Antibiotics release into the water environment through sewage discharge is a significant environmental concern. In the present study, we investigated the removal of ciprofloxacin (CIP) in simulated sewage by biological aeration filter (BAF) equipped with Fe3O4-modified zeolite (Fe3O4@ZF). Fe3O4@ZF were prepared with impregnation method, and the Fe3O4 particles were successfully deposited on the surface of ZF in an amorphous form according to the results of XPS and XRD analysis. The modification also increased the specific surface area (from 16.22 m²/g to 22 m²/g) and pore volume (from 0.0047 cm³/g to 0.0063 cm³/g), improving the adsorption efficiency of antibiotics. Fe3O4 modified ZF improved the treatment performance significantly, and the removal efficiency of CIP in BAF-Fe3O4@ZF was 79%±2.4%. At 10ml/L CIP, the BAF-Fe3O4@ZF reduced the relative abundances of antibiotics resistance genes (ARGs) int, mexA, qnrB and qnrS in the effluent by 57.16%, 39.59%, 60.22%, and 20.25%, respectively, which effectively mitigate the dissemination risk of ARGs. The modification of ZF increased CIP-degrading bacteria abundance, such as Rhizobium and Deinococcus-Thermus, and doubled bacterial ATP activity, promoting CIP degradation. This study offers a viable, efficient method to enhance antibiotic treatment and prevent leakage via sewage discharge.

  • Fe3O4 modification can improve the removal efficiency of CIP by BAF.

  • Fe3O4@ZF reduced the released resistance genes in the effluent.

  • Fe3O4@ZF enriched the abundance of CIP-degrading bacteria.

  • Fe3O4@ZF increased the microbial activity on the fillers rather than the biomass.

The horizontal gene transfer (HGT) of antibiotic-resistant genes (ARGs), which may create resistant pathogens that make patients difficult to treat in clinical settings, have become a global crisis of environmental safety (Cacace et al. 2019; Khan et al. 2019). It is essential to remove the antibiotics from sewage for the safe management of antibiotic crisis, but the microbial inhibition limited the treatment performance of antibiotics in wastewater treatment plant (WWTPs) (Yi et al. 2017; Yao et al. 2021). The removal efficiency of erythromycin in mixed municipal and hospital sewage in Hangzhou was even only 1–2%, and the removal efficiency of cefazolin was 63–72% (Kato et al. 2012). It is urgent to develop more efficient method for the treatment of antibiotics in WWTPs and has been serious concerned by the environmental researchers and managers.

Biological aeration filter (BAF), an efficient biofilm water treatment combined with physical filtration, is a promising technology for the treatment of antibiotics in WWTPs. Their enriched microbial population on the filter material cultivates functional flora for intensive antibiotic treatment, boosting efficiency and reducing antibiotic inhibition on microbes (Marsidi et al. 2018). Yang et al. (2016) found that an up-flow anaerobic/aeration filter system with sludge and cinder ceramic particles could remove 90% of tetracycline. Researchers also modified the surface characteristics of fillers to enhance the treatment performance of antibiotics by BAF and iron oxides are the most common used materials. BAF equipped with iron oxide-based porous ceramsite had superiority in removing ammonia nitrogen, total nitrogen (TN), total organic carbon, and phosphorus (Bao et al. 2017). Iron oxides may facilitate electron exchange between dissimilating Fe(III)-reducing bacteria and Fe(II)-oxidizing photosynthetic bacteria, which could aid antibiotic degradation (Byrne et al. 2015). However, the mechanisms behind filter modification and preventing ARG leakage in BAFs, crucial for minimizing HGT, remain underexplored.

The present study aims to achieve several key objectives. First, we aim to develop a straightforward method for modifying zeolite (ZF) with Fe3O4, resulting in the creation of Fe3O4@ZF. Second, we aim to investigate the treatment performance of ciprofloxacin (CIP), a widely used and representative antibiotic, using a BAF filled with the modified ZF filler. This investigation will focus on the removal efficiency of CIP and the management of ARGs. To provide a comparative analysis, the performances of BAFs filled with unmodified ZF and Fe2O3-modified zeolite (Fe2O3@ZF) will also be studied. Furthermore, genomics and transcriptomics analyses are performed to reveal the alterations in community structure and metabolic state in the presence of Fe3O4. Finally, the present study may provide researchers and engineers with a practical and effective method to mitigate the risk of antibiotic release through wastewater discharge.

Chemicals

An analytical grade of CIP was provided by Macklin Co. (Shanghai, China). ZF was obtained from Hongxing Material Factory (Gongyi, China). Other chemicals, including potassium persulfate, sulfanilamide, ascorbic acid, glucose, NaOH, KNaC4H12O10·4H2O, NH4Cl, CaCl2, KH2PO4, and FeCl3, were of analytical grade and provided by local suppliers.

Modification of ZF

Pre-treatment of ZF

The ZF fragments with the diameter of 2–4 mm were screened with a Tyler standard sieve of 6 mesh and washed several times with deionized water (DI) water to remove ash and impurities. Then, the ZF was soaked in 0.1 mol/L hydrochloric acid solution for 24 h, and then rinsed with DI water until the pH of the effluent is neutral.

Preparation of nano-Fe3O4@ZF

The pretreated ZF was mixed with diluted commercial nano-sized Fe3O4 dispersion (1 mL/g) at a volume ratio of 1:1. The mixture was placed in a customized sealing device for stirring and mixing evenly under nitrogen protection conditions and dried at 100 °C in an oil bath pot. The dried ZF was calcined in a tubular furnace with nitrogen protection at 250 °C for 4 h, after which the ZF was washed with DI water to rinse off unloaded Fe3O4 particles and eventually put into oven at 110 °C for drying.

Preparation of Fe2O3@ZF

The pretreated ZF and FeCl3 (2 mol/L) was mixed and stirred well with a dosing ratio of 0.5 mL/g and dried in oven at 110 °C for 24 h without any stirring. The mixture was calcined in muffle at 250 °C for 3 h. The chilled ZF was wash with DI water to rinse off unloaded Fe2O3 particles and eventually put into oven at 110 °C for drying.

Reactor design and operating conditions

The reactor is divided into two sections, both filled with ZF or modified ZF with a thickness of 11.5 cm. The total height of the reactor is 51 cm, and the inner diameter is 3 cm (Figure 1). Consequently, the total effective volume of the reactor is 255 mL, with a hydraulic retention time (HRT) set at 8 h. These BAFs were filled with Nano-Fe3O4@ZF, Fe2O3@ZF, and ZF, respectively, and were labeled as BAF-Fe3O4@ZF, BAF-Fe2O3@ZF, and BAF-ZF for identification purposes.
Figure 1

(a) Structural diagram of the BAF reactor design and (b) image of actual reactor operation.

Figure 1

(a) Structural diagram of the BAF reactor design and (b) image of actual reactor operation.

Close modal

The wastewater quality simulated the effluent of antibiotics production secondary wastewater treatment. Synthetic wastewater with glucose, NH4Cl, and KH2PO4 was used as an organic influent substrate. The ratio of COD: : was 100:5:1. Table 1 shows the amount of the trace elements as inorganic nutrient were supplemented. As usage of CIP continues to increase, its detection frequency in the environment has also risen significantly. CIP concentrations in water bodies often range from ng/L to mg/L, with concentrations reaching up to 4.35 mg/L or even higher in pharmaceutical wastewater (Wang et al. 2021). CIP was introduced into the influent, and its concentration was gradually increased from 2 to 10 mg/L.

Table 1

Water quality of simulated wastewater

Simulated wastewater (mg/L)
MgSO4 100 KI 0.2 CuSO4·5H20.8 
CaCl2 100 CoCl2·6H20.5 MnCl2·4H20.8 
FeCl3 H3BO3 0.5 ZnCl2 0.2 
COD 200 DO 4.2–5.6   
Simulated wastewater (mg/L)
MgSO4 100 KI 0.2 CuSO4·5H20.8 
CaCl2 100 CoCl2·6H20.5 MnCl2·4H20.8 
FeCl3 H3BO3 0.5 ZnCl2 0.2 
COD 200 DO 4.2–5.6   

Analytical methods

Dissolved oxygen (DO), pH, TN, nitrate nitrogen (), nitroso nitrogen (), ammonia nitrogen (NH3-N), chemical oxygen demand (COD), and total phosphorus (TP) were determined according to the APHA standard methods. CIP was determined using high performance liquid chromatography (HPLC) equipped with a Shim-pack GIST C18-AQHP column (100 × 2.1 mm, 2 μm). The mobile phase consisted of acetonitrile and 0.08% phosphoric acid solution at a ratio of 2:8 (v:v). The flow rate was 1mL/min at a column temperature of 40 °C. The injection volume was 20 μL, and the detection wavelength was set at 278 nm.

The crystallinity of ZF, Fe2O3@ZF and Fe3O4@ZF was identified by X-ray diffraction (XRD) analysis (SHIMADZU, XRD-6100, Japan). The specific surface area and pore size distribution of ZF, Fe2O3@ZF and Fe3O4@ZF were detected by specific surface area analyzer (ASAP 2460, Micrometrics Co., USA) and calculated by Brunauer–Emmett–Teller (BET) method and Barret–Joyner–Halenda (BJH) method, respectively. Surface morphology was observed by scanning electron microscope (SEM) (JEM-2010, FEI Co., USA). The elemental composition and valence of each element were detected by X-ray photoelectron spectroscopy (XPS) (ESCALAB 250Xi, Thermo Fisher Scientific Co., USA). Zeta potential was measured using a Zeta potentiometer (Nano-zs90, Malvern, Shanghai) to analyze their surface charges.

Microbial community analysis

Water sample was filtered with 0.22-μm Millipore polycarbonate membrane for microbe enrichment. The DNA was extracted with the FastDNA spin kit for soil DNA extraction (MP Biomedicals, USA) according to its operational requirements. DNA samples were quantitatively amplified using the 7,300 quantitative PCR system (QuantStdioTM 3 Real-Time PCR System, Thermo Fisher Co., Ltd) for the measurement of resistance genes and 16S rRNA total bacterial genes.

High-throughput sequencing was conducted at Meiji Biomedical Technology Co., Ltd (Shanghai, China). In this study, the V3 and V4 regions of the total bacterial 16sTRNA in the DNA samples were amplified and sequenced to analyze the microbial community structure and diversity. ATP was detected with the BacTiter-Glo™ Bioluminescence Kit (Promega, Beijing, China) according to its operational requirements.

Statistical analysis

Data analysis was performed with Microsoft Excel 2016 (Microsoft, USA), and visualized through Origin 9.0 (OriginLab, USA). Statistical analysis was conducted using SPSS 20 (Chicago, IL, USA). The independent Student's t-test was employed to examine statistically significant differences in mean values between two independent groups. The significance level was set at p = 0.05.

Characteristics of Fe2O3@ZF and Fe3O4@ZF

Unmodified ZF is a silicon-aluminum oxide with a small amount of Fe as the impurity (Britannica 2023). XPS analysis showed that the Fe contents on the surface of ZF, Fe2O3@ZF and Fe3O4@ZF were 0.64, 5.5 and 13.3%, respectively. Fe2+ and Fe3+ were detectable on the surface of Fe3O4@ZF, accounting for 29 and 71%, respectively (Figure 2(e)), while on the surface of ZF and Fe2O3@ZF Fe3+ was the primary component of Fe (Figure 2(d) and 2(f)).
Figure 2

High-resolution XPS spectra of (a) ZF, (b) Fe3O4@ZF, (c) Fe2O3@ZF, and (d,e, f) their Fe2p orbits.

Figure 2

High-resolution XPS spectra of (a) ZF, (b) Fe3O4@ZF, (c) Fe2O3@ZF, and (d,e, f) their Fe2p orbits.

Close modal
XRD analysis showed that there were no diffraction peaks of iron on the surface of ZF modified with Fe2O3 or Fe3O4, which implied that the deposited Fe was amorphous (Figure 3(d)). In the present study, the calcination temperature was restricted to avoid the collapse of ZF by overheating, so the deposited Fe may not form a stable crystalline structure.
Figure 3

The nitrogen adsorption–desorption curves of (a) ZF, (b) Fe3O4@ZF, (c) Fe2O3@ZF, and (d) XRD patterns of ZF, Fe3O4@ZF and Fe2O3@ZF.

Figure 3

The nitrogen adsorption–desorption curves of (a) ZF, (b) Fe3O4@ZF, (c) Fe2O3@ZF, and (d) XRD patterns of ZF, Fe3O4@ZF and Fe2O3@ZF.

Close modal

The nitrogen absorption and desorption curve of ZF loaded with iron metal was also obtained by BET analysis (Figure 3(a)–3(c)). The pore size of the Fe3O4@ZF and Fe2O3@ZF was the same as that of ZF, which is between 2 and 50 nm and belonged to mesoporous materials (Thommes 2010). The desorption curves of the three materials were similar, and they were all H3-type hysteresis ring isotherms (Chen et al. 2019; Gibson et al. 2020). Compared with ZF, the specific surface area of the Fe3O4@ZF increased from 16.22 to 22 m2/g, while the Fe2O3@ZF decreased to 14.77 m2/g. Furthermore, the pore volume of the Fe3O4@ZF increased from 0.0047 to 0.0063 cm3/g, while the Fe2O3@ZF remained basically unchanged (0.0041 cm3/g, Supplementary material, Table S1). The results showed that the modification with Fe3O4 increased the specific surface area and pore volume of ZF, both of which could improve the adsorption performance of Fe3O4@ZF and be beneficial to the growth of microorganisms.

The morphology of ZF, Fe2O3@ZF and Fe3O4@ZF was analyzed by SEM (Figure 4(a)–4(c)). The surface of ZF was rough with irregular particles, and there were obvious crystals around the pores. The surface of Fe3O4@ZF was uniformly covered with 10–50 nm spherical particles, which may come from the deposition of aggregated Fe3O4, and the pores were well preserved. This phenomenon implied that Fe3O4@ZF had a developed pore structure may provide a larger specific surface area, which was consistent with the results of BET analysis. The surface of Fe2O3@ZF was covered with 20–200 nm bulk particles, which showed long plate-like structures with a length of 10 μm, and the modification significantly blocked the pores, resulting in a reduction in specific surface area, as shown in Supplementary material, Table S1. After the modification, the color of the surface became brown as shown in Figure 4(d) and 4(e), which implied that Fe had been successfully loaded onto the ZF.
Figure 4

Morphology of ZF, Fe3O4@ZF and Fe2O3@ZF analyzed by SEM at the scales of (a–c) 1 μm and the photos of (d–f) ZF, Fe3O4@ZF and Fe2O3@ZF.

Figure 4

Morphology of ZF, Fe3O4@ZF and Fe2O3@ZF analyzed by SEM at the scales of (a–c) 1 μm and the photos of (d–f) ZF, Fe3O4@ZF and Fe2O3@ZF.

Close modal

Performance of BAF with difference fillers

The BAFs were started with the secondary sludge from a WWTP in Guangdong, China and fed with simulated sewage water. CIP was added into the inflow when the removals of COD and were all greater than 95% and maintained almost constant. The presence of CIP at low concentrations (2–10 mg/L) did not affect the treatment performances of BAFs, and there were no statistic differences among the removal efficiency of COD, NH3-N, TN and TP in the three reactors as shown in Supplementary material, Figure S1. The results indicated that the trace antibiotics in sewage could not significantly inhibit the degradation of conventional pollutants by microorganisms in BAFs.

However, the treatment efficiencies of CIP among the three reactors were different (Figure 5). At the very beginning, high removal efficiency of CIP (approximately 97%) were obtained in the three reactors, which may be attributed to the adsorption of CIP by the fillers and attached biofilms. Then, the removal efficiency decreased significantly in BAF-Fe2O3@ZF, while it only reduced by approximately 5% in BAF-ZF and BAF-Fe3O4@ZF. This phenomenon indicated that the microorganisms on ZF and Fe3O4@ZF had a much better adaptability to the stress of CIP than those on Fe2O3@ZF. When the concentration of CIP was 10 mg/L, the removal efficiency in BAF-Fe3O4@ZF was 79 ± 2.4%, approximately 15 and 30% higher than those in BAF-ZF and BAF-Fe2O3@ZF, respectively.
Figure 5

Removal efficiency of CIP by BAF-ZF, BAF-Fe3O4@ZF and BAF-Fe2O3@ZF during the domestication process.

Figure 5

Removal efficiency of CIP by BAF-ZF, BAF-Fe3O4@ZF and BAF-Fe2O3@ZF during the domestication process.

Close modal

Control of resistance genes

The contents of four typical ARG, including Int, mexA, qnrB and qnrS, both in the effluent and on the surface of fillers were investigated (Figure 6). There were always the lowest ARGs in the effluent from BAF-Fe3O4@ZF as compared with the other two reactors, and the Mann–Whitney U test showed a significant gap among the three reactors (p < 0.05). For example, the content of gene Int in the effluent of BAF-Fe3O4@ZF at the CIP concentration of 10 mg/L was 2.38 × 1010 gene copies/mL, less than those of the other two reactors (Figure 6(a)). However, the case on the surface of fillers were different. BAF-Fe3O4@ZF has the lowest Int and mexA but the highest qnrB and qnrS on the surface of fillers as shown in Figure 6(c) and 6(d).
Figure 6

The contents of gene Int, mexA, qnrB and qnrS in (a and c) BAF effluent and (b and d) on the surface of fillers at the CIP concentrations of 10 mg/L in the influent.

Figure 6

The contents of gene Int, mexA, qnrB and qnrS in (a and c) BAF effluent and (b and d) on the surface of fillers at the CIP concentrations of 10 mg/L in the influent.

Close modal

Variations of microbial community and activity in BAF with different fillers

The compositions of the microbial community in the three reactors were significantly different after the domestication process (Figure 7(a)). Fe3O4 modification increased the proportion of Actinobacteriota (8.4–13.4%), Acidobacteriota (1.4–3.3%) and Planctomycetota (0.3–4.6%) and decreased the proportion of Proteobacteria (70.6–60.9%). These increased microbial communities may promote the degradation of CIP in BAF-Fe3O4@ZF and enhance the resistance of microorganisms to the toxic effect of CIP. As a comparison, Fe2O3 modification increased the proportion of Chloroflexi (4.4–15.8%), Verrucomicrobiota (1.8–10.6%) and Planctomycetota (0.3–1.8%), but the potential CIP-degrading bacteria, Actinobacteriota (8.4–8.3%) and Acidobacteriota, even decreased. These differences may result in a poor removal efficiency of CIP in BAF-Fe2O3@ZF.
Figure 7

The relative abundance of bacteria at the (a) phylum level and (b) genus level in the three reactors and (c) ATP content on the surface of fillers at the CIP concentration of 4,6,8 and 10 mg/L in the influent.

Figure 7

The relative abundance of bacteria at the (a) phylum level and (b) genus level in the three reactors and (c) ATP content on the surface of fillers at the CIP concentration of 4,6,8 and 10 mg/L in the influent.

Close modal

The relative abundance of bacteria at genus level in three reactors was shown in Figure 7(b). The proportion of Rhizobia in BAF-Fe3O4@ZF was higher than that in the other two reactors. The top five relative abundant genera in BAF-Fe3O4@ZF are Acidovorax (47.48%), Galbitalea (8.33%), AAP99 (8.25%), Xanthomonadaceae (3.92%) and Caldilineaceae (3.57%). The relative abundance of two bacteria, Deinococcus-Thermus and Rhizobia, with the potential to degrade CIP (Nguyen et al. 2018; Pan et al. 2018), in BAF-Fe3O4@ZF was much higher than that in the other two reactors. Both Rhizobia and Deinococcus are capable of degrading CIP, which may be one of the reasons that BAF-Fe3O4@ZF showed a better CIP degradation efficiency than the other two reactors.

The surface morphology of the fillers in the three reactors after the domestication process was analyzed (Supplementary material, Figure S6), and the results showed that the proportion of bacillus and coccus attached to the surface of the three materials was different, which may be related to the varied relative abundance of microorganisms. Apparently, the proportion of cocci on the surface of Fe3O4@ZF was more abundant than that on the surface of ZF and Fe2O3@ZF. This great difference was in accordance with the results of microbial community analysis, and it also implied that the compositions of the microbial community would have been changed because of the iron modification.

ATP analysis (Figure 7(c)) showed that the content of ATP on the filler of BAF-Fe3O4@ZF was greatly enhanced and the value was more than twice that on the fillers of the other two reactors when the concentration of CIP was no less than 6 mg/L. Meanwhile, there was no statistic difference (p > 0.05) of the biomass on the surface of fillers among the three reactors according to the results of 16S rRNA analysis (Supplementary material, Figure S7). These phenomena indicated that Fe3O4 modification enhanced the microbial metabolism activity but did not increase the biomass.

Generally, the Fe3O4 modification greatly improve the removal efficiency of CIP. In the present study, the removal efficiency increased to 79% at a CIP loading of 10 mg/L. Previous studies also found similar results (Yang et al. 2018; Lin & Lee 2020).

Fe3O4 modification also reduced the released ARGs in the effluent, which should be beneficial to reducing the risk of HGT. In the present study, we selected four typical ARGs to investigate the management of ARGs by BAF. Int, the most studied integraton, is regard as the key to the HGT of ARGs (Partridge et al. 2009). MexA is an important efflux pump gene of CIP, which may help bacteria excrete the antibacterial drugs entering the cells through the efflux pump system, resulting in a decrease in the concentration of drugs in cells to produce drug resistance (Tandukar et al. 2013). Both qnrB and qnrS are specific fluoroquinolone-resistant genes, which may represent the resistance of bacterial to toxic effect of CIP (Rutgersson et al. 2014). Fe3O4 modification reduced the contents of all the four ARGs by 7.16–60.22% in the effluent as compared with that in BAF@ZF. Obviously, the reduced ARGs in water would be beneficial to limiting the HGT and reducing the appearance of antibiotic-resistant pathogenic bacteria.

However, the variations of ARGs on the surface of fillers showed a contrary tendency. The contents of Int and mexA decreased on the surface of Fe3O4@ZF as compared with those on ZF and Fe2O3@ZF, which implied that the presence of CIP significantly inhibited the activity of microflora on the surface of fillers. Nevertheless, some specific CIP-degrading bacteria, such as Rhizobium and Deinococcus-Thermus, gained superiority with the help of Fe3O4 and became dominant bacteria in the system, resulting increased contents of qnrB and qnrS (Figure 6(b) and 6(d)).

The enhance treatment performance of CIP by BAF-Fe3O4@ZF could be attributed to the improved physical properties after the modification and the introduction of multi-valence iron. Fe3O4 modification increased the specific surface area by 35.6% and the average also increased by 34.0% according to the analytical results of BET and SEM (Supplementary material, Table S1), which effectively enhanced adsorption performance (Danalioglu et al. 2017). These features provided more attachment sites for microorganisms, allowing for the formation of larger biofilms and improved contact with CIP and organic substrates (Lin & Lee 2020).

It should be noted that the valence and structure of iron rather than the loading amount should be critical factors to the successful modification. Fe2O3 modification even worsened the performance of BAF and cannot promote reproduction of CIP-degrading bacteria, though the loading amount of iron was more than twice that of Fe3O4 modification (Figure 2). The electron transfer between Fe(II) and Fe(III) may contribute to the degradation of CIP, which has been extensive reported in the studies of anaerobic biodegradation (Bao & Li 2017). In BAFs, an anaerobic environment may occur at the depth of biofilm, and the introduction of multi-valence iron could promote the direct interspecies electron transfer and improve the degradation of CIP (Xu et al. 2024).

The introduction of multi-valence iron changed the microbial community and facilitated the growth of CIP-degrading bacteria, which was supported by the increased abundance of Rhizobium and Deinococcus-Thermus (Figure 7(b)). Meanwhile, it also greatly enhanced the microbial metabolism activity, which was supported by the higher ATP concentration on the surface of Fe3O4@ZF (Figure 7(c)). The concentration of ATP could be regarded as a reliable and fast indicator of microbial metabolism activity, and a high concentration may promote the degradation of xenobiotic pollutants (Yan et al. 2019). The fact that Fe3O4 modification improved the microbial activity may be another reason for the increased removal efficiency of CIP.

It is interesting that the biomass on the surface of fillers did not change significantly, and there were no statistic difference (p > 0.05) of the biomass on the surface of fillers among the three reactors (Supplementary material, Figure S7). This phenomenon was in accordance with the facts that the removal efficiencies of COD, NH3-N, TN, and TP were identical among the three reactors (Supplementary material, Figure S1). These phenomena implied that the facilitation of Fe3O4 on the growth of microorganism may be selective. Only some specific bacteria could obtain the benefit from Fe3O4 modification, and the CIP-degrading bacteria happened to belong these ‘lucky microbes’. However, the underlying mechanism is still unclear and further study is needed.

The complex composition of pharmaceutical wastewater, rich in suspended solids (SS), poses challenges for reactors. These SS can clog filter pores, compete for adsorption sites, and disrupt biological processes, affecting reactor performance (Wu et al. 2018). To address these issues, comprehensive studies with real wastewater are crucial for determining optimal BAF filling ratios and pre-treatment SS requirements, aiming to reduce maintenance and enhance removal efficiency.

In the present study, we developed a simple method for the modification of ZF by Fe3O4. The removal of CIP by BAF filled with Fe3O4@ZF could be significantly improved and its removal efficiency was approximately 15% higher than BAF filled with unmodified ZF at the influent CIP concentration of 10 mg/L. In the effluent of Fe3O4@ZF amended BAF reactor, the expression of resistance genes Int, mexA, qnrB and qnrS was the least and the relative abundance of qnrB and qnrS on the surface of filter was the most, which indicated that the HGT of ARGs can be prevented. Fe3O4 modification could enrich the relative abundance of some CIP-degrading bacteria, such as Rhizobium and Deinococcus-Thermus, in the reactor and increase the content of ATP, which may be the reasons that enhancing the treatment performance of CIP.

This work is supported by the National Natural Science Foundation of China (No. 52270002), Guangdong Basic and Applied Basic Research Foundation (2021A1515010508), Science and Technology Project of Guangzhou (202102010448), and Hundred Talent Program of Guangzhou University (RQ2020050).

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

The authors declare there is no conflict.

Bao
P.
&
Li
G. X.
2017
Sulfur-driven iron reduction coupled to anaerobic ammonium oxidation
.
Environmental Science & Technology
51
(
12
),
6691
6698
.
Bao
T.
,
Chen
T. H.
,
Ezzatahmadi
N.
,
Rathnayake
S. I.
,
Chen
D.
,
Wille
M. L.
&
Frost
R.
2017
A performance evaluation of a new iron oxide-based porous ceramsite (IPC) in biological aerated filters
.
Environmental Technology
38
(
7
),
827
834
.
Britannica
T.
Editors of Encyclopaedia
2023
zeolite. Encyclopedia Britannica
.
Byrne
J. M.
,
Klueglein
N.
,
Pearce
C.
,
Rosso
K. M.
,
Appel
E.
&
Kappler
A.
2015
Redox cycling of Fe(II) and Fe(III) in magnetite by Fe-metabolizing bacteria
.
Science
347
(
6229
),
1473
1476
.
Cacace
D.
,
Fatta-Kassinos
D.
,
Manaia
C. M.
,
Cytryn
E.
,
Kreuzinger
N.
,
Rizzo
L.
,
Karaolia
P.
,
Schwartz
T.
,
Alexander
J.
,
Merlin
C.
,
Garelick
H.
,
Schmitt
H.
,
de Vries
D.
,
Schwermer
C. U.
,
Meric
S.
,
Ozkal
C. B.
,
Pons
M. N.
,
Kneis
D.
&
Berendonk
T. U.
2019
Antibiotic resistance genes in treated wastewater and in the receiving water bodies: A pan-European survey of urban settings
.
Water Research
162
,
320
330
.
Chen
L.
,
Zuo
L.
,
Jiang
Z. X.
,
Jiang
S.
,
Liu
K. Y.
,
Tan
J. Q.
&
Zhang
L. C.
2019
Mechanisms of shale gas adsorption: Evidence from thermodynamics and kinetics study of methane adsorption on shale
.
Chemical Engineering Journal
361
,
559
570
.
Danalioglu
S. T.
,
Bayazit
S. S.
,
Kerkez
Ö.
,
Alhogbi
B. G.
&
Salam
M. A.
2017
Removal of ciprofloxacin from aqueous solution using humic acid- and levulinic acid- coated FeO nanoparticles
.
Chemical Engineering Research & Design
123
,
259
267
.
Gibson
N.
,
Kuchenbecker
P.
,
Rasmussen
K.
,
Hodoroaba
V.-D.
,
Rauscher
H.
,
2020
Chapter 4.1 - Volume-specific surface area by gas adsorption analysis with the BET method
. In:
Hodoroaba
V.-D.
,
Unger
W. E. S.
&
Shard
A. G.
(eds). Characterization of Nanoparticles,
Elsevier
, Amsterdam, The Netherlands, pp.
265
294
.
Kato
S.
,
Hashimoto
K.
&
Watanabe
K.
2012
Microbial interspecies electron transfer via electric currents through conductive minerals
.
Proceedings of the National Academy of Sciences of the United States of America
109
(
25
),
10042
10046
.
Khan
F. A.
,
Söderquist
B.
&
Jass
J.
2019
Prevalence and diversity of antibiotic resistance genes in Swedish aquatic environments impacted by household and hospital wastewater
.
Frontiers in Microbiology
10
, 441998.
Marsidi
N.
,
Abu Hasan
H.
&
Abdullah
S. R. S.
2018
A review of biological aerated filters for iron and manganese ions removal in water treatment
.
Journal of Water Process Engineering
23
,
1
12
.
Pan
L. J.
,
Li
J.
,
Li
C. X.
,
Tang
X. D.
,
Yu
G. W.
&
Wang
Y.
2018
Study of ciprofloxacin biodegradation by a sp isolated from pharmaceutical sludge
.
Journal of Hazardous Materials
343
,
59
67
.
Partridge
S. R.
,
Tsafnat
G.
,
Coiera
E.
&
Iredell
J. R.
2009
Gene cassettes and cassette arrays in mobile resistance integrons
.
Fems Microbiology Reviews
33
(
4
),
757
784
.
Rutgersson
C.
,
Fick
J.
,
Marathe
N.
,
Kristiansson
E.
,
Janzon
A.
,
Angelin
M.
,
Johansson
A.
,
Shouche
Y.
,
Flach
C. F.
&
Larsson
D. G. J.
2014
Fluoroquinolones and genes in sediment, water, soil, and human fecal flora in an environment polluted by manufacturing discharges
.
Environmental Science & Technology
48
(
14
),
7825
7832
.
Tandukar
M.
,
Oh
S.
,
Tezel
U.
,
Konstantinidis
K. T.
&
Pavlostathis
S. G.
2013
Long-term exposure to benzalkonium chloride disinfectants results in change of microbial community structure and increased antimicrobial resistance
.
Environmental Science & Technology
47
(
17
),
9730
9738
.
Thommes
M.
2010
Physical adsorption characterization of nanoporous materials
.
Chemie Ingenieur Technik
82
(
7
),
1059
1073
.
Wang
K.
,
Zhuang
T.
,
Su
Z. X.
,
Chi
M. H.
&
Wang
H. C.
2021
Antibiotic residues in wastewaters from sewage treatment plants and pharmaceutical industries: Occurrence, removal and environmental impacts
.
Science of the Total Environment
788
, 147811.
Wu
C. Y.
,
Zhou
Y. X.
,
Sun
X. M.
&
Fu
L. Y.
2018
The recent development of advanced wastewater treatment by ozone and biological aerated filter
.
Environmental Science and Pollution Research
25
(
9
),
8315
8329
.
Xu
H.
,
Zhang
L.
,
Xu
R. H.
,
Yang
B.
&
Zhou
Y.
2024
Iron cycle-enhanced anaerobic ammonium oxidation in microaerobic granular sludge
.
Water Research
250
, 121022.
Yan
W. W.
,
Lu
D.
,
Liu
J. B.
&
Zhou
Y.
2019
The interactive effects of ammonia and carbon nanotube on anaerobic digestion
.
Chemical Engineering Journal
372
,
332
340
.
Yang
K. L.
,
Yue
Q. Y.
,
Kong
J. J.
,
Zhao
P.
,
Gao
Y.
,
Fu
K. F.
&
Gao
B. Y.
2016
Microbial diversity in combined UAF-UBAF system with novel sludge and coal cinder ceramic fillers for tetracycline wastewater treatment
.
Chemical Engineering Journal
285
,
319
330
.
Yang
Z. M.
,
Shi
X. S.
,
Dai
M.
,
Wang
L.
,
Xu
X. H.
&
Guo
R. B.
2018
Co-metabolic removal of ciprofloxacin under condition of interaction between microbes and FeO
.
Chemical Engineering Journal
333
,
649
656
.
Yao
S. J.
,
Ye
J. F.
,
Yang
Q.
,
Hu
Y. R.
,
Zhang
T. Y.
,
Jiang
L.
,
Munezero
S.
,
Lin
K. F.
&
Cui
C. Z.
2021
Occurrence and removal of antibiotics, antibiotic resistance genes, and bacterial communities in hospital wastewater
.
Environmental Science and Pollution Research
28
(
40
),
57321
57333
.
Yi
K. X.
,
Wang
D. B.
,
Yang
Q.
,
Li
X. M.
,
Chen
H. B.
,
Sun
J.
,
An
H. X.
,
Wang
L. Q.
,
Deng
Y. C.
,
Liu
J.
&
Zeng
G. M.
2017
Effect of ciprofloxacin on biological nitrogen and phosphorus removal from wastewater
.
Science of the Total Environment
605
,
368
375
.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY 4.0), which permits copying, adaptation and redistribution, provided the original work is properly cited (http://creativecommons.org/licenses/by/4.0/).

Supplementary data