends
.
Journal of Membrane Science
73
(
1
),
25
35
.
Pruden
A.
Pei
R.
Storteboom
H.
Carlson
K. H.
2006
Antibiotic resistance genes as emerging contaminants: studies in Northern Colorado
.
Environmental Science & Technology
40
(
23
),
7445
7450
.
Shen
Y.
2014
Theoretical Research on Ultrasonic Cavitation and Simulation Analysis of Affecting Factors
.
Northeastern University
,
Liaoning
,
China
.
Shi
Z.
Bi
C.
Zhou
S. Q.
Bu
L. J.
2017
Degradation of iron electrode as sodium persulfate activator for orange G
.
Chinese Journal of Environmental Engineering
11
(
03
),
1335
1340
.
Su
Y.
2018
Remediation of Sulfanilic Acid Wastewater with Activation of Persulfate Based on Iron/Copper Nanocomposite and Heat
.
Jilin University
,
Jilin
,
China
.
Wang
M. L.
2018
Distribution and Removal of Typical Antibiotic Resistance Genes During Sludge Pretreatment and Anaerobic Digestion Processes
.
Tianjin University
,
Tianjin
,
China
.
Wen
H. Q.
Shi
J.
Xun
H.
Deng
H. P.
2015
Distribution,dissemination and removal of antibiotic resistant genes (ARGs) in the aquatic environment
.
Chinese Journal of Applied Ecology
26
(
02
),
625
635
.
Wu
Y.
Chen
X.
Han
Y.
Yue
D. T.
Cao
X. D.
Zhao
Y. X.
Qian
X. F.
2019
Highly efficient utilization of Nano-Fe(0) embedded in mesoporous carbon for activation of peroxydisulfate
.
Environmental Science & Technology
53
(
15
),
9081
9090
.
Xia
D.
Yin
R.
Sun
J.
An
T. C.
Li
G. Y.
Wang
W. J.
Zhao
H. J.
Po
K. W.
2017
Natural magnetic pyrrhotite as a high-Efficient persulfate activator for micropollutants degradation: radicals identification and toxicity evaluation
.
Journal of Hazardous Materials
340
,
435
444
.
Xiao
Y.
2016
Study on Phenol Wastewater Treatment by Activated Persulfate Oxidation Method
.
Shenyang Jianzhu University
,
Liaoning
,
China
.
Yan
W.
Herzing
A. A.
Kiely
C. J.
Zhang
W. X.
2010
Nanoscale zero-valent iron (nZVI): aspects of the core-shell structure and reactions with inorganic species in water
.
Journal of Contaminant Hydrology
118
(
3
),
96
104
.
Zhang
M.
2018
The Degradation of Iodinated Pollutants in Water by Ferrous/Persulfate Advanced Oxidation Technology
.
Taiyuan University of Technology
,
Shanxi
,
China
.
Zhang
X.
Zhang
T.
Fang
H. H. P.
2009
Antibiotic resistance genes in water environment
.
Applied Microbiology and Biotechnology
82
(
3
),
397
414
.
Zhang
Y. J.
Li
D. P.
Cao
T. J.
2012
Research on the oxidation degradation of phenol by activating ammonium persulfate with zero-valent iron
.
Industrial Water Treatment
32
(
09
),
13
16
.
Zhang
Y. X.
Liu
H. L.
Xin
Y. P.
Shen
Y. P.
Wan
J.
Cai
C.
Wang
M. M.
2019a
Erythromycin degradation and ERY-resistant gene inactivation in erythromycin mycelial dreg by heat-activated persulfate oxidation
.
Chemical Engineering Journal
358
,
1446
1453
.
Zhang
Y.
Gao
C. Y.
Chen
C. Z.
2019b
Degradation of PAHs in contaminated soil by zero valent iron activated sodium persulfate system
.
Chinese Journal of Environmental Engineering
13
(
04
),
955
962
.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY-NC-ND 4.0), which permits copying and redistribution for non-commercial purposes with no derivatives, provided the original work is properly cited (http://creativecommons.org/licenses/by-nc-nd/4.0/).