Journal of Environmental Management 377 (2025) 124657

Contents lists available at ScienceDirect

Journal of Environmental Management

journal homepage: www.elsevier.com/locate/jenvman

Research article

Enhanced pharmaceutical removal from building wastewater by the novel

integrated system of anaerobic baffled biofilm-membrane bioreactor and
UV/O3: Microbial community, occurrence of bio-intermediates and
post-treatment
Tanissorn Buakaew a,b, Chavalit Ratanatamskul a,b,*
a
Department of Environmental and Sustainable Engineering, Faculty of Engineering, Chulalongkorn University, Bangkok, 10330, Thailand
b
Center of Excellence in Innovative Waste Treatment and Water Reuse, Faculty of Engineering, Chulalongkorn University, Bangkok, 10330, Thailand

A R T I C L E I N F O A B S T R A C T

Keywords: This research aimed to develop the novel integrated system of anaerobic baffled biofilm-membrane bioreactor
Anaerobic baffled biofilm membrane (AnBB-MBR) (with and without microaeration) and UV/O3 for removal of target pharmaceuticals (ciprofloxacin
bioreactor (AnBB-MBR) (CIP), caffeine (CAF), sulfamethoxazole (SMX) and diclofenac (DCF)) from building wastewater. The investi­
Bio-intermediate products
gation was performed to elucidate how microaeration affected the removal performances, degradation kinetics
Integrated system
and pathways of bio-intermediates of the AnBB-MBR. Two AnBB-MBR reactors - R1: AnBB-MBR (without
Microaeration
Pharmaceuticals removal microaeration) and R2: AnBB-MBR with microaeration at 0.93 LO2/LFeed - were operated at the same hydraulic
UV/O3 retention time (HRT) of 30 h. The UV/O3 was selected as the post-treatment system. While UV alone slightly
removed CIP without the removal of other compounds. After 150 min of the UV/O3, the R1 with UV/O3 achieved
97.31–100% removal efficiency of targeted pharmaceuticals and increased to 99.47–100% with the R2 inte­
grated with UV/O3. The obtained pseudo-first order kinetic rate constants of the UV/O3 in treating the permeate
of R1 were 0.0235, 0.004, 0.0423 and 0.097 min− 1 for CIP, CAF, SMX and DCF, respectively. Whereas the
obtained pseudo-first order kinetic rate constants of the UV/O3 in treating the permeate of R2 were 0.021,
0.0338, 0.0511 and 0.0527 min− 1 for CIP, CAF, SMX and DCF, respectively. For the major microorganisms
involved in targeted pharmaceutical removal in the R2 under microaerobic conditions included ammonia
oxidizing bacteria (AOB) and methanotrophs, while Bacillus, Longilinea, Clostridium and Lactivibrio were possibly
responsible for pharmaceutical removal in the R1 under anaerobic conditions. The differences of bio-
intermediates between anaerobic and microaerobic conditions were exclusively identified. In addition, the
integration of AnBB-MBR with microaeration and UV/O3 was more effective in removing a wide variety of bio-
intermediates than the case of the integrated system without microaeration. Therefore, the integrated system of
AnBB-MBR with microaeration and UV/O3 can be a promising technology for pharmaceutical removal from
building wastewater.

1. Introduction into treatment plants where these substances have been identified at
various concentrations from ng/L to μg/L (Carneiro et al., 2019; Frederic
The contamination of pharmaceuticals in environment has been the and Yves, 2014). Among these pharmaceuticals, antibiotics, stimulants
subject of comprehensive research owing to their frequent identification and analgesics have been indicated as the most used pharmaceuticals
in wastewater and water sources and the potential risks they pose to around the world and they have been globally detected in the river and
human health through agricultural products and water consumption canal (Tewari et al., 2013), water basin (Zhao et al., 2024a) and the
(Ben Mordechay et al., 2022; Siri et al., 2024; Tamura et al., 2017). effluent discharges from municipal and hospital-wastewater treatment
Hospitals, households, industries and agricultural facilities and livestock (Sinthuchai et al., 2021).
constitute the main origins of pharmaceuticals, releasing their waste For antibiotics, CIP and SMX are widely consumed by humans and

* Corresponding author. Department of Environmental and Sustainable Engineering, Faculty of Engineering, Chulalongkorn University, Bangkok, 10330, Thailand.
E-mail address: [email protected] (C. Ratanatamskul).

https://doi.org/10.1016/j.jenvman.2025.124657
Received 19 October 2024; Received in revised form 13 February 2025; Accepted 18 February 2025
Available online 25 February 2025
0301-4797/© 2025 Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies.
T. Buakaew and C. Ratanatamskul Journal of Environmental Management 377 (2025) 124657

animals (Chatila et al., 2016). CIP is a significant fluoroquinolone of compartments and following by MBR. It could operate with constant
antibiotic, which is characterized by a high adsorption capacity to flux for long experimental runs with a low membrane fouling rate, and
biomass (Nguyen et al., 2019). It can also exert microbial lethality, the avoiding sludge washout by its configurations of baffles, fixed films,
development of bacterial and gene resistances, mutagenic risks, and moving beds and membrane (Buakaew and Ratanatamskul, 2021;
potential carcinogenicity (Girardi et al., 2011). SMX is a sulfonamide Ratanatamskul and Katasomboon, 2017). The use of AnBB-MBR inte­
antibiotic with variable biodegradability by microorganisms. (Xu et al., grated with ozoned-based AOP process as the post-treatment might offer
2022), and it has been identified by WHO as a carcinogenic substance the advantage of eliminating suspended solids by the AnBB-MBR, which
(Zhang et al., 2023a). CAF is a legal stimulant that is easily biodegrad­ would otherwise interfere with the oxidation process. Additionally,
able through biological processes, but demonstrates persistence when it AnBB-MBR might offer high removal of pollutant loads, allowing
was exposed to oxidative reagents (Chen et al., 2018). The demand of ozoned-based AOP process more efficiently. Although numerous studies
CAF consumption has been increasing every year (Raj et al., 2021). It has have utilized ozone-based processes to remove residual pharmaceuticals
been reported that CAF at concentrations in the range of hundreds of from biological treatment systems (Gorito et al., 2021; Kim and Tanaka,
μg/L can cause acute toxicity to Daphnia magna (Pashaei et al., 2023). 2010; Paucar et al., 2019; Wang et al., 2019), the direct comparison of
DCF is an analgesic known for being a recalcitrant compound in the integrating ozone-based processes with anaerobic and microaerobic
environment (Wu et al., 2020), and it poses harmful cytological effects biological processes in treating the residual pharmaceuticals still re­
in liver, kidney, gills and intestine of rainbow trout even at low range mains unclear in terms of degradation kinetics and performance,
concentration (μg/L) (Triebskorn et al., 2004). removal mechanism and pathways, and bio-intermediate product
In recent years, anaerobic treatment methods have gained wide­ (Paucar et al., 2019; Wang et al., 2019). This necessity arises from the
spread adoption for managing wastewater because of their cost- divergent nature of organics and nutrients in various types of
effectiveness and high efficiency (Meng et al., 2009). Anaerobic mem­ wastewaters.
brane bioreactor (AnMBR) is an integration of anaerobic digestion (AD) This work aimed to develop a novel integrated system of AnBB-MBR
with membrane separation, which can keep high biomass in the system (with and without microaeration) and UV/O3 to enhance the removal of
with a prolonged sludge retention time (SRT). Moreover, this process is targeted pharmaceuticals (CIP, CAF, SMX and DCF), which are
challenged by lower energy consumption, efficient organic matter commonly detected in building wastewater. In this study, UV and O3
degradation, and less biosolid production (Ji et al., 2020). The quality of alone, and UV/O3 photoreactor were initially investigated for the
the effluent from AnMBR is better than that from the conventional AD removal of targeted pharmaceuticals in order to further select a post-
process (Aslam et al., 2022). The AnMBR system demonstrated signifi­ treatment of the AnBB-MBR system, and the UV/O3 was selected. This
cantly higher removal efficiency for CIP and SMX compared to the study also investigated the UV/O3 degradation kinetics and removal
conventional AD system. Moreover, the AnMBR system consistently pathways for treatment of the permeate from the AnBB-MBR system. To
achieved a COD removal rate of approximately 84%, whereas the COD reveal the difference of possible transformation pathways of the targeted
removal efficiency in the AD system remains below 70% (Ji et al., 2020). pharmaceutical removal by the AnBB-MBR under anaerobic and
However, the limitation of the anaerobic condition is low removal of microaerobic conditions, the intermediate products were identified.
nutrients and complex molecular structures due to the exclusive thriving Moreover, the reduction in the intermediate products by the UV/O3
only anaerobic microorganisms under oxygen-deficient conditions system as the post-treatment was intensively investigated.
(Ratanatamskul and Katasomboon, 2017).
Low amounts of oxygen within the range of 1 LO2/LFeed called 2. Materials and methods
microaeration have been employed in AD for boosting the activity and
diversity of microbial communities, fostering synergistic interactions 2.1. Set-up the integrated system of AnBB-MBR and UV/O3 photoreactor
among various microbial groups, contributing to the establishment of a system
more robust and stable AD process (Jenicek et al., 2014; Nguyen and
Khanal, 2018). Microaeration applied in the AD process at 0.10–0.15 The two novel integrated treatment systems are illustrated in Fig. 1.
LO2/LFeed has been reported to enhance the biodegradation of recalci­ The two identical AnBB-MBR systems were made of plexiglass with total
trant compounds including SMX and DCF up to 90% as compared to the volume of 40.5 L. Each system consists of three compartments. The
sole AD process (20%). The application of microaeration also can enrich volume of the first compartment (C1), second compartment (C2), and
microaerophilic genera and promote specific oxygenase enzymes, which third compartment (C3) were 11.25 L, 11.25 L, and 18 L, respectively. In
involve in the biotransformation of pharmaceuticals (do Nascimento C1 and C2 compartments, fiber glass filters (Glass filter mat, AQUA-
et al., 2021). It could alleviate membrane fouling rate 14% by lowering NEAT, Japan) with a specific area of 500 m2 m− 3 and the working
soluble protein substances (Buakaew and Ratanatamskul, 2023). volume of 4.5 L were installed. Moving bed (MVB) media, Mutag Bioship
However, the non-complete elimination of pharmaceuticals is the 25 TM PE with a circular length of 2.5 cm, thickness of 0.11 cm, and
main shortcoming of biological processes since their resistant molecular surface area of 4850 m2m− 3 (Multi Umwelttechnologies AG, Germany)
structures impede their biological breakdown (Gu et al., 2018; Tiwari were introduced into the C3 compartment at a volume ratio of 30%
et al., 2017). Advanced oxidation processes (AOPs) have been recog­ (Vmedia: Vcompartment) to encourage the growth of biofilm. A sonic
nized to be able to dismantle the molecular structures of persistent pump (AP1600) was utilized to circulate the MVBs in the C3 zone for
pollutants and reduce toxicity in the discharged effluent (Das et al., membrane scouring and mixing. A ceramic flat sheet microfiltration
2024; Gorito et al., 2021). Ozone-based processes were characterized by membrane (Ceraflo, Singapore) with a pore size of 0.5 μm and an
its potent oxidation capability, and environmentally friendly by gener­ average surface area of 0.107 m2 was immersed in the C3 solution within
ating hydroxyl free radicals (OH⋅), superoxide free radicals (O.-2) and each bioreactor. This membrane was linked to a vacuum pressure gauge
singlet oxygen (1O2) and direct O3 with dissolved pollutants to degrade to monitor transmembrane pressure (TMP). Metering pumps (Promi­
them to the products that are more biodegradable in aquatic ecosystem. nent, Germany) were utilized for the feeding and permeate processes.
(Poyatos et al., 2009). Although ozone-based processes are effective in Level sensors were employed to regulate the wastewater levels within
removal of pollutants, some disadvantages such as high costs, energy each of the two bioreactors. From Fig. 1, reactor 1 (R1) is the anaerobic
demands, toxic by-product formation, and sensitivity to water matrix baffled biofilm-Membrane Bioreactor (AnBB-MBR) and reactor 2 (R2) is
components such as suspended solids, organics and specific ions makes the AnBB-MBR with microaeration of 0.1 L min− 1 (DO in the range of
them less scalable and complex to maintain (Wardenier et al., 2019). 0.3–0.7 mg/L) in the C3 compartment.
The development of an anaerobic baffled biofilm-membrane biore­ To conduct experiments with UV, O3 and UV/O3 as the integrated
actor (AnBB-MBR) involved separating biomass by baffles in a sequence post treatment, the O3 generator with capacity of 400 mgO3/h with gas

2
T. Buakaew and C. Ratanatamskul Journal of Environmental Management 377 (2025) 124657

Fig. 1. Integrated systems of R1: AnBB-MBR with UV/O3 photoreactor (a), and R2: AnBB-MBR with microaeration with UV/O3 photoreactor (b).

flowrate of 1 L/min (Hi-400i, Glory Ozoner) and the UVC lamp removal performance by the developed novel integrated system.
(Lamptan) with the wavelength of 254 nm and UV flux of 0.03 mW/cm2 Anaerobic sludges sourced from a textile wastewater treatment plant
were used. The reactor was conducted in a 2 L-batch cylindrical glass were introduced into the C1 and C2 compartments of two bioreactors,
reactor with controlling the temperature by a water bath at 35 C◦ in a occupying 30% of the reactor volume. These sludges underwent a
dark box. The residual O3 gas was eliminated by immersion in a 20% KI continuous acclimation process for over three months, during which
solution before being released into the external environment. In the case they were exposed to high-rise building wastewater containing targeted
of testing UV alone, nitrogen gas was employed as a substitute for O3 at a pharmaceutical compounds. This acclimation period preceded the
flow rate of 1 L/min to control the turbulence in the system, and this was commencement of the experimental phase. Two AnBB-MBR systems (R1
then compared with O3 and UV/O3 systems. and R2) were operated simultaneously under experimental conditions
with HRT of 30 h for 115 d, corresponding to a permeate flux of 12.6 L/
2.2. Wastewater and experimental investigations m2/h, and SRT of 100 days was controlled in C3 compartment. The
permeate flow rates were intermittently managed with a cycle of 8 min
This study utilized high-rise building wastewater from toilets and the of suction followed by 2 min of relaxation to avoid excessive stress on
canteen within Mahitaladhibesra Building at Chulalongkorn University. the membrane. When TMP reached a range of 50–53 kPa, the membrane
The wastewater was pumped from a septic tank of the building into 700 was removed for cleaning by scrubbing to remove fouling substances,
L of influent tank. Wastewater samples were collected from each sam­ then immersed in a 0.4% sodium hypochlorite (NaOCl) solution for 8 h.
pling point using pre-cleaned plastic bottles. Bottles were rinsed several Following this, the membranes were washed with tap water and soaked
times with deionized (DI) water and then with the wastewater sample in 0.4% nitric acid for an additional 8 h. The membranes were rinsed
prior to collection. To protect wastewater characteristics, samples were again with tap water before being reused in the AnBB-MBR systems.
immediately placed in Styrofoam containers to minimize photo­ In order to select the post treatment method among UV photoreactor,
degradation and immediately transported to the laboratory for analysis. ozonation (O3) and UV/O3 photoreactor (UV/O3), the targeted phar­
Samples were refrigerated at 4 ◦ C to prevent biodegradation and maceuticals were spiked in DI water at 200 μg/L each to comparatively
analyzed within 4 h of collection. The targeted pharmaceuticals (CIP, observe the best method which provided the shortest contact time with
CAF, SMX and DCF) supplied by Sigma-Aldrich, were individually fastest kinetic degradation and highest removal efficiency. After that,
formulated into stock solutions at a concentration of 2 g/L. These so­ the best method was integrated with the AnBB-MBRs in both R1 and R2
lutions were stored at − 20 ◦ C before being introduced as spikes. CIP was systems, aiming to compare the removal performances of organics, nu­
dissolved in a solution containing 0.1% formic acid. CAF was prepared trients, color and targeted pharmaceuticals between anaerobic and
using Milli-Q water, and SMX and DCF were dissolved in pure methanol. microaerobic conditions. The pharmaceutical concentrations, dissolved
All targeted pharmaceuticals were spiked to reach the same concentra­ O3 concentration, chemical oxygen demand (COD), nitrogen in various
tions of 250 μg/L, which were higher than the detected concentrations in forms, phosphorous and color were analyzed during the treatment time
the influent of municipal wastewater treatment plants (WWTPs) (Tran to observe kinetic degradations of targeted pharmaceuticals and O3
et al., 2018) in order to investigate the feasibility of the pharmaceutical profiles. For quantification of the samples from ozone-based treatments,

3
T. Buakaew and C. Ratanatamskul Journal of Environmental Management 377 (2025) 124657

nitrogen gas was used for removing the residual O3 in wastewater (Patel USA) equipped with Orbitrap high resolution, and a heated electrospray
et al., 2021). ionization source (H-ESI II). 0.1% formic acid in Milli-Q water (solvent
In this research, the pseudo-first order kinetic model was applied to A) and 0.1% formic acid in methanol (solvent B) were mobile phases.
describe the removal of pharmaceuticals using ozone-based treatment, The gradient elution program with a flowrate of 0.2 mL/min started
as shown in Eq. (1): with 20% solvent B for 1 min and increased linearly to 90% in 4 min, and
held at 90% for 2 min, after that, reduced to 20% for 3 min for equili­
dC Ct
r= − = ln = − kt (1) brate. A Thermo Hypersil GOLDTM C18 column (150 mm × 2.1 mm, 5
dt C0
μm particle size, 175 Å pore size) was applied at constant temperature at
Where r is the pharmaceutical degradation rate, C0 is the initial 40 ◦ C. The injection volume 5 μL was applied. The polarity was operated
concentration of pharmaceutical (μg/L), Ct is the concentration of in positive mode of ESI for MS parameters. HESI source parameters
pharmaceutical at time t (μg/L) and k is the pseudo-first order kinetic consist of Auxiliary gas heater temperature 300 ◦ C, capillary tempera­
rate constant (min− 1). ture 250 ◦ C, S-len RF level 50 and spray voltage 2.8 kV. Ultrapure ni­
trogen was used both as the source gas for ions and as the gas used for
2.3. Analytical procedures collisions with utilized in the following proportions: 40 au for sheath
gas, 10 au for auxiliary gas, and 2 au for sweep gas. The Orbitrap was
COD, Total Kjeldahl nitrogen (TKN), ammonia nitrogen (NH+ 4 -N), operated in Parallel reaction monitoring (PRM) mode with 35,000 res­
nitrite (NO−2 -N), nitrate (NO−3 -N), phosphate (PO3-
4 -P), color, mixed li­ olution targeted-MS2. The mass range in PRM was scanned at
quor suspended solids (MLSS), and mixed liquor volatile suspended 50.0–280.0 m/z. Maximum injection time was automated, and the
solids (MLVSS) were assessed using established standard methods automatic gain control (AGC) target in PRM was established at a value of
(APHA, 2017). Oxidation reduction potential (ORP) and pH were 5 × 104 ions. Normalized collision energy (NCE) was stepped at 20 and
measured using a pH/ORP meter (Hanna instrument, model no. HI 30 V for PRM, while an isolation window was set at 3.0 m/z. The
98121). DO concentration was measured by a DO meter (DO-5512SD, acquisition state for instrument control was operated in Xcalibur, and
Lutron). The dissolved O3 concentration in water solution was analyzed TraceFinder was used for processing the data. The SPE recovery was
by indigo disulfonate method with a spectrophotometry using a wave­ validated by the difference of the peak areas of the target compounds in
length of 600 nm (Zhang et al., 2023a). Fluorescence the mixed standards before SPE (solution A) and after SPE (solutions B):
excitation-emission matrix (FEEM) was conducted to determine the R (%) = (A/B) × 100. The SPE recovery of SMX was 80.55%, while the
organic substances that undergo reactions upon exposure to fluorescent other compounds were 100%. Limit of detection (LOD) and limit of
light in the permeates of AnBB-MBR systems and the effluents of UV/O3. quantification (LOQ) were calculated as the standard concentrations in
The samples were filtered with a 0.45 μm filter; and then using a fluo­ line with to a signal to noise ratio (S/N) of 3:1 and 10:1, respectively.
rescence spectrofluorometer (Aqualog®, HORIBA Scientific., USA) at The standard calibration curves of each compound are provided at 0.5,
excitation wavelengths of 220–500 nm with a 3 nm increment and 1, 5, 10, 25, 50, 100, 200, 500 and 1000 μg/L with high level of accuracy
emission wavelengths from 250 to 550 nm. The integration time was 3 (R2 > 0.99). The method of LC-MS/MS Orbitrap analysis was described
min, and a specific single-channel detector (SCD1) was used as a de­ in the earlier study (Buakaew and Ratanatamskul, 2024).
tector with a grating density of 1200 and 285 for excitation and emission The analysis of bio-intermediate products from the permeates of R1
wavelengths, respectively. Dissolved methane in the permeate of each and R2 and the intermediate products of targeted pharmaceuticals after
bioreactor was also analyzed using a headspace method as described in the UV/O3 process could be achieved by injecting samples that have
(DelSontro et al., 2016). Briefly, A 500 mL syringe, equipped with a passed through SPE into a UHPLC-MS/MS operating in full-scan of ESI
stopcock valve, was used to collect 250 mL of permeate. Subsequently, positive mode for untargeted analysis. This process was analyzed trip­
250 mL of nitrogen gas was injected into the headspace, and the stop­ licate to detect the mass of compounds that experienced fragmentation
cock valve was immediately closed. The syringe was then agitated for 5 within the m/z range of 50–500, followed by their subsequent frag­
min to establish equilibrium between the liquid and gas phases, in mentation into MS2. Subsequently, the obtained files were imported into
accordance with Henry’s law. The gas in the headspace and from the Compound Discoverer 3.3 SP2 Metabolism software. This involved
AnBB-MBR systems were then analyzed using gas chromatography with adding the compounds of interest obtained from literature by adding
a thermal conductivity detector (GC-TCD). The calculation of dissolved their chemical formulas, structures, and ion masses to the expected
methane from analysis is shown in Text S1 in Supplementary data. compound lists for enhancing the program’s sensitivity in detecting
these compounds. Subsequently, a process workflow was created and
2.4. Quantification of pharmaceuticals and identification of their processed the files from three injections per sample to conduct statistical
intermediate products analysis. Group the samples, remove blank interference masses that
occurred from DI water injection. Furthermore, the software matched
The pharmaceutical concentration in wastewater is extracted by the obtained data with compounds in the environmental database and
solid phase extraction protocol (SPE) by using 6 mL, 200 mg Oasis HLB matched them with the reactions that are likely based on the specified
cartridges (Waters, Milford, MA, USA) equipped with a vacuum mani­ parent compounds (CIP, CAF, SMX, and DCF).
fold pump. Before extraction, the samples were filtered by nylon syringe
filter 0.45 μm to eliminate suspended solids and diluted to 100 mL with 2.5. Microbial consortium analysis
Milli-Q water. The cartridges were first conditioned with 5 mL methanol
followed by 5 mL Milli-Q water. Without getting the column dry, the The biofilms and suspended sludges in compartment 3 of each AnBB-
samples were immediately loaded into the cartridge at the flowrate MBR system were processed with DNA extraction by a DNeasy Power­
around 4 drops/sec. After finishing sample loading, the cartridges were Soil Pro DNA Kit (Qiagen, USA). The 16S rRNA gene, specifically tar­
washed with 5 mL of Milli-Q water and dried under vacuum for 30 min. geting the V3-V4 regions of bacteria and archaea, was amplified using
The analytes were then gently eluted with 5 mL of methanol. The eluents the primers 341F (TCGTCGGCAGCGTCAGATGTGTATAAGAGACAG-
were gently dried with a stream of pure nitrogen to nearly dryness and CCTACGGGNGGCWGCAG) and 805R (GTCTCGTGGGCTCGGA­
reconstituted in 5 ml of 50% methanol in Milli-Q water. Lastly, the GATGTGTATAAGAGA- CAGGACTACHVGGGTATCTAATCC), along
samples were filtered through PP membrane syringe filter (0.22 μm). with 2 × sparQ HiFi polymerase chain reaction (PCR) Master Mix
The samples were then subjected to an ultra-high-performance liquid (QuantaBio, USA). PCR amplification started with an initial denatur­
chromatography with tandem mass spectrometry (UHPLC-MS/MS), Q ation at 98 ◦ C for 2 min, followed by 30 cycles. Each cycle included
Exactive™ Focus MS, UltiMate™ 3000 RSLC System (Thermo Scientific, denaturation at 98 ◦ C for 20 s, annealing at 60 ◦ C for 30 s, and extension

4
T. Buakaew and C. Ratanatamskul Journal of Environmental Management 377 (2025) 124657

at 72 ◦ C for 1 min. A final extension step was performed at 72 ◦ C for 1 kinetic rates. It was observed that UV/O3 exhibited the highest degra­
min after the PCR amplification. The resulting 16S amplicons were then dation rates for CIP, SMX and DCF with k values of 0.0914 min− 1,
purified using sparQ Puremag Beads (QuantaBio, USA). Each sample 0.1989 min− 1 and 0.7038 min− 1 for CIP, SMX and DCF, respectively
was subsequently combined with 5 μL of Nextera XT index primers in a compared to 0.0761 min− 1, 0.1155 min− 1 and 0.3751 min− 1 for O3
50 μL PCR reaction, followed by 8–10 additional amplification cycles treatment of CIP, SMX and DCF, respectively. For CAF, identical
using the same PCR conditions. Bioinformatics analysis of the amplicon degradation rate constants were obtained for both UV/O3 and O3
sequences was conducted using QIIME2 version 2020.8 (Bokulich et al., treatments, with the same k values of 0.012 min− 1.
2018). CAF poses resistance to O3 and hydroxyl radicals compared to other
pharmaceuticals since it comprises stable fused aromatic structure, and
3. Results and discussion the aromatic rings in CAF, which are not highly reactive to ozone. This
might be due to the electron-withdrawing nature of the amide groups,
3.1. The selection of UV, O3 and UV/O3 as the post-treatment system which reduce the electron density (Kim and Tanaka, 2010).
The reason behind the fastest removal observed in UV/O3 system is
From Supplementary Data Fig. S1, UV, O3 and UV/O3 were con­ that the utilization of UV with O3 accelerated the generation of hydroxyl
ducted by introducing 200 μg/L of targeted pharmaceuticals to the three radicals (OH⋅) and H2O2,which can enhance oxidation as Eq. (2) and Eq.
systems in order to investigate which method could yield the highest (3) (Das et al., 2024):
degradation rate of targeted pharmaceuticals over 120 min. The results
show that both O3 and UV/O3 were effective in removal of all targeted O3 + H2O + hν (λ > 310 nm) → O2 + H2O2 (2)
compounds, while UV alone slightly removed CIP without the removal O3 + H2O2 → HO+
2 + OH⋅ + O2 (3)
of other compounds. UV alone relies solely on direct photolysis, which is
less efficient for many pharmaceuticals, and it does not supply sufficient During the photolysis of the ozone, H2O2 as an intermediate product
energy for significant molecular breakdown (Challis et al., 2014; Chin could produce two OH⋅ molecules (Eq. (4)):
and Bérubé, 2005). Liu et al. (2021a) found that independent UV lamps
H2O2 + hν → 2OH⋅ (4)
had minimal impact on CIP removal, whereas O3 and UV/O3 treatments
were able to completely eliminate CIP within 60 min. Notably, the Previous research has proved that the elimination of various phar­
UV/O3 process achieved the highest mineralization of dissolved organic maceutical pollutants in the AOPs primarily occurs via hydroxyl radical
carbon at 62.4%, compared to 31.6% for O3 alone leading to the pro­ oxidation (Wardenier et al., 2019). Moreover, the increased
duction of intermediates and small organics (Liu et al., 2021a). The self-decomposition of ozone also can lead to the formation of hydroxyl
degradation of all targeted pharmaceuticals by O3 and UV/O3 were radicals, which become the primary mechanism for pharmaceutical
appropriately fitted to the pseudo first order kinetic with R2 > 0.90 as removal. Ozone could self-decompose in water to form hydroxyl radical
shown in Fig. 2 to select the best approach that provided the fastest and react with hydroxide ions which naturally occurred in neutral pH

Fig. 2. Degradation kinetics by pseudo first order of O3 and UV/O3 in removal of targeted pharmaceuticals at 200 μg/L in DI water.

5
T. Buakaew and C. Ratanatamskul Journal of Environmental Management 377 (2025) 124657

(Koundle et al., 2024), as shown in Eq. (5) and Eq. (6): Supplementary Data Fig. S2. With the microaeration condition, the ef­
ficiency of COD removal slightly increased from 81.82% to 84.13%.
O3 + H2O → 2HO2 ◦
(5) It is noticeable that almost all COD removal occurred through the
O3 + OH⋅ → O2 + O−2 (6) biological processes with the AnBB-MBR systems. When the UV/O3
system was applied, there was no significant reduction in COD. This
In both ozonation and UV/O3 processes, ozone also can react with the might be because the transformation of the pharmaceutical compounds
free radicals formed, resulting in the generation of hydroxyl radicals, as and other organic molecules resulted in the intermediate products as
shown in Eq. (7) and Eq. (8): further discussed in Section 3.7, rather than complete oxidation into
H2O2 and CO2. This is consistent with previous research findings where
O3 + O−2 → O2 + O−3 (7)
assessments of total organic carbon (TOC) during O3 trials indicated
O−3 + H → O2 + OH⋅
+
(8) limited mineralization, even when CAF was completely eliminated into
various intermediate compounds (Rosal et al., 2009). The removal of
Thus, combining UV with the O3 system led to a notable improve­ nitrogen and phosphorus demonstrated that biological treatment pro­
ment in the removal rate and mineralization efficiency. Previous cesses using the AnBB-MBR systems were significantly more effective
research found that O3 treatment at 30 mg/L/min and UV/O3 (with UV than UV/O3. Based on the dominant microorganisms, nitrogen removal
intensity of 2.6 mW/cm2 and O3 at 30 mg/L/min) could similarly in the AnBB-MBR system with microaeration (R2) was highly efficient,
destroy CIP at 100 mg/L within 60 min. However, in the UV/O3 con­ facilitated by Candidatus Brocadia, Nitrosomonas, Nitrospira, Deni­
dition, dissolved organic carbon was reduced by 64.2%, compared to tratisoma, and Thiobacillus. In contrast, nitrogen removal in R1 primarily
31.6% in the O3 alone condition which demonstrated more effective of resulted from nitrogen assimilation for cell growth. For phosphorus
UV/O3 process than O3 alone (Liu et al., 2021a). Moreover, UV/O3 removal, Bacillus was responsible in R1, while Denitratisoma played a key
process enhanced SMX removal compared to O3 alone by leveraging role in R2. A detailed explanation of the functional mechanisms of these
non-selective oxidation through OH⋅ formation (Liu et al., 2021c). This microbial groups can be found in Section 3.3 and is illustrated in Fig. 5.
study also confirmed that the possible formation of OH⋅ from O3 Moreover, although ammonia and nitrite can be partially oxidized by
photolysis of UV/O3 was likely attributed to the enhancement of phar­ ozone or OH⋅ in the UV/O3 process, the final product remains nitrate,
maceutical degradations. Therefore, UV/O3 has been selected as the post which was still a form of nitrogen in wastewater. Similar result of the
treatment system for removal of targeted pharmaceutical residues from residual nitrogen after the ozonation was also reported by Luo et al.
the two AnBB-MBR systems. (2015). Both nitrate and phosphorus are chemically stable and unreac­
tive with ozone or hydroxyl radicals (Zhang et al., 2023b), rendering
3.2. Performances of the novel integrated system of AnBB-MBRs (with nitrogen and phosphorus removal through the UV/O3 process insuffi­
and without microaeration) and UV/O3 photoreactor for building ciently effective.
wastewater treatment For color removal, it was observed that the overall color removal
efficiency of R1 could achieve up to 57.6% after the utilization of UV/
Table 1 illustrates the overall treatment efficiencies of the integra­ O3, and the removal efficiency could increase up to 76.3% with the
tion of the two AnBB-MBR systems. It was observed that the R1 with UV/ integration of UV/O3 system with the microaeration system of R2. The
O3 could eliminate the four targeted pharmaceuticals in the range of results proved that the integration of AnBB-MBR system with microa­
97.31–100%, while the R2 with UV/O3 could eliminate them in the eration followed by UV/O3 system can enhance the removal of targeted
range of 99.47–100%. This indicates the high efficiency of the proposed pharmaceuticals performance compared to anaerobic systems which
integrated AnBB-MBR and UV/O3 system, regardless of anaerobic or was attributed to the reduction of organic and nutrient elements in
microaerobic conditions of the AnBB-MBR system. It was found that DCF wastewater under microaerobic conditions, as illustrated in Table 1. The
could be completely eliminated (100% removal) for both conditions presence of high organic compounds and ammonia concentrations, in
within 150 min, while SMX was completely removed when UV/O3 was anaerobic effluent can lead to higher consumption of ozone and OH⋅ in
coupled with the R2 under microaerobic conditions. For the removal of the oxidation of these substances, rather than being directly utilized for
CIP, it was found that both conditions were not significantly different pharmaceutical removal (Luo et al., 2015).
(99%). Additionally, the efficacy of CAF removal by the R1 with UV/O3 For the analysis of COD balance, as depicted in Fig. 3(a) and (b), it
was 97.31%. Whereas the efficiency of CAF removal by the R2 with UV/ was conducted to understand the change in COD within the two AnBB-
O3 was up to 99.67%. MBR systems and to demonstrate the effects of microaeration on COD
The removal of COD, nitrogen, phosphorous and colors through UV/ conversion by using Eq. (9) and Eq. (10):
O3 treatment of the permeates from R1 and R2 are shown in

Table 1
Removal performances of AnBB-MBRs in anaerobic and microaerobic integrated with UV/O3.
Parameters Influent R1: AnBB-MBR + UV/O3 photoreactor R2: AnBB-MBR with microaeration + UV/O3 photoreactor

Permeate of R1 150 min UV/O3 % Overall removal efficiency Permeate of R2 150 min UV/O3 % Overall removal efficiency

CIP (μg/L) 250 30.52 ± 1.87 1.12 ± 0.02 99.55% 35.02 ± 1.54 1.32 ± 0.09 99.47%
CAF (μg/L) 250 7.41 ± 0.11 6.73 ± 0.31 97.31% 3.44 ± 0.16 0.83 ± 0.16 99.67%
SMX (μg/L) 250 47.38 ± 0.74 0.70 ± 0.37 99.72% 8.70 ± 0.31 NF 100%
DCF (μg/L) 250 129.13 ± 4.22 NF 100% 67.28 ± 0.72 NF 100%
COD (mg/L) 245.80 ± 94.18 45.67 ± 1.15 44.67 ± 4.93 81.82% 37.67 ± 1.53 39.00 ± 1.73 84.13%
NH+4 -N (mgN/L) 83.30 ± 20.10 59.00 ± 0.50 54.83 ± 0.29 34.17% 17.17 ± 0.29 17.17 ± 0.29 78.91%
NO−2 -N (mgN/L) 0.01 ± 0.01 0.02 ± 0.02 0.01 ± 0.01 – 1.18 ± 0.03 0.12 ± 0.03 –
NO−3 -N (mgN/L) 1.50 ± 1.10 2.10 ± 1.30 2.40 ± 1.25 – 9.53 ± 0.12 13.60 ± 0.11 –
PO3-
4 -P (mgP/L) 11.80 ± 6.00 5.50 ± 0.12 5.50 ± 0.03 53.39% 6.43 ± 0.12 6.23 ± 0.23 47.20%
SS (mg/L) 255.83 ± 20.61 0 0 100% 0 0 100%
VSS (mg/l) 165.00 ± 31.10 0 0 100% 0 0 100%
pH 7.43 ± 0.01 7.33 ± 0.10 7.35 ± 0.10 – 6.61 ± 0.20 6.82 ± 0.20 –
Color (ADMI) 80.19 ± 15.4 82.33 ± 1.53 34.00 ± 1.00 57.60% 61.00 ± 2.65 19.00 ± 0.57 76.30%

Note: NF is not found.

6
T. Buakaew and C. Ratanatamskul Journal of Environmental Management 377 (2025) 124657

Fig. 3. COD balance in: (a) R1: AnBB-MBR and (b) R2: AnBB-MBR with microaeration and (c) distributions of nitrogen species along the two integrated systems.

proportion of COD in the sludge was only 1%.

InfluentCOD = EffluentCOD + SludgeCOD + DissolvedCH4COD + The systems with the presence of oxygen could result in higher
GasCH4COD + BiodegradedCOD (9) sludge growth yield compared to anaerobic systems (Koshikawa et al.,
2022). Kassab et al. (2010) reported the sludge yields of 0.06
gVSS/gCOD and 0.4 gVSS/gCOD for anaerobic and aerobic processes,

1.42 mgCOD 4 mgCOD 16g/mol 4 mgCOD

Qin × CODin = Qout × CODout + MLSS × × Qw + CDCH4 × × Qout × %CH4 × QCH4 × × +
mgMLSS mgCH4 22.4L/mol mgCH4 (10)
Biodegraded COD

where Qin and Qout are the influent and effluent flowrates, respectively; respectively. The average MLSS and MLVSS concentrations in each
CODin and CODout are COD concentrations of the influent and effluent, compartment of the two AnBB-MBR systems are shown in Supplemen­
respectively; MLSS is mixed liquor suspended sludge; Qw is sludge tary data Fig. S3. Additionally, microaeration significantly increased the
withdrawal rate; CDCH4 is dissolved methane concentration in the proportion of biodegraded COD by 31% (from 42% to 73%), and
effluent; %CH4 is methane percentage in biogas and QCH4 is biogas reduced the proportion of dissolved methane in the effluent from 38% to
production rate. It was found that both R1 and R2 systems exhibited a 5%. Geng et al. (2023) found that 20% and 4.3% of the influent COD
unique proportion of COD in the effluent at 18–19%. However, with the from the integrated anaerobic biofilm reactor-step feed MBR system was
application of microaeration, the proportion of COD in the sludge converted into dissolved methane and sludge, respectively.
increased to 4%, whereas in the absence of microaeration, the The distribution of nitrogen species along the two integrated systems

7
T. Buakaew and C. Ratanatamskul Journal of Environmental Management 377 (2025) 124657

Fig. 4. 3DFEEM spectra of permeate of R1 (a); permeate of R2 (b); R1 with UV/O3 for 150 min (c); and R2 with UV/O3 for 150 min (d).

is shown in Fig. 3 (c). It was found that ammonium nitrogen concen­ tryptophan protein-like substances) was observed, decreasing from
trations decreased by 35% and 79.21% for R1 and R2, respectively 2200 ATU of R1 to 1600 ATU of R2. It is clearly demonstrated that the
(Table 1). This result proved that the use of microaeration can signifi­ utilization of microaeration could reduce protein content, aligning with
cantly enhance the efficiency of nitrogen removal. Additionally, the our previous research findings that highlighted the significance of pro­
effluent from the UV/O3 system showed an increase in nitrate levels teins originating from anaerobic systems as crucial contributors to
from the oxidation of ammonia by ozone and OH⋅ to form nitrate as the membrane fouling compared to microaerobic conditions (Buakaew and
final products (Luo et al., 2015). Ratanatamskul, 2023). In line with this study, it was observed that the
In addition, 3D FEEM spectra was used to provide evidence of the average membrane fouling rate of R2 was 21.73 ± 8.89 mmHg/d,
remaining organic compounds. Fig. 4 illustrates the organic compounds whereas the membrane fouling rate of R1 was 61.17 ± 49.19 mmHg/d
which were found in the treated effluents from the integrated systems of (Supplementary Data Fig. S4). When the permeates from both R1 and R2
AnBB-MBR and UV/O3. The fluorescent peak A, with the location of EX/ were further treated by the UV/O3 process, the reduction in the intensity
EM at 220–250 nm/330–380 nm, represents tryptophan protein-like of all peaks was observed as shown in Fig. 4(c) and (d). This indicates the
substances, and Peak B at EX/EM of 250–300 nm/300–350 nm is efficacy of UV/O3 in reducing various major organic groups detectable
observed for the soluble microbial product-like substances (Chen et al., through 3D FEEM, which confirms its efficiency by degrading a broad
2003). For humic-like substances indicated as peak C, it is located at spectrum of pollutants.
EX/EM in both of 300–350 nm/400–450 nm and 260 nm/380–460 nm
(Coble, 1996). The fluorescence area of peak D at EX/EM between 225 3.3. Microbial community inside the AnBB-MBR system
and 300 nm/325–400 nm arises from the presence of its conjugated
aromatic ring (Zhang et al., 2023a). To understand the removal mechanisms, Fig. 5 illustrates % relative
In comparing the permeates between R1 (Fig. 4 (a)) and R2 (Fig. 4 abundance of microbial consortium at the genus level, with values
(b)), a notable reduction in the peak intensity of peak A (associated with greater than 1%, in both suspended sludges and biofilms in the

8
T. Buakaew and C. Ratanatamskul Journal of Environmental Management 377 (2025) 124657

Fig. 5. Relative abundance at the genus level of microorganisms in C3 compartment samples with values greater than 1% of (a) R1: AnBB-MBR and (b) R2: AnBB-
MBR with microaeration.

9
T. Buakaew and C. Ratanatamskul Journal of Environmental Management 377 (2025) 124657

compartment 3 of each AnBB-MBR system. compartment 3 of R2, which has been reported to be abundant under
From Fig. 5(a), the microbial groups responsible for the degradation low-nutrient conditions (Leonel et al., 2022). Also, this aligns with the
of organic matter in R1 include Caldisericum, Romboutsia, Clos­ microaerobic environmental condition.
tridium_sensu_stricto_1, Mycobacterium, and Methanosaeta, which play a The microbial groups involved in enhancement targeted pharma­
role in methane production (Mori et al., 2009; Tiwari et al., 2021), with ceutical removal in R2 include AOB (Nitrosomonas), which was reported
the observed concentrations exceeding 1% in R1. Saccharimonadales, to have co-metabolism of CIP and SMX with the oxygen availability of
Ferruginibacter, OLB14, and Pir4 lineage are the primary microbial 0.037 L/g.VSS/d and 0.069 L/g.VSS/d, respectively (Li et al., 2022).
communities responsible for the degradation of organic matter in R2 AOB and methanotrophs can stimulate the co-metabolism of CAF during
(Kochetkova et al., 2020; Yang et al., 2021b). Methylosarcina, found nitrification activity and methanotrophic activity, respectively with
more than 1% in compartment 3 of R2 under microaeration conditions, 100% removal within 200 h compared to only 20% removal in the
plays a critical role by specializing in the methane metabolism (Wei inhibited condition (Wang et al., 2022). SMX was reported to be
et al., 2016). As a result, the methane emissions could be significantly removed by the co-metabolism of methanotrophs (Benner et al., 2015).
reduced as depicted in Fig. 5(a) and (b). Nitrifying activity was correlated with enhanced biotransformation of
It was found that R1 lacked microorganisms associated with nitrogen DCF with transformation rate at 7.4–11.1 μg/g.biomass compared to
removal. Therefore, it is hypothesized that the reduction in nitrogen 0.4–3.4 μg/g.biomass as nitrifying activity was inhibited (Kolakovic
(34%) could be associated with nitrogen utilization by microbial cell et al., 2022). For anaerobic microbial groups in R1, Bacillus was reported
growth and adsorption. A reduction in total nitrogen at 32.5% was also for the possible biodegradation of antibiotics (Wu et al., 2024). It was
observed in the staged anaerobic fluidized membrane bioreactor system reported that Bacillus sp. could degrade CIP at a rate of 0.25 mg/day,
treating municipal wastewater (Chen et al., 2019). In compartment 3 of while Bacillus subtilis could degrade CIP at a rate of 0.35 mg/day, in a
R2, the microorganisms responsible for nitrogen removal were detected medium containing CIP at a concentration of 5 mg/L (Liyanage and
in both the suspended sludge and biofilms (Fig. 5 (b)). The process began Manage, 2018). Longilinea,which is facultative anaerobic microbes, was
with Candidatus Brocadia, a microorganism responsible for ammonia reported for enhancing SMX removal in AnMBR (Wang et al., 2023).
oxidation under anaerobic conditions (Anammox), by reacting with ni­ Clostridium, Bacillus and Lactivibrio linked to co-metabolic biodegrada­
trite to convert ammonia directly into nitrogen gas and a small amount tion of CIP and SMX in an anaerobic fixed bed biofilm reactor (Carneiro
of nitrate (Meng et al., 2019) leading to the increase of nitrate concen­ et al., 2020). Some species of Mycobacterium also could degrade complex
tration in the effluent from R2 (Fig. 5 (c)). In addition, Nitrosomonas, an aromatic hydrocarbon molecules (Boorgula et al., 2024).
ammonia oxidizing bacteria (AOB), converts ammonia to nitrite in the
presence of oxygen. This was followed by nitrite oxidizing bacteria
(NOB) by Nitrospira, which converted nitrite into nitrate. Subsequently, 3.4. Degradation kinetics of targeted pharmaceuticals by UV/O3 of the
denitrifying bacteria such as Denitratisoma and Thiobacillus transformed AnBB-MBR permeates under anaerobic and microaerobic conditions
the nitrate into nitrogen gas, effectively, which can remove nitrogen
from wastewater through the process of denitrification. The denitrifi­ The experimental results of the hybrid model coupling UV/O3 system
cation process typically occurs under anoxic conditions, which are more and AnBB-MBR system under both anaerobic condition (R1) and
prevalent in the deeper layers of biofilms or large flocs, where oxygen microaerobic condition (R2) for the removal of the four targeted phar­
penetration is limited (Liu et al., 2021b). Denitratisoma had also been maceuticals are shown in Supplementary data Fig. S5. The obtained
reported to perform simultaneous denitrification and phosphorus up­ results fitted well with the pseudo first order reaction kinetics (R2 >
take under anoxic conditions (Shi and Lee, 2007), and it is likely the 0.89) for all compounds (Fig. 6(a) and (b)). Since the initial concen­
main microbial group responsible for phosphorus removal in R2. For trations of the remaining targeted pharmaceuticals in the permeate of
phosphorus removal in R1, Bacillus, known for their versatility, some of R1 and R2 before subjected to UV/O3 process were different, direct
their species could be able to accumulate phosphorus under specific comparison of reaction rate kinetics between anaerobic and micro­
environmental conditions (Prakash and Arora, 2019). In addition, ni­ aerobic conditions was not feasible.
trogen and phosphorus removal under anaerobic conditions can occur From Fig. 6 (a), the observed highest k value was 0.0907 min− 1 for
primarily through assimilation for microbial growth and adsorption DCF in the permeate of R1 (under anaerobic conditions), whereas the
onto anaerobic sludge (An et al., 2023). RBG-13-54-9 was detected in lowest k value was 0.004 min− 1 for CAF. This discrepancy was likely
attributed to the high sensitivity of DCF to O3, whereas CAF has been

Fig. 6. Degradation kinetics by UV/O3 of targeted pharmaceuticals in the permeate of R1 (anaerobic) (a) and R2 (microaerobic) (b).

10
T. Buakaew and C. Ratanatamskul Journal of Environmental Management 377 (2025) 124657

reported to exhibit greater resistance to O3 compared to all other tar­ formation of intermediate products with m/z = 348.13 on the benzene
geted pharmaceuticals (Paucar et al., 2019). For the application of ring, potentially giving rise to IP-CIP1a, or on the piperazine ring,
UV/O3 in treating the permeate from the microaerobic R2 system as resulting in IP-CIP1b. Previous studies on the biodegradation of CIP in
shown in Fig. 6 (b), it was found that CAF exhibited the k value at 0.0338 anaerobic sulfate-reducing bacteria and anaerobic digestion also found
min− 1, which is significantly greater than 0.004 min− 1 in the permeate the possible formation of the compounds IP-CIP1a and IP-CIP1b (Jia
of R1. This was likely due to the presence of lower concentrations of et al., 2018; Tang et al., 2022). Additionally, the aerobic granular sludge
organic matter, nutrients, and pharmaceuticals in the permeate from the process, composed of both aerobic and anaerobic microbes, could
R2 compared to the permeate from the R1. Consequently, the integra­ facilitate the production of hydroxylation products at the piperazine
tion of UV/O3 as the post-treatment for the microaerobic R2 system ring (Zhao et al., 2024b). Therefore, it is reasonable to expect the for­
resulted in significantly higher reaction rates of CAF. For CIP, it was mation of these compounds in both R1 and R2 systems, which contain
observed that the initial concentrations prior to entering the UV/O3 anaerobic microbial communities. IP-CIP1c (m/z = 348.13) is
system were relatively equal, approximately 30 μg/L. Experimental re­ desethylene-N-acetylciprofloxacin, which can occur by the desethyla­
sults further revealed that the rate constants (k) were also similar, with tion of C2H2 in the piperazine ring of CIP molecule, which was also
values close to 0.020 min− 1. found in both R1 and R2 systems. The fungus Pestalotiopsis guepini was
reported for producing this product during the biotransformation of CIP
3.5. Ozone concentration profiles (Parshikov et al., 2001). Moreover, the acetylation product of CIP on the
piperazine ring might be generated as IP-CIP5 with m/z 374.15, which
From Supplementary Data Fig. S5, it illustrates the dissolved O3 has been previously detected in Thermus thermophilus. This thermo­
concentration over time of continuous O3 supply to assess the amount of philic anaerobic bacterium, capable of degrading CIP, was isolated from
O3 utilized in the O3 and UV/O3 systems for treatment of the targeted sludge sampled from a pharmaceutical factory (Pan et al., 2018).
pharmaceuticals. The concentrations of the dissolved O3 gradually For the intermediate products of CIP degradation, the intermediate
increased over time until reaching a steady state in all experiments. product IP-CIP2a (m/z = 330.14), likely forms through defluorination of
When comparing the dissolved O3 between the O3 and UV/O3 processes IP-CIP1a, possibly accompanied by hydroxylation on the benzene ring.
in the DI water, it was found that the dissolved O3 concentration was IP-CIP2a (m/z = 330.14) is probably formed by defluorination of IP-
lower through the experimental runs of the UV/O3 system. This is CIP1a or defluorination followed by hydroxylation on benzene ring. In
attributed to the UV radiation reacting with O3 to produce 2 molecules addition, it was possible to replace fluorine atom of CIP with hydrox­
of OH⋅. This can serve as an aiding factor in enhancing the efficiency of ylation, which was resulted in the production of IP-CIP2b (m/z =
pollutant removal in wastewater (Zhu et al., 2023), as it can generate 330.14). N-formylciprofloxacin (IP-CIP4 with m/z = 360.13) and
two OH⋅ molecules from a single O3 molecule. For O3, it comprises direct cleavage on the quinolone ring of CIP to form IP-CIP3 (m/z = 353.15)
O3 and indirect O3, where O3 reacts with water to form OH⋅ (Chiang could be found in the anaerobic permeate of R1 in this study. These
et al., 2006) which occurs in both O3 and UV/O3 processes. products were also reported in anaerobic digestion of CIP in various
The dissolved O3 concentration of the UV/O3 system was lowered in anaerobic treatment systems (Carneiro et al., 2022; Tang et al., 2022).
in the permeate of R1 than in the permeate of R2 through the operating The biotransformation pathways of CAF exhibited similarity in both
time and reached constant values of 0.11 mgO3/L and 0.37 mgO3/L at R1 and R2 permeates by the oxidative pathways (Fig. 7 (b)). CAF was
120 min for the permeates of R1 and R2, respectively. This might be due possibly oxidized to form 1, 3, 7 trimethyl uric acid and then possibly
to the higher amount of O3 was used to oxidize the remaining organics degraded to IP-CAF1 and IP-CAF2 (m/z = 213.09). In addition, IP-CAF3
and pollutants in the permeate of R1, which had higher concentrations (m/z = 201.097) or 3, 6, 8-trimethylallantoin could also be formed by
of residual pollutants than in the permeate of R2 system (Table 1). When 1,3,7 trimethyl uric acid likewise the observation of CAF catabolism in
comparing the application of UV/O3 for the targeted pharmaceuticals the microbial cultures comprising Klebsiella and Rhodococcus
between DI water and real wastewater, it was observed that the (Mohapatra et al., 2006).
remaining dissolved O3 concentration in the DI water was significantly From Fig. 8 (a), both similarities and differences were observed in
higher, becoming evident after a duration exceeding 45 min. Despite the the biodegradation of SMX between the R1 and R2 permeates. For the
higher concentration of the targeted pharmaceuticals in the DI water, bio-intermediates found in both permeates, SMX could be probably
this was attributed to the increased utilization of O3 for oxidation due to converted to N4-hydroxyl-SMX, and a possible conversion of N4-
the presence of various pollutants in the real building wastewater. hydroxyl-SMX to IP-SMX1 (m/z 258.09) by attacking the sulfonyl
group, which was in accordance with the SMX removal in anoxic con­
3.6. Bio-intermediate products of targeted pharmaceuticals of the AnBB- dition of anoxic/oxic moving bed biofilm reactor (Saidulu et al., 2024).
MBR systems under anaerobic and microaerobic conditions This was in accordance with this study, which involves both anaerobic
and aerobic microorganisms in anaerobic and microaerobic conditions.
The experimental results for identifying the intermediate products of The ring cleavage of N-O bond by hydroxylation in reductive conditions
the targeted pharmaceuticals were obtained from the permeate water of SMX could form unstable intermediate (SMX+) followed by IP-SMX2
from R1: AnBB-MBR and R2: AnBB-MBR with microaeration at steady (m/z 254.059) that were also found in both of R1 and R2 permeates.
state. This was achieved by utilizing formulas, reference m/z ions, and This has been observed in previous research with the environmental
transformation reactions generated from the Compound discoverer conditions similar to the current study of the biological conversion of
program 3.3 SP2 metabolism to match the intermediate products found SMX in an autotrophic denitrification system (Zhang et al., 2020).
in previous literatures. This data was input into the program to assist in IP-SMX3c (m/z 256.07) was found in anaerobic sulfate-reducing bac­
identifying masses that correspond to the desired substances, as shown teria sludge system (Jia et al., 2017) and anaerobic fixed bed bioreactor
in Supplementary Data Tables S1, S2, S3 and S4 in order to propose the (Carneiro et al., 2022). The breakdown of isoxazole ring by hydroge­
possible mechanisms and biodegradation pathways of R1 and R2 and the nation, hydroxylation and reduction might lead to the formation of m/z
degradation pathways by the UV/O3 photoreactor. = 256.07 which were IP-SMX3a, IP-SMX3b and IP-SMX3c, respectively.
The experimental results depicted in Fig. 7 (a) reveal significant Additionally, various hydroxylated intermediate products (m/z =
differences in the intermediate products generated from the CIP removal 270.05) could be generated in many positions of SMX such as IP-SMX4a,
pathways in the R1 and R2 systems. Specifically, it was observed that the IP-SMX4b, IP-SMX4c, and IP-SMX4d. Some previous reports can confirm
intermediate products generated by R1 were more abundant compared this finding (Carneiro et al., 2022; Yang et al., 2021a; Zhang et al.,
to the case of R2 with microaeration. Both R1 and R2 systems yielded 2020). All these mentioned bio-intermediate products were found in
intermediate products through the hydroxylation process, leading to the both R1 and R2 permeates, which might be due to the occurrence of the

11
T. Buakaew and C. Ratanatamskul Journal of Environmental Management 377 (2025) 124657

Fig. 7. Bio-intermediate products of CIP (a) and CAF (b) in the permeates from R1: AnBB-MBR (Blue arrow) and from R2: AnBB-MBR with microaeration
(Red arrow).

12
T. Buakaew and C. Ratanatamskul Journal of Environmental Management 377 (2025) 124657

Fig. 8. Bio-intermediate products of SMX (a) and DCF (b) in the permeates from R1: AnBB-MBR (Blue arrow) and from R2: AnBB-MBR with microaeration
(Red arrow).

13
T. Buakaew and C. Ratanatamskul Journal of Environmental Management 377 (2025) 124657

same anaerobic microbes in both R1 and R2 systems, which also can IP-CAF1 was detected in the urban sewer WWTP system. Moreover,
thrive under microaerobic conditions. IP-DCF1a, IP-DCF1b, IP-DCF1c, IP-DCF3a, IP-DCF5, IP-DCF6a, and
The intermediate IP-SMX5 (m/z 296.07), identified only in the IP-DCF6b have been detected in full-scale municipal WWTPs (Jewell
microaeration permeate of R2, was likely formed through SMX acety­ et al., 2016; Nguyen et al., 2021). For IP-DCF1a, IP-DCF1b, IP-DCF1c,
lation, which was also observed in oxygen-utilizing in the fixed-film and IP-DCF5, they have been reported to be originated from the
activated sludge membrane bioreactors IFAS-MBR systems (Hou et al., biodegradation of DCF by nitrifying and heterotrophic bacteria (Wu
2022). After that, IP-SMX5 tended to have the cleavage between sulfur et al., 2020).
and benzene ring followed by hydroxylation to be IP-SMX7 called Due to the difference in wastewater characteristics, microorganisms,
N-(2-hydroxyphenyl) acetamide (m/z = 152.07). reactor configurations, pharmaceutical types and concentrations, the
For analysis of the anaerobic products in the permeate of R1, SMX fate and removal pathways of pharmaceuticals can be different. There­
was possibly oxidized, followed by SO2 elimination and nitroso substi­ fore, it is crucial for future research to thoroughly investigate the fate of
tution to form IP-SMX6 (m/z 231.04). The similar observation through pharmaceuticals and their metabolic products in the anaerobic digestion
the heterologous expression of laccases from anaerobic bacteria of SMX of sewage sludge and full-scale WWTPs under various operating condi­
was also confirmed by (Yang et al., 2021a). From Fig. 8 (b), the tions. This will provide the necessary data for designing more efficient
biotransformation of DCF in both permeates of the R1 and R2 systems wastewater treatment systems aimed at the removal of pharmaceutical
could lead to possible hydroxylation reactions, resulting in the forma­ contaminants from wastewater.
tion of compounds with m/z = 312.02 at multiple positions. For
example, the hydroxylation product that occurred in the phenyl group 3.7. Reduction of intermediate products of the targeted pharmaceuticals
devoid of chlorine led to the formation of 5-hydroxyldiclofenac by the UV/O3 post-treatment system
(IP-DCF1a), and the occurrence within the phenyl group containing
two chlorine atoms resulted in the formation of either 4-hydroxyldiclo­ The examination of intermediate products of the targeted pharma­
fenac (IP-DCF1b) or 3-hydroxyldiclofenac (IP-DCF1c). ceuticals remaining after the UV/O3 process involved testing the water
Besides the hydroxylation route in both R1 and R2, DCF could also be samples at the steady state condition of the UV/O3 treatment of the R1
converted to IP-DCF5 or DCF-lactam (m/z = 278.01) through direct and R2 permeates (after 150 min). This investigation reveals the inter­
amidation, which was found in both R1 and R2 permeates. Nevertheless, mediate compounds that could be generated from the UV/O3 treatment
DCF-lactam has been identified as an unstable transformation product, as listed from the literature as shown in Supplementary Data Tables S1,
as it has the potential to revert to DCF (Wu et al., 2020). The hydrox­ S2, S3 and S4. Moreover, the types of bio-intermediates, which could be
ylated products had also been identified in the activated sludge pro­ highly eliminated through the UV/O3 process, could be identified.
cesses, under both nitrification and heterotrophic conditions and in From Fig. 9 (a), the occurrence of IP-CIP6 (m/z = 333.14), which
anaerobic fermentation of waste activated sludge (Nguyen et al., 2021; was likely attributed to the hydroxylation at the piperazine ring (Liu
Poirier-Larabie et al., 2016; Yang et al., 2022), which allowed for the et al., 2012), was the initial step of the CIP degradation in the permeates
detection in both anaerobic and microaerobic conditions of R1 and R2, from the R1 and R2 systems via the UV/O3 process. Then, the next step
respectively. It has been speculated that hydroxylation was facilitated by followed by the detachment of an OH molecule from the carboxylic
the cytochrome P450 system as well as mono- and dioxygenases group, resulting in a positively charged structure and subsequent
(Kolakovic et al., 2022). defluorination reaction by superoxide anion radical. Furthermore,
Moreover, the results from Fig. 8 (b) show that various microaerobic additional hydroxylation could occur, leading to the formation of
intermediate products of DCF generated from the R2 system were IP-CIP7 (m/z = 329.14). IP-CIP8 with m/z = 315.12, produced from the
observed. For example, 5-hydroxyldiclofenac (DCF-B1a) could possibly reaction involving singlet oxygen and superoxide anion radical at the
be converted to IP-DCF2 (m/z 312.02) via oxidation at benzene ring piperazine ring, could result in the formation of a keto-derivative of the
under microaerobic conditions. This oxidation of the benzene ring in piperazine ring (Liu et al., 2021a). It was found that the UV/O3 reaction
oxygen availability could lead to ring cleavage, facilitating the complete for treatment of the R1 and R2 permeates led to the formation of
degradation of DCF in biological wastewater treatment (Wu et al., IP-CIP7. Moreover, this possibly resulted in the formation of the final
2020). Oxidation via dehydrogenation of hydroxyl group of 5-hydroxyl­ product of IP-CIP8, which was also found by Liu et al. (2021a). Toxicity
diclofenac and 4-hydroxyldiclofenac were capable for the formation of testing during the application of the UV/O3 process revealed that
IP-DCF3a and IP-DCF3b, respectively, and the conversion of DCF-lactam IP-CIP7 and IP-CIP8 exhibited lower toxicity compared to CIP (Liu et al.,
could generate these two intermediate products. Similarly, IP-DCF4 2021a).
(m/z = 326.03) could be generated in the microaerobic system of R2 From Fig. 9 (b), the compound IP-CAF4 (m/z = 243.07) is the pro­
through the hydroxylation followed by a subsequent methylation reac­ posed intermediate product resulting from the reaction of ozonation,
tion. Moreover, the intracellular amidation of 4-hydroxyldiclofenac and leading to the ring opening at the N9=C8 position of the purine structure
5-hydroxyldiclofenac in the R2 were likely attributed to the formation of which could generate a molecule known as molozonide before trans­
IP-DCF6a and IP-DCF6b with m/z = 294.01. These microaerobic prod­ forming into the compound IP-CAF4 subsequently (Rosal et al., 2009).
ucts mentioned above were detected only in the R2 system and they This compound was found only in the case of CAF removal by the
were not found under the anaerobic conditions of the R1 system. This UV/O3 treatment of the DI water containing CAF. The compound IP-
finding aligns with previous studies on wastewater treatment systems CAF5 (m/z = 199.08), which was found only in the UV/O3 treatment
involving the presence of oxygen in the heterotrophic sludge systems, of the DI water containing CAF, is referred to as 6-amino-5-(N-formyl­
enhanced biological phosphorus removal processes, and hybrid bio­ methylamino)-1,3-dimethyluracil. This compound could be derived
film–activated sludge processes (Jewell et al., 2016; Nguyen et al., 2021; from the loss of CO-N-CH3 in the six-membered ring (Rosal et al., 2009).
Wu et al., 2020). The compound IP-CAF6 (m/z = 181.07) was a product of the deme­
When comparing the intermediate products of targeted pharma­ thylation process, which could be detected in the DI water, treated with
ceuticals in the AnBB-MBR systems from this study with those detected the UV/O3 system. Subsequently, under microaerobic conditions, the
in the full-scale systems, it was found that there is a lack of studies on the compound IP-CAF6 might not be stable. And then it could probably
intermediate products during full-scale anaerobic digestion of sewage undergo oxidation, leading to the opening of the imidazole ring and the
sludge. Additionally, there are only a limited number of studies inves­ formation of IP-CAF7 with m/z = 197.07 (Rosal et al., 2009). However,
tigating intermediate products in full-scale WWTPs (Akay et al., 2024; the related intermediate products of CAF under the UV/O3 oxidation in
Jewell et al., 2016; Nguyen et al., 2021; Ren et al., 2021). From a pre­ the anaerobic condition were not found in this study.
vious study by Ren et al. (2021), the same intermediate compound of As shown in Fig. 10 (a), the formation of the IP-SMX8 compound (m/

14
T. Buakaew and C. Ratanatamskul Journal of Environmental Management 377 (2025) 124657

Fig. 9. Feasible intermediate products of CIP (a) and CAF (b) found in the effluents by UV/O3 degradation; of DI water (Black arrow), permeate of R1 (Blue arrow)
and permeate of R2 (Red arrow).

z = 128.07), which was an intermediate product observed under both confirmed the possibility of this intermediate occurrence.
anaerobic (R1) and microaerobic (R2) conditions, is likely due to the In addition, 5-hydroxyldiclofenac was unstable under oxidative
cleavage of the N7-S8 bond. Subsequently, the benzene ring undergoes conditions (Alharbi et al., 2022) and then it could be further oxidized to
hydroxylation. The bond between sulfur (S) and the benzene ring is then form IP-DCF2 or DCF-2, 5-iminoquinone as appeared in this study.
cleaved, followed by sequential hydroxylation reactions. This is also However, these hydroxylated products also could be generated through
confirmed with a previous study by (Zhang et al., 2023a), who applied the AnBB-MBR systems under microaerobic (R2) conditions (Fig. 10
the microbubble ozonation method for SMX degradation. Furthermore, (b)). IP-DCF8, named as 2-((2,6-dichlorophenyl) amino) benzaldehyde,
the degradation of the isoxazole ring structure of the SMX molecule was found in the treated effluent from the UV/O3 treatment of the R1
could lead to the formation of another intermediate product, IP-SMX9 effluent. Alharbi et al. (2022) also reported this intermediate product
compound with m/z = 230.06 as illustrated in Fig. 10 (a). from the ozonation process of DCF, which could involve reactions with
From Fig. 10 (b), it was observed that the intermediate products from O3 or OH⋅ at the methylene position on the carbocyclic moiety, followed
the reaction of the UV/O3 system for DCF degradation in the effluent by the release of CO2. The IP-DCF9 (m/z = 282.08) was likely a product
from the anaerobic R1 system exhibit a diverse range of compounds, that was formed from the reaction of ozone on the acetate position,
with the highest abundance being compounds resulting from the hy­ leading to the release of CO2 and the oxidation of the carbonyl carbon on
droxylation process. This hydroxylation could occur through either the the benzene ring to form an aldehyde group, followed by a subsequent
ozone addition or the generation of hydroxyl radicals on benzene rings hydroxylation process. This intermediate product was also reported by
or nitrogen molecules within its structure. (Patel et al., 2021).
This process could result in the formation of hydroxylated products The intermediate products from the UV/O3 post-treatment of the R1
(m/z = 312.02) such as 4-hydroxyldiclofenac (IP-DCF1b) from benzene and R2 permeates included IP-DCF3a and IP-DCF10. According to IP-
rings with two chlorine atoms, 5-hydroxyldiclofenac (IP-DCF1a) from DCF3a (m/z = 310), it might be formed via the interaction between
benzene rings without two chlorine atoms and IP-DCF7 on nitrogen an aminyl radical and ozone. Both intermediate products could occur at
atom. A previous work by (Hong et al., 2020), who used UV/Fenton also the para-position relative to nitrogen on the aromatic ring that is more

15
T. Buakaew and C. Ratanatamskul Journal of Environmental Management 377 (2025) 124657

Fig. 10. Feasible intermediate products of SMX (a) and DCF (b) found in the effluents by UV/O3 degradation; of DI water (Black arrow), permeate of R1 (Blue arrow)
and permeate of R2 (Red arrow).

electron-rich due to the presence of acetate ion as an electron-donating DCF1b have been reported to exhibit logarithmic lethal concentration
group (Alharbi et al., 2022). Additionally, the occurrence of C-N 50% values (LC50) below 1 categorized as toxic to aquatic organisms
cleavage between the benzene ring, followed by ring-opening reactions, (Kolakovic et al., 2022). Whereas, IP-CIP3 was less toxic compared to
resulted in the formation of IP-DCF10 with m/z = 135.03. CIP as structural modifications to the quinolone ring (Tang et al., 2022).
As can be seen from Supplementary Data Tables S1, S2, S3 and S4, The results suggests that the integration of the AnBB-MBR with
after integrating the UV/O3 as the post treatment of the AnBB-MBR microaeration and UV/O3 is very effective for the significant removal of
systems (R1 and R2), it was found that several bio-intermediates of bio-intermediate compounds.
targeted pharmaceuticals were completely removed. This confirms that
the UV/O3 can be considered as the effective approach for removing bio- 4. Conclusion
intermediates of the targeted pharmaceuticals. However, IP-CIP3, IP-
CAF3, IP-DCF1a, IP-DCF1b, IP-DCF1c, IP-DCF2 were still found only in This study successfully developed the novel integrated system of
the UV/O3 effluent, treated for the R1 permeate. IP-DCF1a and IP- AnBB-MBR (with and without microaeration) and post-treatment of UV/

16
T. Buakaew and C. Ratanatamskul Journal of Environmental Management 377 (2025) 124657

O3 for the removal of CIP, CAF, SMX and DCF from building wastewater. contaminants in fresh produce: dietary exposure and human health concerns. Water
Res. 223, 118986. https://doi.org/10.1016/j.watres.2022.118986.
The UV/O3 showed superior degradation rates for CIP, CAF, SMX and
Benner, J., De Smet, D., Ho, A., Kerckhof, F.M., Vanhaecke, L., Heylen, K., Boon, N.,
DCF. AOB and methanotrophs in the AnBB-MBR system with microa­ 2015. Exploring methane-oxidizing communities for the co-metabolic degradation of
eration (R2) could facilitate co-metabolism of the targeted pharmaceu­ organic micropollutants. Appl. Microbiol. Biotechnol. 99, 3609–3618. https://doi.
ticals. While Bacillus, Longilinea, Clostridium and Lactivibrio were the org/10.1007/s00253-014-6226-1.
Bokulich, N.A., Kaehler, B.D., Rideout, J.R., Dillon, M., Bolyen, E., Knight, R., Huttley, G.
anaerobes that likely could remove targeted pharmaceuticals in the A., Gregory Caporaso, J., 2018. Optimizing taxonomic classification of marker-gene
AnBB-MBR (R1). It was observed that bio-intermediates generated by R1 amplicon sequences with QIIME 2’s q2-feature-classifier plugin. Microbiome 6, 90.
were more abundant compared to the case of R2 with microaeration. https://doi.org/10.1186/s40168-018-0470-z.
Boorgula, G.D., Gumbo, T., Singh, S., McShane, P.J., Philley, J.V., Srivastava, S., 2024.
This indicated that microaeration could improve the biodegradation and Omadacycline drug susceptibility testing for non-tuberculous mycobacteria using
removal efficiency of the targeted pharmaceuticals. The UV/O3 proved oxyrase to overcome challenges with drug degradation. Tuberculosis 147, 102519.
to be highly effective with 97–100% removal efficiency for the targeted https://doi.org/10.1016/j.tube.2024.102519.
Buakaew, T., Ratanatamskul, C., 2023. Effects of microaeration and sludge recirculation
pharmaceuticals from both anaerobic and microaerobic AnBB-MBR on VFA and nitrogen removal, membrane fouling reduction and microbial
systems. The UV/O3 system could effectively degrade a wide variety community of the anaerobic baffled biofilm-membrane bioreactor in treating
of bio-intermediates from the AnBB-MBR systems. Therefore, this novel building wastewater. Sci. Total Environ. 903, 166248. https://doi.org/10.1016/j.
scitotenv.2023.166248.
integrated system of AnBB-MBR with microaeration and UV/O3 can be Buakaew, T., Ratanatamskul, C., 2021. Effects of novel anaerobic baffled biofilm
considered as a promising technology for pharmaceutical removal from membrane bioreactor configurations on membrane fouling mitigation and microbial
building wastewater, offering a potential solution for mitigating the community in treating liquor condensate. Bioresour. Technol. 335, 125310. https://
doi.org/10.1016/j.biortech.2021.125310.
impact of pharmaceutical contaminants in the environment. Moreover,
Buakaew, T., Ratanatamskul, C., 2024. Unveiling the influence of microaeration and
future study on the toxicity reduction by this integrated system is sludge recirculation on enhancement of pharmaceutical removal and microbial
recommended. community change of the novel anaerobic baffled biofilm - membrane bioreactor in
treating building wastewater. Sci. Total Environ. 927, 172420. https://doi.org/
10.1016/j.scitotenv.2024.172420.
CRediT authorship contribution statement Carneiro, R.B., Mukaeda, C.M., Sabatini, C.A., Santos-Neto, A.J., Zaiat, M., 2020.
Influence of organic loading rate on ciprofloxacin and sulfamethoxazole
Tanissorn Buakaew: Writing – original draft, Formal analysis, Data biodegradation in anaerobic fixed bed biofilm reactors. J. Environ. Manag. 273,
111170. https://doi.org/10.1016/j.jenvman.2020.111170.
curation, Conceptualization. Chavalit Ratanatamskul: Writing – re­ Carneiro, R.B., Sabatini, C.A., Santos-Neto, A.J., Zaiat, M., 2019. Feasibility of anaerobic
view & editing, Writing – original draft, Supervision, Project adminis­ packed and structured-bed reactors for sulfamethoxazole and ciprofloxacin removal
tration, Methodology, Investigation, Conceptualization. from domestic sewage. Sci. Total Environ. 678, 419–429. https://doi.org/10.1016/j.
scitotenv.2019.04.437.
Carneiro, R.B., Sabatini, C.A., Titato, G.M., Ferreira, T.C.R., Zaiat, M., Santos-Neto, Á.J.,
Declaration of competing interest 2022. New biotransformation pathways from sulfamethoxazole and ciprofloxacin
removal in sewage treatment along the spatial profile of an anaerobic fixed bed
bioreactor. Bioresour. Technol. Rep. 17, 100944. https://doi.org/10.1016/j.
The authors declare that they have no known competing financial biteb.2022.100944.
interests or personal relationships that could have appeared to influence Challis, J.K., Hanson, M.L., Friesen, K.J., Wong, C.S., 2014. A critical assessment of the
photodegradation of pharmaceuticals in aquatic environments: defining our current
the work reported in this paper.
understanding and identifying knowledge gaps. Environmental Science: Process.
Impacts 16, 672–696. https://doi.org/10.1039/C3EM00615H.
Acknowledgements Chatila, S., Amparo, M.R., Carvalho, L.S., Penteado, E.D., Tomita, I.N., Santos-Neto, A.J.,
Lima Gomes, P.C., Zaiat, M., 2016. Sulfamethoxazole and ciprofloxacin removal
using a horizontal-flow anaerobic immobilized biomass reactor. Environ. Technol.
This research is funded by Thailand Science research and Innovation 37, 847–853. https://doi.org/10.1080/09593330.2015.1088072.
Fund Chulalongkorn University (DIS66210019) as well as the Ratchada- Chen, R., Jiang, H., Li, Y.-Y., 2018. Caffeine degradation by methanogenesis: efficiency
phiseksomphot Endowment Fund for Center of Excellence, Chula­ in anaerobic membrane bioreactor and analysis of kinetic behavior. Chem. Eng. J.
334, 444–452. https://doi.org/10.1016/j.cej.2017.10.052.
longkorn University. Chen, W.-H., Wong, Y.-T., Huang, T.-H., Chen, W.-H., Lin, J.-G., 2019. Removals of
pharmaceuticals in municipal wastewater using a staged anaerobic fluidized
Appendix. ASupplementary data membrane bioreactor. Int. Biodeterior. Biodegrad. 140, 29–36. https://doi.org/
10.1016/j.ibiod.2019.03.008.
Chen, W., Westerhoff, P., Leenheer, J.A., Booksh, K., 2003. Fluorescence
Supplementary data to this article can be found online at https://doi. Excitation− Emission matrix regional integration to quantify spectra for dissolved
org/10.1016/j.jenvman.2025.124657. organic matter. Environ. Sci. Technol. 37, 5701–5710. https://doi.org/10.1021/
es034354c.
Chiang, Y.-P., Liang, Y.-Y., Chang, C.-N., Chao, A.C., 2006. Differentiating ozone direct
Data availability and indirect reactions on decomposition of humic substances. Chemosphere 65,
2395–2400. https://doi.org/10.1016/j.chemosphere.2006.04.080.
Chin, A., Bérubé, P.R., 2005. Removal of disinfection by-product precursors with ozone-
Data will be made available on request. UV advanced oxidation process. Water Res. 39, 2136–2144. https://doi.org/
10.1016/j.watres.2005.03.021.
References Coble, P.G., 1996. Characterization of marine and terrestrial DOM in seawater using
excitation-emission matrix spectroscopy. Mar. Chem. 51, 325–346. https://doi.org/
10.1016/0304-4203(95)00062-3.
Akay, C., Ulrich, N., Rocha, U., Ding, C., Adrian, L., 2024. Sequential anaerobic–aerobic
Das, P.P., Dhara, S., Samanta, N.S., Purkait, M.K., 2024. Advancements on ozonation
treatment enhances sulfamethoxazole removal: from batch cultures to observations
process for wastewater treatment: a comprehensive review. Chemical Engineering
in a large-scale wastewater treatment plant. Environ. Sci. Technol. 58,
and Processing - Process Intensification 202, 109852. https://doi.org/10.1016/j.
12609–12620. https://doi.org/10.1021/acs.est.4c00368.
cep.2024.109852.
Alharbi, S.K., Ansari, A.J., Nghiem, L.D., Price, W.E., 2022. New transformation products
DelSontro, T., Perez, K.K., Sollberger, S., Wehrli, B., 2016. Methane dynamics
from ozonation and photolysis of diclofenac in the aqueous phase. Process Saf.
downstream of a temperate run-of-the-river reservoir. Limnol. Oceanogr. 61,
Environ. Prot. 157, 106–114. https://doi.org/10.1016/j.psep.2021.10.050.
S188–S203. https://doi.org/10.1002/lno.10387.
An, Z., Zhu, J., Zhang, M., Zhou, Y., Su, X., Lin, H., Sun, F., 2023. Anaerobic membrane
do Nascimento, J., Silva, E.V.A., Dos Santos, A.B., da Silva, M.E.R., Firmino, P.I.M., 2021.
bioreactor for the treatment of high-strength waste/wastewater: a critical review and
Microaeration improves the removal/biotransformation of organic micropollutants
update. Chem. Eng. J. 470, 144322. https://doi.org/10.1016/j.cej.2023.144322.
in anaerobic wastewater treatment systems. Environ. Res. 198, 111313. https://doi.
APHA, 2017. Standard Methods for the Examination of Water and Wastewater American
org/10.1016/j.envres.2021.111313.
Public Health Association., Washington, DC.
Frederic, O., Yves, P., 2014. Pharmaceuticals in hospital wastewater: their ecotoxicity
Aslam, A., Khan, S.J., Shahzad, H.M.A., 2022. Anaerobic membrane bioreactors
and contribution to the environmental hazard of the effluent. Chemosphere 115,
(AnMBRs) for municipal wastewater treatment-potential benefits, constraints, and
31–39. https://doi.org/10.1016/j.chemosphere.2014.01.016.
future perspectives: an updated review. Sci. Total Environ. 802, 149612. https://doi.
Geng, Y.-K., Gu, J., Zhang, X., Lim, Z.K., Jiang, Y., Zhang, M., Zhou, Y., Liu, Y., 2023.
org/10.1016/j.scitotenv.2021.149612.
Multi-parameter control-based operation strategy for mainstream deammonification
Ben Mordechay, E., Sinai, T., Berman, T., Dichtiar, R., Keinan-Boker, L., Tarchitzky, J.,
Maor, Y., Mordehay, V., Manor, O., Chefetz, B., 2022. Wastewater-derived organic

17
T. Buakaew and C. Ratanatamskul Journal of Environmental Management 377 (2025) 124657

in an integrated anaerobic biofilm reactor-step feed MBR. Chemosphere 333, Luo, X., Yan, Q., Wang, C., Luo, C., Zhou, N., Jian, C., 2015. Treatment of ammonia
138941. https://doi.org/10.1016/j.chemosphere.2023.138941. nitrogen wastewater in low concentration by two-stage ozonization. Int J Environ
Girardi, C., Greve, J., Lamshoft, M., Fetzer, I., Miltner, A., Schaffer, A., Kastner, M., 2011. Res Public Health 12, 11975–11987. https://doi.org/10.3390/ijerph120911975.
Biodegradation of ciprofloxacin in water and soil and its effects on the microbial Meng, F., Chae, S.R., Drews, A., Kraume, M., Shin, H.S., Yang, F., 2009. Recent advances
communities. J. Hazard Mater. 198, 22–30. https://doi.org/10.1016/j. in membrane bioreactors (MBRs): membrane fouling and membrane material. Water
jhazmat.2011.10.004. Res. 43, 1489–1512. https://doi.org/10.1016/j.watres.2008.12.044.
Gorito, A.M., Pesqueira, J.F.J.R., Moreira, N.F.F., Ribeiro, A.R., Pereira, M.F.R., Meng, J., Li, J., Li, J., Nan, J., Deng, K., Antwi, P., 2019. Effect of temperature on
Nunes, O.C., Almeida, C.M.R., Silva, A.M.T., 2021. Ozone-based water treatment nitrogen removal and biological mechanism in an up-flow microaerobic sludge
(O3, O3/UV, O3/H2O2) for removal of organic micropollutants, bacteria reactor treating wastewater rich in ammonium and lack in carbon source.
inactivation and regrowth prevention. J. Environ. Chem. Eng. 9. https://doi.org/ Chemosphere 216, 186–194. https://doi.org/10.1016/j.chemosphere.2018.10.132.
10.1016/j.jece.2021.105315. Mohapatra, B.R., Harris, N., Nordin, R., Mazumder, A., 2006. Purification and
Gu, Y., Huang, J., Zeng, G., Shi, L., Shi, Y., Yi, K., 2018. Fate of pharmaceuticals during characterization of a novel caffeine oxidase from Alcaligenes species. J. Biotechnol.
membrane bioreactor treatment: status and perspectives. Bioresour. Technol. 268, 125, 319–327. https://doi.org/10.1016/j.jbiotec.2006.03.018.
733–748. https://doi.org/10.1016/j.biortech.2018.08.029. Mori, K., Yamaguchi, K., Sakiyama, Y., Urabe, T., Suzuki, K.-i., 2009. Caldisericum exile
Hong, M., Wang, Y., Lu, G., 2020. UV-Fenton degradation of diclofenac, sulpiride, gen. nov., sp. nov., an anaerobic, thermophilic, filamentous bacterium of a novel
sulfamethoxazole and sulfisomidine: degradation mechanisms, transformation bacterial phylum, Caldiserica phyl. nov., originally called the candidate phylum
products, toxicity evolution and effect of real water matrix. Chemosphere 258, OP5, and description of Caldisericaceae fam. nov., Caldisericales ord. nov. and
127351. https://doi.org/10.1016/j.chemosphere.2020.127351. Caldisericia classis nov. Int. J. Syst. Evol. Microbiol. 59, 2894–2898. https://doi.org/
Hou, H., Mengting, Z., Duan, L., Zhao, Y., Zhang, Z., Yao, M., Zhou, B., Zhang, H., 10.1099/ijs.0.010033-0.
Hermanowicz, S.W., 2022. Removal performance and biodegradation mechanism of Nguyen, D., Khanal, S.K., 2018. A little breath of fresh air into an anaerobic system: how
sulfonamides antibiotic contained wastewater by IFAS-MBR bioreactor. J. Mol. Liq. microaeration facilitates anaerobic digestion process. Biotechnol. Adv. 36,
367, 120572. https://doi.org/10.1016/j.molliq.2022.120572. 1971–1983. https://doi.org/10.1016/j.biotechadv.2018.08.007.
Jenicek, P., Celis, C., Krayzelova, L., Anferova, N., Pokorna, D., 2014. Improving Nguyen, P.Y., Carvalho, G., Reis, M.A.M., Oehmen, A., 2021. A review of the
products of anaerobic sludge digestion by microaeration. Water Sci. Technol. 69, biotransformations of priority pharmaceuticals in biological wastewater treatment
803–809. processes. Water Res. 188, 116446. https://doi.org/10.1016/j.watres.2020.116446.
Jewell, K.S., Falås, P., Wick, A., Joss, A., Ternes, T.A., 2016. Transformation of diclofenac Nguyen, T.T., Bui, X.T., Dang, B.T., Ngo, H.H., Jahng, D., Fujioka, T., Chen, S.S., Dinh, Q.
in hybrid biofilm–activated sludge processes. Water Res. 105, 559–567. https://doi. T., Nguyen, C.N., Nguyen, P.T., 2019. Effect of ciprofloxacin dosages on the
org/10.1016/j.watres.2016.08.002. performance of sponge membrane bioreactor treating hospital wastewater.
Ji, J., Kakade, A., Yu, Z., Khan, A., Liu, P., Li, X., 2020. Anaerobic membrane bioreactors Bioresour. Technol. 273, 573–580. https://doi.org/10.1016/j.biortech.2018.11.058.
for treatment of emerging contaminants: a review. J. Environ. Manag. 270, 110913. Pan, L.-j., Li, J., Li, C.-x., Tang, X.-d., Yu, G.-w., Wang, Y., 2018. Study of ciprofloxacin
https://doi.org/10.1016/j.jenvman.2020.110913. biodegradation by a Thermus sp. isolated from pharmaceutical sludge. J. Hazard
Jia, Y., Khanal, S.K., Shu, H., Zhang, H., Chen, G.-H., Lu, H., 2018. Ciprofloxacin Mater. 343, 59–67. https://doi.org/10.1016/j.jhazmat.2017.09.009.
degradation in anaerobic sulfate-reducing bacteria (SRB) sludge system: mechanism Parshikov, I., Heinze, T., Moody, J., Freeman, J., Williams, A., Sutherland, J., 2001. The
and pathways. Water Res. 136, 64–74. https://doi.org/10.1016/j. fungus Pestalotiopsis guepini as a model for biotransformation of ciprofloxacin and
watres.2018.02.057. norfloxacin. Appl. Microbiol. Biotechnol. 56, 474–477. https://doi.org/10.1007/
Jia, Y., Khanal, S.K., Zhang, H., Chen, G.-H., Lu, H., 2017. Sulfamethoxazole degradation s002530100672.
in anaerobic sulfate-reducing bacteria sludge system. Water Res. 119, 12–20. Pashaei, R., Dzingelevičienė, R., Putna-Nimane, I., Overlinge, D., Błaszczyk, A.,
https://doi.org/10.1016/j.watres.2017.04.040. Walker, T.R., 2023. Acute toxicity of triclosan, caffeine, nanoplastics, microplastics,
Kim, I., Tanaka, H., 2010. Use of ozone-based processes for the removal of and their mixtures on Daphnia magna. Mar. Pollut. Bull. 192, 115113. https://doi.
pharmaceuticals detected in a wastewater treatment plant. Water Environ. Res. 82, org/10.1016/j.marpolbul.2023.115113.
294–301. https://doi.org/10.2175/106143009x12487095236630. Patel, S., Majumder, S.K., Ghosh, P., 2021. Ozonation of diclofenac in a laboratory scale
Kochetkova, T.V., Zayulina, K.S., Zhigarkov, V.S., Minaev, N.V., Chichkov, B.N., bubble column: intermediates, mechanism, and mass transfer study. J. Water Proc.
Novikov, A.A., Toshchakov, S.V., Elcheninov, A.G., Kublanov, I.V., 2020. Eng. 44, 102325. https://doi.org/10.1016/j.jwpe.2021.102325.
Tepidiforma bonchosmolovskayae gen. nov., sp. nov., a moderately thermophilic Paucar, N.E., Kim, I., Tanaka, H., Sato, C., 2019. Effect of O3 dose on the O3/UV
Chloroflexi bacterium from a Chukotka hot spring (Arctic, Russia), representing a treatment process for the removal of pharmaceuticals and personal care products in
novel class, Tepidiformia, which includes the previously uncultivated lineage secondary effluent. ChemEngineering 3.
OLB14. Int. J. Syst. Evol. Microbiol. 70, 1192–1202. https://doi.org/10.1099/ Poirier-Larabie, S., Segura, P.A., Gagnon, C., 2016. Degradation of the pharmaceuticals
ijsem.0.003902. diclofenac and sulfamethoxazole and their transformation products under controlled
Kolakovic, S., Salgado, R., Freitas, E.B., Bronze, M.R., Sekulic, M.T., Carvalho, G., environmental conditions. Sci. Total Environ. 557–558, 257–267. https://doi.org/
Reis, M.A.M., Oehmen, A., 2022. Diclofenac biotransformation in the enhanced 10.1016/j.scitotenv.2016.03.057.
biological phosphorus removal process. Sci. Total Environ. 806, 151232. https://doi. Poyatos, J.M., Muñio, M.M., Almecija, M.C., Torres, J.C., Hontoria, E., Osorio, F., 2009.
org/10.1016/j.scitotenv.2021.151232. Advanced oxidation processes for wastewater treatment: state of the art. Water Air
Koshikawa, T., Abe, M., Nagi, M., Miyazaki, Y., Takemura, H., 2022. Biofilm-formation Soil Pollut. 205, 187–204. https://doi.org/10.1007/s11270-009-0065-1.
capability depends on environmental oxygen concentrations in Candida species. Prakash, J., Arora, N.K., 2019. Phosphate-solubilizing Bacillus sp. enhances growth,
J. Infect. Chemother. 28, 643–650. https://doi.org/10.1016/j.jiac.2022.01.010. phosphorus uptake and oil yield of Mentha arvensis L. 3 Biotech. 9, 126. https://doi.
Koundle, P., Nirmalkar, N., Momotko, M., Makowiec, S., Boczkaj, G., 2024. Tetracycline org/10.1007/s13205-019-1660-5.
degradation for wastewater treatment based on ozone nanobubbles advanced Raj, R., Tripathi, A., Das, S., Ghangrekar, M.M., 2021. Removal of caffeine from
oxidation processes (AOPs) – Focus on nanobubbles formation, degradation kinetics, wastewater using electrochemical advanced oxidation process: a mini review. Case
mechanism and effects of water composition. Chem. Eng. J., 156236 https://doi. Studies in Chemical and Environmental Engineering 4, 100129. https://doi.org/
org/10.1016/j.cej.2024.156236. 10.1016/j.cscee.2021.100129.
Leonel, L.P., Bize, A., Mariadassou, M., Midoux, C., Schneider, J., Tonetti, A.L., 2022. Ratanatamskul, C., Katasomboon, J., 2017. Effect of sludge recirculation pattern on
Impacts of disinfected wastewater irrigation on soil characteristics, microbial biological nutrient removal by a prototype IT/BF-MBR (inclined tube/biofilm-
community composition, and crop yield. Blue-Green Systems 4, 247–271. https:// membrane bioreactor) and microbial population characteristics. Int. Biodeterior.
doi.org/10.2166/bgs.2022.126. Biodegrad. 124, 26–35. https://doi.org/10.1016/j.ibiod.2017.07.010.
Li, S., Peng, L., Yang, C., Song, S., Xu, Y., 2022. Cometabolic biodegradation of Ren, B., Shi, X., Jin, X., Wang, X.C., Jin, P., 2021. Comprehensive evaluation of
antibiotics by ammonia oxidizing microorganisms during wastewater treatment pharmaceuticals and personal care products (PPCPs) in urban sewers: degradation,
processes. J. Environ. Manag. 305, 114336. https://doi.org/10.1016/j. intermediate products and environmental risk. Chem. Eng. J. 404, 127024. https://
jenvman.2021.114336. doi.org/10.1016/j.cej.2020.127024.
Liu, C., Nanaboina, V., Korshin, G.V., Jiang, W., 2012. Spectroscopic study of Rosal, R., Rodriguez, A., Perdigon-Melon, J.A., Petre, A., Garcia-Calvo, E., Gomez, M.J.,
degradation products of ciprofloxacin, norfloxacin and lomefloxacin formed in Aguera, A., Fernandez-Alba, A.R., 2009. Degradation of caffeine and identification of
ozonated wastewater. Water Res. 46, 5235–5246. https://doi.org/10.1016/j. the transformation products generated by ozonation. Chemosphere 74, 825–831.
watres.2012.07.005. https://doi.org/10.1016/j.chemosphere.2008.10.010.
Liu, H., Gao, Y., Wang, J., Ma, D., Wang, Y., Gao, B., Yue, Q., Xu, X., 2021a. The Saidulu, D., Agrawal, S., Bhatnagar, A., Gupta, A.K., 2024. Sulfamethoxazole removal
application of UV/O3 process on ciprofloxacin wastewater containing high salinity: from wastewater via anoxic/oxic moving bed biofilm reactor: degradation pathways
performance and its degradation mechanism. Chemosphere 276, 130220. https:// and toxicity assessment. Bioresour. Technol. 392, 129998. https://doi.org/10.1016/
doi.org/10.1016/j.chemosphere.2021.130220. j.biortech.2023.129998.
Liu, T., Jia, G., Xu, J., He, X., Quan, X., 2021b. Simultaneous nitrification and Shi, H.P., Lee, C.M., 2007. Phosphate removal under denitrifying conditions by
denitrification in continuous flow MBBR with novel surface-modified carriers. Brachymonas sp. strain P12 and Paracoccus denitrificans PP15. Can. J. Microbiol.
Environ. Technol. 42, 3607–3617. 53, 727–737. https://doi.org/10.1139/w07-026.
Liu, X., Huang, Y., Su, X., Tian, S., Li, Y., Yuan, R., 2021c. Oxidation of sulfadiazine and Sinthuchai, D., Boontanon, S.K., Piyaviriyakul, P., Boontanon, N., Jindal, R.,
sulfamethoxazole through O3, UV, and UV/O3 processes. Desalination Water Treat. Polprasert, C., 2021. Fate and mass loading of antibiotics in hospital and domestic
222, 346–353. https://doi.org/10.5004/dwt.2021.27088. wastewater treatment plants in Bangkok, Thailand. J. Water, Sanit. Hyg. Dev. 11,
Liyanage, G.Y., Manage, P.M., 2018. Removal of Ciprofloxacin (CIP) by bacteria isolated 959–971. https://doi.org/10.2166/washdev.2021.092.
from hospital effluent water and identification of degradation pathways. Siri, Y., Bumyut, A., Precha, N., Sirikanchana, K., Haramoto, E., Makkaew, P., 2024.
International Journal Of Medical, Pharmacy And Drug Research 2, 37–47. https:// Multidrug antibiotic resistance in hospital wastewater as a reflection of antibiotic
doi.org/10.22161/ijmpd.2.3.1.

18
T. Buakaew and C. Ratanatamskul Journal of Environmental Management 377 (2025) 124657

prescription and infection cases. Sci. Total Environ. 908, 168453. https://doi.org/ Wei, M., Qiu, Q., Qian, Y., Cheng, L., Guo, A., 2016. Methane oxidation and response of
10.1016/j.scitotenv.2023.168453. Methylobacter/Methylosarcina methanotrophs in flooded rice soil amended with
Tamura, I., Yasuda, Y., Kagota, K.-i., Yoneda, S., Nakada, N., Kumar, V., Kameda, Y., urea. Appl. Soil Ecol. 101, 174–184. https://doi.org/10.1016/j.apsoil.2016.01.009.
Kimura, K., Tatarazako, N., Yamamoto, H., 2017. Contribution of pharmaceuticals Wu, G., Geng, J., Shi, Y., Wang, L., Xu, K., Ren, H., 2020. Comparison of diclofenac
and personal care products (PPCPs) to whole toxicity of water samples collected in transformation in enriched nitrifying sludge and heterotrophic sludge:
effluent-dominated urban streams. Ecotoxicol. Environ. Saf. 144, 338–350. https:// transformation rate, pathway, and role exploration. Water Res. 184, 116158.
doi.org/10.1016/j.ecoenv.2017.06.032. https://doi.org/10.1016/j.watres.2020.116158.
Tang, T., Liu, M., Du, Y., Chen, Y., 2022. Deciphering the internal mechanisms of Wu, Z., Zhang, L., Lin, H., Zhou, S., 2024. Enhanced removal of antibiotic resistance
ciprofloxacin affected anaerobic digestion, its degradation and detoxification genes during chicken manure composting after combined inoculation of Bacillus
mechanism. Sci. Total Environ. 842, 156718. https://doi.org/10.1016/j. subtilis with biochar. J. Environ. Sci. 135, 274–284. https://doi.org/10.1016/j.
scitotenv.2022.156718. jes.2022.12.002.
Tewari, S., Jindal, R., Kho, Y.L., Eo, S., Choi, K., 2013. Major pharmaceutical residues in Xu, D., Xie, Y., Li, J., 2022. Toxic effects and molecular mechanisms of sulfamethoxazole
wastewater treatment plants and receiving waters in Bangkok, Thailand, and on Scenedesmus obliquus. Ecotoxicol. Environ. Saf. 232, 113258. https://doi.org/
associated ecological risks. Chemosphere 91, 697–704. https://doi.org/10.1016/j. 10.1016/j.ecoenv.2022.113258.
chemosphere.2012.12.042. Yang, J., Duan, A., Wang, J., Yang, X., Liu, X., Xiao, F., Qin, F., Yu, Y., Wang, D., 2022.
Tiwari, B., Sellamuthu, B., Ouarda, Y., Drogui, P., Tyagi, R.D., Buelna, G., 2017. Review The fate of diclofenac in anaerobic fermentation of waste activated sludge.
on fate and mechanism of removal of pharmaceutical pollutants from wastewater J. Environ. Manag. 322, 116098. https://doi.org/10.1016/j.jenvman.2022.116098.
using biological approach. Bioresour. Technol. 224, 1–12. https://doi.org/10.1016/ Yang, L.-H., Qiao, B., Xu, Q.-M., Liu, S., Yuan, Y., Cheng, J.-S., 2021a. Biodegradation of
j.biortech.2016.11.042. sulfonamide antibiotics through the heterologous expression of laccases from
Tiwari, B., Sellamuthu, B., Piche-Choquette, S., Drogui, P., Tyagi, R.D., Vaudreuil, M.A., bacteria and investigation of their potential degradation pathways. J. Hazard Mater.
Sauve, S., Buelna, G., Dube, R., 2021. Acclimatization of microbial community of 416, 125815. https://doi.org/10.1016/j.jhazmat.2021.125815.
submerged membrane bioreactor treating hospital wastewater. Bioresour. Technol. Yang, S., Xiao, J., Liang, T., He, W., Tan, H., 2021b. Response of soil biological properties
319, 124223. https://doi.org/10.1016/j.biortech.2020.124223. and bacterial diversity to different levels of nitrogen application in sugarcane fields.
Triebskorn, R., Casper, H., Heyd, A., Eikemper, R., Kohler, H.R., Schwaiger, J., 2004. AMB Express 11, 172. https://doi.org/10.1186/s13568-021-01331-4.
Toxic effects of the non-steroidal anti-inflammatory drug diclofenac. Part II: Zhang, J., Lv, S., Yu, Q., Liu, C., Ma, J., Jia, M., Fang, S., 2023a. Degradation of
cytological effects in liver, kidney, gills and intestine of rainbow trout sulfamethoxazole in microbubble ozonation process: performance, reaction
(Oncorhynchus mykiss). Aquat. Toxicol. 68, 151–166. https://doi.org/10.1016/j. mechanism and toxicity assessment. Separ. Purif. Technol. 311, 123262. https://doi.
aquatox.2004.03.015. org/10.1016/j.seppur.2023.123262.
Wang, J., Zhang, C., Poursat, B.A.J., de Ridder, D., Smidt, H., van der Wal, A., Sutton, N. Zhang, L., Sun, F., Wu, D., Yan, W., Zhou, Y., 2020. Biological conversion of
B., 2022. Unravelling the contribution of nitrifying and methanotrophic bacteria to sulfamethoxazole in an autotrophic denitrification system. Water Res. 185, 116156.
micropollutant co-metabolism in rapid sand filters. J. Hazard Mater. 424, 127760. https://doi.org/10.1016/j.watres.2020.116156.
https://doi.org/10.1016/j.jhazmat.2021.127760. Zhang, Y., Guo, L., Hoffmann, M.R., 2023b. Ozone- and hydroxyl radical-mediated
Wang, K., Ma, H., Shen, Y., Shen, Y., Shu, J., Zeng, X., Liu, M., Wang, H., 2023. Magnetic oxidation of pharmaceutical compounds using Ni-doped Sb–SnO2 anodes:
biochar derived from peanut shells facilitates antibiotic degradation and fouling degradation kinetics and transformation products. ACS ES&T Engineering 3,
control in anaerobic membrane bioreactor for antibiotic pharmaceutical wastewater 335–348. https://doi.org/10.1021/acsestengg.2c00337.
treatment. J. Water Proc. Eng. 55, 104249. https://doi.org/10.1016/j. Zhao, F., Wang, Q., Shi, W., Shan, Y., Wang, X., Li, Q., Liu, D., Du, J., Wu, Y., 2024a.
jwpe.2023.104249. Global at-risk basins of socio-ecological impacts from water scarcity. Sci. Total
Wang, Y., Li, H., Yi, P., Zhang, H., 2019. Degradation of clofibric acid by UV, O3 and UV/ Environ., 176944 https://doi.org/10.1016/j.scitotenv.2024.176944.
O3 processes: performance comparison and degradation pathways. J. Hazard Mater. Zhao, X., Li, X., Pei, W., Li, X., Wei, Q., Liu, W., 2024b. Characterization, degradation
379, 120771. https://doi.org/10.1016/j.jhazmat.2019.120771. pathway and microbial community of aerobic granular sludge with ciprofloxacin at
Wardenier, N., Liu, Z., Nikiforov, A., Van Hulle, S.W.H., Leys, C., 2019. Micropollutant environmentally relevant concentrations. Biochem. Eng. J. 205, 109255. https://doi.
elimination by O3, UV and plasma-based AOPs: an evaluation of treatment and org/10.1016/j.bej.2024.109255.
energy costs. Chemosphere 234, 715–724. https://doi.org/10.1016/j. Zhu, J., Wang, H., Duan, A., Wang, Y., 2023. Mechanistic insight into the degradation of
chemosphere.2019.06.033. ciprofloxacin in water by hydroxyl radicals. J. Hazard Mater. 446, 130676. https://
doi.org/10.1016/j.jhazmat.2022.130676.

19