Water Research 261 (2024) 122065

Contents lists available at ScienceDirect

Water Research
journal homepage: www.elsevier.com/locate/watres

Activation of peracetic acid by electrodes using biogenic electrons: A novel

energy- and catalyst-free process to eliminate pharmaceuticals
Rusen Zou a , Wenqiang Yang b , Babak Rezaei c , Kai Tang a , Kuangxin Guo a , Pingping Zhang d ,
Stephan Sylvest Keller c , Henrik Rasmus Andersen a , Yifeng Zhang a, *
a
Department of Environmental & Ressource Engineering, Technical University of Denmark, DK-2800, Lyngby, Denmark
b
Department of Physics, Technical University of Denmark, Lyngby, DK 2800, Denmark
c
National Centre for Nano Fabrication and Characterization, DTU Nanolab, Technical University of Denmark, 2800 Kgs., Lyngby, Denmark
d
School of Environment and Architecture, University of Shanghai for Science and Technology, Shanghai, 200093, PR China

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

Keywords: Peracetic acid (PAA) has received increasing attention as an alternative oxidant for wastewater treatment.
Peracetic acid activation However, existing processes for PAA activation to generate reactive species typically require external energy
Bioelectrochemical system input (e.g., electrically and UV-mediated activation) or catalysts (e.g., Co2+), inevitably increasing treatment
Organic radicals
costs or introducing potential new contaminants that necessitate additional removal. In this work, we developed
Hydroxyl radicals
Pharmaceuticals removal
a catalyst-free, self-sustaining bioelectrochemical approach within a two-chamber bioelectrochemical system
(BES), where a cathode electrode in-situ activates PAA using renewable biogenic electrons generated by anodic
exoelectrogens (e.g., Geobacter) degrading biodegradable organic matter (e.g., acetic acid) in wastewater at the
anode. This innovative BES-PAA technique achieved 98 % and 81 % removal of 2 µM sulfamethoxazole (SMX) in
two hours at pH 2 (cation exchange membrane) and pH 6 (bipolar membrane) using 100 μM PAA without
external voltage. Mechanistic studies, including radical quenching, molecular probe validation, electron spin
resonance (ESR) experiments, and density functional theory (DFT) calculations, revealed that SMX degradation
was driven by reactive species generated via biogenic electron-mediated O–O cleavage of PAA, with CH3C(O)
OO• contributing 68.1 %, •OH of 18.4 %, and CH3C(O)O• of 9.4 %, where initial formation of •OH and CH3C(O)
O• rapidly reacts with PAA to produce CH3C(O)OO•. The presence of common water constituents such as anions
(e.g., Cl− , NO−3 , and H2PO−4 ) and humic acid (HA) significantly hinders SMX removal via the BES-PAA technique,
whereas CO2− 3 and HCO3 ions have a comparatively minor impact. Additionally, the study investigated the
−

removal of various pharmaceuticals present in secondary treated municipal wastewater, attributing differences
in removal efficiency to the selective action of CH3C(O)OO•. This research demonstrates a novel PAA activation
method that is ecologically benign, inexpensive, and capable of overcoming catalyst deactivation and secondary
pollution issues.

1. Introduction a low degradation rate (below 30 %) in typical WWTPs due to its

resistant properties (Mirzaei et al. 2021; Yang et al. 2017). Accordingly,
Pharmaceutical residues in urban wastewater pose significant chal­ the detection concentration of SMX in sewage treatment plant effluents
lenges for conventional wastewater treatment plants (WWTPs), which has been reported to range from 0.3 ng L− 1 to 783 ng L− 1 (da Silva
often fail to effectively remove these persistent compounds. Existing Rodrigues et al. 2020). Excessive antibiotic exposure can spread
guidelines rarely set threshold limits for such contaminants, leading to antibiotic-resistance genes, making these compounds emergent pollut­
their presence in surface, ground, and potable water (Fan et al., 2020; Lu ants of public concern (Luo et al. 2022). Therefore, it is crucial to
et al. 2021; Zou et al. 2021b). Prolonged exposure to these residues develop more robust, cost-effective, and environmentally friendly
threatens aquatic life and human health (Chaves et al. 2020; Pazda et al. technologies for removing these residues.
2019). For example, sulfamethoxazole (SMX), a common antibiotic, has Advanced oxidation processes (AOPs) are recognized for their

* Corresponding author at: Department of Environmental & Resource Engineering, Technical University of Denmark, Denmark.
E-mail address: [email protected] (Y. Zhang).

https://doi.org/10.1016/j.watres.2024.122065
Received 23 April 2024; Received in revised form 22 June 2024; Accepted 8 July 2024
Available online 8 July 2024
0043-1354/© 2024 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
R. Zou et al. Water Research 261 (2024) 122065

effective contaminant removal. Peracetic acid (PAA, CH3C(O)OOH), a conductive pili for long-range transfer (Wu et al. 2024). Our previous
colorless organic peroxy acid, is widely used as a disinfectant, bacteri­ study demonstrated that the electrons generated at the anode of BES
cide, and bleach (Ao et al. 2021; Henao et al. 2018). PAA is effective in could activate PS at the cathode forming •OH and •SO−4 to remove
microbial inactivation, offering potent bactericidal effects, low pH de­ contaminants (Zou et al. 2021a). Inspired by this, BES could be an ideal
pendency, easy storage and application, and minimal harmful source of electrical energy to activate the O–O bonds of PAA, elimi­
by-products in treated wastewater (Ao et al. 2021; Kim and Huang 2020; nating the need for external electrical energy to drive the electro­
Kim et al. 2019). Its high pKa (8.2) enables it to exist mainly as a neutral chemical activation process. To our knowledge, direct PAA activation at
species, which has a higher oxidizing capability than the anionic form the cathode of BES receiving electrons from the bio-anode has not been
(Cai et al. 2017; Kim et al. 2019). Additionally, PAA has a high redox reported.
potential (E0 = 1.06–1.96 V), making it a promising oxidizing agent In this work, BES was innovatively utilized as an energy- and
(Kim and Huang 2020; Wang et al. 2019). However, PAA can only catalyst-free bioelectronically activated PAA process for the efficient
directly oxidize a limited range of micropollutants with low efficiency removal of pharmaceuticals (e.g., SMX) from wastewater. The effect of
(Kim and Huang 2020). Consequently, there is interest in activating PAA operational factors including input voltage, pH, PAA dose, and co-
to produce highly reactive species such as hydroxyl radicals (•OH, (E0 = existing substances (Cl− , NO−3 , CO2− 3−
3 , HCO3 , H2PO4 , PO4 and humic
− −

1.80–2.72 V)) and organic radicals (R-O•, e.g., CH3C(O)OO•, (E0 = 2.24 acid (HA)) on PAA activation and SMX elimination was studied.
V) and CH3C(O)O•, (E0 = 1.60 V)) (Du et al. 2022; Kim et al. 2020; Meanwhile, the main reactive species, transformation products, and
Zhang et al. 2021). PAA-based AOPs have thus emerged as an exciting ecotoxicity of the treated effluent were systematically analyzed using
technique (Ao et al. 2021). Various methods for activating PAA include quenching experiments, chemical probes, electron spin resonance (ESR)
radiation (e.g., ultraviolet (UV), sunlight, and ultrasonic) (Rizzo et al. analysis, density functional theory (DFT) calculations, LC-MS detection,
2018; Zhang and Huang 2020; Zhao et al. 2021), heat (Wang et al. 2020) and the Toxicity Estimation Software Tool (T.E.S.T.), respectively.
and catalysts (transition metals like iron, cobalt and copper) (Kim et al. Lastly, the applicability of this novel system was further assessed with
2019; Wang et al. 2019; Wu et al. 2024), which is a critical step in the WWTP secondary effluent containing 21 model pharmaceuticals.
oxidation of aqueous micropollutants.
However, conventional UV lamps with mercury are often used to 2. Experimental
activate PAA but pose risks if ruptured (Nguyen et al. 2019). LED UV
lamps offer energy efficiency, extended lifespan, and design flexibility 2.1. Chemicals
but have higher initial costs and limited spectral wavelengths (Bertagna
Silva et al. 2021; Matafonova and Batoev 2018). Both traditional and Commercial peracetic acid stock solution (38–40 % w/w of peracetic
LED UV lamps consume energy throughout the treatment process. In acid, 5% w/w of H2O2 and 38–40 % w/w of acetic acid), sulfamethox­
addition, transition metal catalysts, like cobalt, require no external en­ azole (≥99 %), H2O2 (30% wt), sodium acetic acid (≥99 %), and sodium
ergy but may pose toxicity risks (Hu and Long 2016). Besides, solar sulfate (≥99 %) were purchased from Sigma-Aldrich (Denmark). The
activation of PAA is environmentally beneficial but less efficient than comprehensive information about other chemicals utilized in this work
UV activation due to lower photon flux (Ao et al. 2021). Furthermore, can be found in Text S1 in the Supplementary Information (SI).
thermal and ultrasonic activation are efficient but have high energy
consumption, limiting their application (Wang et al. 2012). Therefore,
2.2. BES installation and operation
there is an imperative to devise a novel, catalyst-free, climate-inde­
pendent, and cost-effective PAA activation method for micropollutant
A lab-scale dual-chamber BES reactor was built, with each chamber
eradication.
having an effective dimension of 8 cm in length, 5 cm in width and 5 cm
Electrochemical methods have been used to activate perox­
in height (Fig. S1). Details of the BES reactor setup and operation are
ymonosulfate (PMS) and persulfate (PS) due to their simplicity, minimal
described in Text S2.
secondary pollutants, and low energy consumption (Liu et al. 2019; Yao
et al. 2021). The O–O bond energy in PAA is 159 kJ mol-1, which is
lower than for PMS (317 kJ mol-1) and comparable to PS (92 kJ mol-1) 2.3. Analytical methods and calculations
(Duan et al. 2021; Wang et al. 2019). Electrochemical cleavage of the
O–O bond in PAA could be a feasible method for its activation. How­ The concentrations of SMX, its transformation products and other
ever, to the best of our knowledge, the performance and mechanisms of selected pharmaceuticals were analyzed quantitatively or qualitatively
the electrochemically activated PAA system for the degradation of by high-performance liquid chromatography (Agilent 1290 Infinity,
micropollutants such as pharmaceuticals have been scarcely investi­ USA, HPLC) with a tandem mass spectrometer (Agilent 6470 series,
gated. Recently, only Yang et al. reported an electrochemically activated USA, MS/MS). A detailed description of the parameters for HPLC-MS/
PAA system using a single-chamber reactor with mixed metal oxide MS is given in Text S3. The BES system’s voltage was measured using
electrodes for both the anode and cathode, focusing on the degradation a digital multimeter (Model 2700, Keithley Instruments, Inc., Cleveland,
of propranolol (Yang et al. 2024). Although they confirmed the feasi­ OH, USA). The detailed DFT calculations were described in Text S4. PAA
bility of generating radicals through PAA activation, the process still concentration was analyzed using the colorimetric method described in
suffers from high energy consumption (2.3–6.9 kWh m-3) (Yang et al. our earlier work (Chhetri et al. 2014). The inferred reactive species were
2024). Moreover, the transformation mechanisms of reactive species (e. identified through ESR analysis (Text S5).
g., CH3C(O)OO•) within the system remain inadequately explored.
Thus, understanding the comprehensive mechanism of electrochemical 2.4. Toxicity assays
activation of PAA and its potential for water treatment is crucial, and
exploring renewable energy sources can make this method both The microtox assays were conducted using a bioluminescence inhi­
economically and environmentally viable. bition experiment using Vibrio fischeri as a toxicity indicator as
Recent research has revealed that bioelectrochemical systems (BES) described in our earlier article (Zou et al. 2020b). Besides, the toxicity of
like microbial fuel cells (MFCs) can turn wastewater chemical energy SMX and its transformation products was evaluated using the Toxicity
into electricity or electrons for biosynthesis. Briefly, typical exoelec­ Estimation Software Tool (T.E.S.T.) (Long et al. 2023). The consensus
trogens such as Geobacter transmit electrons to the anode through method was utilized for prediction, covering Daphnia Magna LC50,
diverse mechanisms, including direct interaction with outer membrane bioaccumulation factor, developmental toxicity, and mutagenicity as
cytochromes, self-produced electron mediators (e.g., flavins), and forecasted toxicity endpoints.

2
R. Zou et al. Water Research 261 (2024) 122065

3. Results and discussion potentially present in the PAA solution (e.g., H2O2 and acetic acid)
didn’t contribute to the removal of SMX in the hybrid BES-PAA process.
3.1. PAA activation at the cathode and SMX elimination In addition, the circuit current and cathode potential recorded for the
different operating conditions described above are shown in Fig. S3.
A series of experiments were conducted to demonstrate the viability Notably, the circuit current of the hybrid BES-PAA process (6.82–13.12
of the BES for PAA activation and its application for removing the target mA) in the first two hours was significantly larger than that of the
micropollutant SMX. As shown in Fig. 1, almost no oxidative elimination BES-H2O2 (stable at 3.22 mA) or BES-acetic acid (stable at 3.16 mA)
of SMX was observed with over 2 h of reaction with either BES or PAA processes. The results confirm the capacity of PAA to accept electrons
seperately. PAA without electrochemical activation generally has a poor from the cathode, thereby enhancing the BES’s current flow. In partic­
oxidizing capacity for most pharmaceuticals and it is already known in ular, the decreasing current trend in the BES-PAA system also indicates
the literature it cannot remove SMX (Kim and Huang 2020). Moreover, the decrease in PAA concentration. Correspondingly, the change in the
the final product at the cathode of the BES reactor without PAA is H2O PAA concentration at the BES cathode is shown in Fig. S4. Additionally,
following the four-electron oxygen reduction (4e--ORR). Thus, it had no the steady output current and cathode potential of BES-H2O2 and
degradation effect on SMX. Moreover, these control experiments also BES-acetic acid processes demonstrate the system’s stability. Further­
show that the cathode chamber and the electrodes of the BES reactor more, the SMX in the cathodic solution in the open-circuit condition did
have no adsorption effect on SMX. As anticipated, the hybrid BES-PAA not show any degradation, which again indicates the role of the elec­
process demonstrated excellent SMX removal of nearly 98 % within 2 trons generated by the electrochemically active microorganisms in
h, indicating that the electrons generated by the electrochemically activating the PAA. Similar results have also been observed in studies on
active microorganisms at the anode can effectively activate the PAA at the direct or indirect activation of PS and periodate mediated by
the cathode creating reactive radicals. Moreover, the degradation of biogenic electrons to produce reactive species and subsequent removal
SMX in the hybrid BES-PAA process conformed to the pseudo-first-order of micropollutants, demonstrating the indispensable role of biogenic
kinetics with a rate constant calculated as 0.032 min-1. Since commercial electrons in activating these novel oxidants used in water treatment
PAA solutions usually include some H2O2 and acetic acid, we further processes (Zou et al. 2021a; Zou et al. 2024).
studied the impact of H2O2 and acetic acid at concentrations equal to the
experimentally dosed PAA on SMX elimination. The findings demon­
strated that SMX degradation in the BES-H2O2 and BES-acetic acid 3.2. Underlying mechanisms of the electrode-driven PAA activation and
processes was negligible. These results are reasonable because H2O2 SMX degradation
alone has a weak oxidation capacity and is difficult to be activated by
electrodes (Zhang et al. 2020). Therefore, the individual components 3.2.1. Identification of reactive species
Numerous earlier studies have demonstrated that the abatement of

1.0

BES
0.8
PAA only
BES+Acetic acid
0.6 BES+H2O2
Ct/C0

BES+PAA
0.4

0.2

0.0
0 20 40 60 80 100 120
Time (min)
Fig. 1. Removal of SMX through different treatment strategies. System operational parameters: initial SMX of 2 µM, PAA of 100 µM, Na2SO4 of 50 mM, without
applied voltage and pH of 2.

3
R. Zou et al. Water Research 261 (2024) 122065

micropollutants by PAA-based AOPs can be classified into two major that the major reaction route in the BES-PAA process was the radical
categories: radical and non-radical routes (Dong et al. 2024; Pan et al. route. Quenching studies using tert‑butanol (TBA) and methanol
2024; Shen et al. 2023). Among these, the radical route is characterized (MeOH) were further performed to identify the primary radical species
by the generation of reactive radicals like •OH and R-O• (Dai et al. responsible for SMX abatement. TBA was a frequent •OH quencher with
2022), whereas the non-radical route is characterized mainly by the a high reaction rate constant (k = 6.0 × 108 M-1 s-1) and less reactive
generation of singlet oxygen (1O2) or single-electron transfer (Dai et al. with R-O• (Lin et al. 2021). In addition, the reaction rate constant for
2022). SMX with •OH was about 7.3 × 109 M-1 s-1 (Gao et al. 2020). As a result,
The radical route was explored thoroughly to comprehend the re­ 100 mM TBA was adequate to quench the reaction between SMX at an
action mechanisms in the BES-PAA process. First, a common free radical initial concentration of 2 µM and the possibility formed of •OH. As seen
scavenger, ascorbic acid (AA), was employed to assess whether the BES- in Fig. 2b, dosing 100 mM TBA had a substantial inhibitory impact on
PAA process created free radicals. As shown in Fig. 2a, AA substantially the abatement of SMX (decreased from 98 % to 69.1 %), demonstrating
inhibited SMX degradation in the BES-PAA process. Specifically, that •OH was one of the primarily responsible radicals. The
increasing the AA concentration from 0 to 1 mM reduced the SMX coumarin-fluorescence technique was further used to verify the pro­
degradation rate from 98 % (without AA) to 10 % after 2 h, suggesting duction of •OH (Li et al. 2022a). The •OH-fluorescence intensity can be

Fig. 2. Effect of AA (a), FFA, TBA, MeOH and BQ (b), anaerobic condition (c), on SMX removal. ESR spectrum obtained by spin-trap with DMPO (d), with DIPPMPO
(e) in the BES-PAA system at varying reaction times and contribution of main reactive species to SMX removal (f) in the BES-PAA system. System operational
parameters: initial SMX of 2 µM, PAA of 100 µM, Na2SO4 of 50 mM, without applied voltage and pH of 2.

4
R. Zou et al. Water Research 261 (2024) 122065

another strong evidence for the major role of •OH on the removal of after 2 h of the reaction (Fig. S6). The results signified the presence of
SMX during the BES-PAA process (Fig. S5). MeOH has been used CH3C(O)OO•. In addition, by adding 1.0 mM of Mn2+ and 100 mM TBA,
extensively as a potent R-O• (k value lacks literature reports) and •OH (k the removal efficiency of SMX was 14.8 %, which indicated the presence
= 9.7 × 108 M–1 s–1) scavenger to further the contribution of the possible of CH3C(O)O•.
presence of R-O• to SMX elimination (Wang et al. 2019). In the presence The prior PAA-based processes showed that superoxide radicals
of 100 mM MeOH, the BES-PAA process was shown to be ineffective for (•O−2 ) might be formed through Eq.(10) (Wu et al. 2022). Thus, we
the removal of SMX, exhibiting only about 3 % removal, which indicates introduced benzoquinone (BQ, •O−2 scavenger, k = 2.3 × 108 M-1 s-1) to
that not only •OH but also R-O• was the major free radical generated the BES-PAA process to further confirm the presence of •O−2 . As shown in
(Fig. 2b). Further, considering the results of using 100 mM TBA alone to Fig. 2b, BQ exhibited only a negligible suppression of SMX degradation,
quench •OH, it becomes evident that R-O• contributes more signifi­ demonstrating that •O−2 was not the principal active radical responsible
cantly to the removal of SMX by the BES-PAA process compared to •OH. for SMX degradation. The •O−2 generation was also tested using a probe
Roughly calculated, the combined contribution of •OH and R-O• to SMX (nitrotetrazolium blue chloride, NBT). Normally, colorless NBT com­
removal is about 96 % based on the calculation methods in Text S6. bines with •O−2 to form blue products, but no such alterations were
Therefore, combined with the previous reports that reactive radicals observed in this study (Fig. S7), again demonstrating that •O−2 was not
existed mainly as •OH and R-O• in the PAA-based AOPs, we inferred the the primary radical. Based on the results, it is possible to conclude that
mechanism of PAA activation in terms of the generation of reactive •OH and R-O• (mainly as CH3C(O)OO• and CH3C(O)O•), rather than
radicals in the BES-PAA process (Eq. (1)–(13)) (Li et al. 2022b; Luuk­ •O−2 , were the primary active radicals responsible for the degradation of
konen and Pehkonen 2017; Wang et al. 2020). SMX in this study.
The possible non-radical route (single-electron transfer and 1O2) in
e− + CH3C(O)OOH → CH3C(O)O− + •OH (1) the BES-PAA process was further investigated. The results suggest that
e− + CH3C(O)OOH → CH3C(O)O• + OH− (2) SMX was degraded through a radical pathway rather than a non-radical
pathway in the BES-PAA process (Figs. 2b and S8, Discussion S1).
CH3C(O)OOH + •OH → CH3CO• + H2O + O2 (3) Subsequently, ESR analysis was undertaken as compelling evidence
to assess further the reactive species produced in the BES-PAA system.
CH3C(O)OOH + •OH → CH3C(O)OO• + H2O (4)
Firstly, a distinct DMPO–CH3C(O)OO• six-peak signal was observed,
CH3C(O)OOH + •OH → CH3C(O)OH + HO2• (5) significantly supporting CH3C(O)OO• production (Fig. 2d) (Zhang et al.
2023). Secondly, a multi-peak composite signal was identified in the
CH3C(O)OOH + CH3C(O)O• → CH3C(O)OO• + CH3C(O)OH (6) complex spectrum depicted in Fig. S9, which indicated that this signal
CH3C(O)O• → CH3• + CO2 (7) was composed of numerous radicals including CH3C(O)O•, CH3CO•,
•OH and CH3•. Moreover, Fig. 2e and Table S4 provided plots and fitted
CH3• + O2 → CH3(O)O• (8) data from further spectral simulations for identifying each radical. The
simulations’ capture products include hyperfine splitting constants (aP,
CH3C(O)OO• + CH3C(O)OO• → 2 CH3C(O)O• + O2 (9)
aN, and aH), which offer compelling evidence regarding the existence of
e− + O2 → •O−2 (10) CH3C(O)O•, CH3CO•, •OH and CH3• (Table S4). Besides, similar to
quenching and chemical probe tests, ESR measurements showed no
1
•O−2 + 2H2O → O2 + 2OH + H2O2
−
(11) significant peaks, confirming the lack of •O−2 and 1O2 in the BES-PAA
1 system (Figs. S10 and S11). Based on the quenching, molecular probe
CH3C(O)OOH + CH3C(O)OO → CH3C(O)O + CH3C(O)OH + O2(12)
− −
validation, ESR experiments, and the calculation methods in Text S6, the
CH3CO• + O2 → CH3C(O)OO• (13) involvement of •OH in eliminating SMX within the BES-PAA process
approximately accounted for 18.4 %, while the participation of R-O•
It can be seen that R-O• contains a variety of radicals such as CH3C constituted 77.5 % (with 68.1 % from CH3C(O)OO• and 9.4 % from
(O)OO•, CH3C(O)O•, CH3CO•, CH3(O)O• and CH3•. As shown by Eq. CH3C(O)O•). In addition, it was calculated that CH3C(O)OO• was con­
(8), CH3• reacts fast with O2 to generate CH3(O)O• (k = 2.8–4.1 × 109 verted from approximately 75 % of •OH and 25 % of CH3C(O)O•
M-1 s-1), which has a substantially lower oxidation capacity than CH3C through formulas (4) and (6), respectively, and contributed 51.1 % and
(O)OO• and CH3C(O)O• (Cai et al. 2017; Li et al. 2022b). To further 17 % to the removal of SMX, respectively (Fig. 2f).
determine the contribution of CH3• and CH3(O)O• to SMX degradation, Therefore, it was concluded that the activation mechanism of PAA in
we evaluated the performance of the BES-PAA process under anaerobic the BES-PAA process was mainly through a radical route. Moreover, the
conditions. As shown in Fig. 2c, anaerobic conditions displayed no in­ identified major reaction pathways and the reaction rate constants in the
fluence on SMX degradation in the presence or absence of TBA, showing BES-PAA process were summarized in Table S5, where the most relevant
that the contribution of CH3• and CH3(O)O• to SMX degradation was ones are highlighted.
negligible. Similarly, CH3CO• could then rapidly combine with dis­
solved oxygen to form CH3C(O)OO• via Eq.(13) (k = 1.2 × 109 M–1 s–1) 3.2.2. Theoretical calculations
(Zheng et al. 2021). In addition, previous literature reports also The mechanism of PAA activation in the BES-PAA process was
demonstrated that CH3CO• contributed much less to SMX removal than further elucidated using DFT calculations. As shown in Fig. 3, we can
CH3C(O)OO• and CH3C(O)O• (Xu et al. 2024; Yang et al. 2024). find that the energy of the system of PAA itself keeps rising without the
Therefore, it could be postulated that CH3C(O)OO• and CH3C(O)O• involvement of biogenic electrons, and the energy of the whole reaction
were the key reactive R-O• contributing to the degradation of SMX in the increases by a total of 1.00 eV, indicating that the reaction cannot take
BES-PAA process. Notably, due to the lower activation energy barrier for place. In contrast, the bioelectronic involvement of PAA’s activation is
PAA’s hydrogen abstraction, once CH3C(O)O• and •OH were formed, it an exothermic process; thus, it is thermodynamically capable of pro­
became relatively easy to generate CH3C(O)OO• via Eq.(7) and (9) (Cai ceeding spontaneously. In the context of the BES-PAA process, we
et al. 2017; Zhang and Huang 2020). Moreover, CH3C(O)OO• was the couldn’t find a computationally stable PAA structure with both a
most powerful oxidizing peroxyl radicals among the R-O• (He et al. negative charge and an electron. Our calculations indicate that the
2022). Recently, it has been reported that Mn2+ could be used as a negatively charged PAA will promptly disintegrate when exposed to
specific quencher for CH3C(O)OO• via Eq.(10) (Zhang et al. 2022). biogenic electrons, resulting in either CH3C(O)O− and •OH or CH3C(O)
Introducing 1.0 mM of Mn2+ into the BES-PAA system partially hindered O• and OH− . Moreover, it was further found that the difficulty of bio­
the elimination of SMX, reducing removal efficacy from 98 % to 49.2 % electronic activation of PAA to generate CH3C(O)O• and •OH was

5
R. Zou et al. Water Research 261 (2024) 122065

Fig. 3. The free energy changes along the reaction coordinates during the BES-PAA process.

different, and the generation of •OH was more favored in thermody­ electron-mediated activation of PAA. Subsequently, •OH and CH3C(O)
namics compared with the generation of CH3C(O)O•, with the system O• further react with the PAA parent compound to form CH3C(O)OO•
energy decreasing by 4.62 and 5.12 eV, respectively. Notably, the DFT through Eq. (4) and (6). CH3C(O)OO• is the primary active species
calculations were mainly used to theoretically validate the feasibility of involved in the subsequent degradation of SMX, contributing up to 68.1
the bioelectronic activation of PAA, and as for the generation of CH3C %.
(O)O• and •OH to further react with the parent PAA to generate other CH3C(O)OO• and CH3C(O)O•, collectively referred to as R-O•, were
reactive species like CH3C(O)OO• via Eq. (3)–(13) in the manuscript shown to play the most crucial role in the removal of SMX, followed by
were not further calculated. Instead, their existence and characteristics •OH. In contrast, the CH3CO•, CH3•, •O−2 , 1O2 and CH3(O)O• generated
were verified through a series of quenching experiments and ESR ex­ through Eq (3), (5), (7), (8), (10)–(13) are negligible. Besides, PAA as a
periments. In conclusion, the DFT calculations agreed well with the cathodic electron acceptor can significantly increase the current of the
results of the radical probing assays, confirming the significance of the BES system, thus facilitating the removal of organic matter from the
radical pathway (mainly as •OH and R-O•) in the BES-PAA process. domestic wastewater by the bioanode. The results showed that adding
100 µM PAA to the cathodic chamber resulted in 67 % COD removal in
3.2.3. Mechanisms underlying the SMX degradation the anodic chamber over six operating cycles (12 h), respectively,
Firstly, the potential transformation products (TPs) and accordingly compared with no PAA addition (32 %) (Fig. S16)
pathways for SMX degradation were investigated. Briefly, the reactive
species (R-O• and •OH) could attack various sites in the SMX molecule,
such as the S-N bond, S-C bond, amine group on the aromatic ring, and 3.3. Key limiting factors to the BES-PAA process
benzene ring, resulting in 15 major TPs that can be attributed to four
different degradation pathways including the cleavage of S-N (pathway 3.3.1. Circuit current
1) and S-C bond (pathway 2), amino oxidation (pathway 3) and hy­ It was demonstrated earlier in this study that electrons play a crucial
droxylation (pathway 4) (Fig. 4a, Discussion S2). Besides, the toxicity of role in the activation of PAA in the BES-PAA process. Thus, the circuit
SMX and its transformation products was further calculated, showing current in the BES-PAA system could significantly impact the production
that the BES-PAA process could effectively reduce the overall toxicity of of free radicals, which could subsequently influence the elimination of
SMX (Fig. 4b–e, Discussion S3). Moreover, an evaluation of the BES-PAA SMX. The circuit current in the BES reactor can be varied by operating in
process’s capability to eliminate the total organic carbon (TOC) was MFC (without input voltage) and microbial electrolysis cell (MEC)
conducted, as detailed in Text S7. It was observed that over 41.5 % of the (input voltage: 0.2–1.0 V) modes. Consequently, the BES-PAA process
TOC was effectively eliminated during a 2-hour reaction, as depicted in was studied at various applied voltages (0, 0.2, 0.4, 0.6, 0.8 and 1.0 V),
Fig. S15. corresponding to different circuit currents. Surprisingly, the removal
Scheme 1 summarizes the suggested overall mechanism of SMX efficiency for the SMX did not increase with the applied voltage but
degradation by the BES-PAA process based on the experimental results. tended to decrease (Fig. 5a). Fig. S17 shows that the recorded system
First, the organic matter, such as acetic acid derived from domestic current was increasing for a higher input voltage. The observed decrease
wastewater as the carbon and energy source (electron donor) was con­ in removal efficiency as the voltage was elevated could be explained by
verted into e− , H+, and CO2 by the anodic electroactive microbes (e.g., side reactions (e.g., Eq. (14)-(18)) that may be initiated when the
Geobacter) attached to the carbon brush in the anode chamber (Zou applied voltage rises, resulting in a decrease in the observed formation
et al. 2021a). Then, the resultant e− went through the wire to the Ti of •OH and R-O• (CH3C(O)OO• and CH3C(O)O•). A similar phenome­
mesh cathode, where it acted on the O–O bond of PAA on the Ti mesh non was found previously in systems with bioelectronically activated
electrode surface to achieve the activation of PAA and generate reactive PMS (Li et al. 2019).
radicals, including CH3C(O)O• and •OH via Eqs. (1) and (2). Notably, e− + CH3C(O)O• → CH3C(O)O− (14)
through DFT theoretical calculations and experimental verification, the
cleavage of the O–O bond more easily forms •OH rather than CH3C(O) e− + CH3C(O)OO•→ CH3C(O)OO− (15)
O•, with a ratio of approximately 3:1 during the biogenic
e + •OH + H → H2O
− +
(16)

6
R. Zou et al. Water Research 261 (2024) 122065

Fig. 4. (a) Proposed transformation pathway of SMX during the BES-PAA process, (b) daphnia Magna LC50 (48 h), (c) bioaccumulation factor, (d) developmental
toxicity, and (e) mutagenicity of SMX and transformation byproducts in the BES-PAA process.

7
R. Zou et al. Water Research 261 (2024) 122065

Scheme 1. Proposed mechanism of activating PAA by electrodes using biogenic electrons and subsequent SMX removal by the BES-PAA process.

oxidants, and model compounds (Ao et al. 2021). The impact of various
e− + CH3C(O)OO• + H+ → CH3C(O)OOH (17) pH levels (2, 3, 5, 7, 9 and 11) on SMX elimination was investigated
further to assess the BES-PAA process thoroughly. The results reveal that
e− + CH3C(O)O• + H+ → CH3C(O)OH (18a) the pH has a considerable impact on the elimination of SMX with the
The trend of •OH-fluorescence intensity at various voltages was BES-PAA process (Fig. 5c). In particular, the optimal removal efficiency
further investigated, and the findings (Fig. S18) confirmed that the was obtained when the initial pH was 2. The removal of SMX was
amount of •OH decreased at higher voltages, affecting the elimination of drastically reduced when the initial pH was 3 with the efficiency
SMX, supporting our prior interpretation. decreasing from 98 % to 62 %. Moreover, when the starting pH was
more than or equal to 5, the BES-PAA process proved inefficient with
3.3.2. PAA dosage SMX removal efficiencies corresponding to 15.4 %, 17.2 %, 16.9 % and
In the BES-PAA process, •OH and R-O• were produced primarily due 11.8 % for pH 5, 7, 9 and 11, respectively). In addition, the change of
to PAA activation. Consequently, the PAA dosage is crucial in deter­ catholyte pH with time is shown in Fig. 5d. When the initial pH was 2 the
mining the efficacy of the BES-PAA process for micropollutant abate­ pH fluctuations throughout the reaction were limited, while when the
ment. Accordingly, the effect of various PAA concentrations (25, 50, initial pH was 3 or higher, it increased rapidly to more than 10. Spe­
100, 200, 500 and 1000 µM) on system performance was explored cifically, as the reaction progresses, the 4e− ORR occurring via Eq. (19)
further and the results are summarized in Fig. 5b. Specifically, when the at the cathode leads to a continuous increase in pH, rapidly rising from
PAA dosage was raised from 25 to 100 µM, the removal efficiency of the an initial pH of 3 to over 10 within 60 min. Moreover, we also observed
BES-PAA process for SMX increased constantly from 62 % to 98 %. in Fig. 3c that at an initial pH of 3, the SMX removal efficiency in the first
Clearly, the increase in •OH and R-O• generation via Eqs. (1) and (2) 30 min was slightly improved compared to that at pH 2, after which the
was responsible for the increase in degradation efficiency. However, efficiency decreased sharply to zero. The specific reasons are divided
when the concentration of PAA was further increased to 200, 500 or into the following aspects: 1) When the pH is higher than 8.2, PAA
even 1000 μM, the degradation of SMX was significantly inhibited. The mainly exists in deprotonation (PAA− ). For example, when the pH was 9
inhibition became more obvious with the increase of PAA dosing where and 11, the proportion of PAA− reached 86 % and 99.6 %, respectively,
the removal efficiency decreased from 98 % for 100 µM to 70 %, 64 % and the activation of the PAA− needed for the participation of the H+; 2)
and 37 %, respectively. when the pH exceeded 5.5, PAA began to decompose spontaneously, and
The reduction in SMX removal efficiency caused by an excessive dose at pH levels above 8.2, it further underwent hydrolysis, with the rate of
of PAA may result from the competition of PAA and SMX for the •OH hydrolysis accelerating as pH increased, resulting in a rapid reduction in
and R-O•. Excess PAA apparently eliminated •OH and R-O• via Eq. (4) PAA concentration (Eqs. (19)–(21)) (Ao et al. 2021; Wang et al. 2019;
and (6), thus hindering the breakdown of SMX (Li et al. 2022b; Luuk­ Yuan et al. 1997); 3) the stronger acidic condition (pH 2) could stabilize
konen and Pehkonen 2017; Wang et al. 2020). Furthermore, the mea­ the PAA, resulting in less catalytic activation of the PAA and less pro­
surements showed that the PAA dosage significantly affected the system duction of the reactive radicals compared to that at pH 3 (Wang et al.
currents and positively correlated with the dosage (Fig. S19). Therefore, 2021; Wu et al. 2020). In summary, high SMX removal efficiency could
the reduction of SMX removal in the presence of excess PAA may also still be obtained at a slightly acidic to neutral cathodic pH.
result in a decrease of free radicals due to side reactions (Eq. (14)-(18)) Thus, we proposed and tested the following two strategies to address
caused by a further increase of current as we described in the previous the pH issue for future practical applications, including the use of a bi­
section (Li et al. 2019). polar membrane (BPM) instead of the existing CEM or the use of an
Overall, the findings suggest that small amounts of PAA were additional pH control system. The BES-PAA process demonstrated its
appreciated to induce efficient SMX elimination (98 % decrease at 100 capability to effectively eliminate SMX not only under acidic conditions
µM PAA). The dosage of 100 µM PAA identified as optimal for the BES- but also under near-neutral conditions. As depicted in Fig. 5c, the effi­
PAA process was similar to the concentrations (approx. 7–132 µM) used ciency of SMX removal via the BES-PAA process was notably improved
for wastewater purification (Luukkonen and Pehkonen 2017). at pH 6, particularly when employing the BPM or a controlled pH control
system (81.1 % and 98.7 % for a 2-hour duration). Compared to the pH
3.3.3. Catholyte pH control system, the relatively lower performance of SMX removal with
In the PAA-based AOPs, the catholyte pH is a crucial parameter that the BPM can be primarily attributed to a significant reduction in current
may significantly influence the characteristics of reactive radicals, (see Fig. 5e), which in turn may lead to a slower PAA activation rate and
the production of relatively fewer reactive radicals, including •OH and

8
R. Zou et al. Water Research 261 (2024) 122065

Fig. 5. Effect of circuit current (a), PAA dose (b), pH (c), and the impact of pH on system performance (pH variation (d) and system current (e)) during the BES-PAA
process on SMX removal. System operational parameters: initial SMX of 2 µM and Na2SO4 of 50 mM.

CH3C(O)OO− + H2O2 → CH3C(O)O− + H2O + O2 (21)

R-O•. Of course, the advantage of using BPM is no need for additional
acid-base solutions. As illustrated in Fig. 5d, the pH of the BES-PAA To conclude, controlling the pH of the catholyte is crucial when using
system with BPM remained relatively stable (from 6 to 6.4) over 2 h. the BES-PAA technique. Additionally, Table S8 provides an overview of
Furthermore, when the pH was controlled at 6, the BES-PAA process commonly employed PAA-based AOPs for SMX removal. While BES-PAA
exhibited a slightly higher SMX removal efficiency than pH 3. The result showed comparable or even higher SMX removal efficiency than pre­
was primarily attributed to the fact that, at pH 6, SMX (pKa 1.8 and 5.6) vious methods, its removal rate was slower than other methods except
predominantly existed in its deprotonated form (71.3 % of SMX− ), for sunlight activation. Nonetheless, the previously reported methods
which exhibited a stronger reactivity with the reactive radicals, generally necessitated substantial quantities of catalysts (ranging from
including •OH and R-O• compared to SMX at pH 3 (Li et al. 2021; Zhu tens to hundreds of milligrams per liter) in the wastewater, significant
et al. 2020). energy consumption, or susceptibility to weather conditions and the
color of the wastewater.
O2 + 4H+ + 4e− → 2H2O (18b)

CH3C(O)OO + CH3C(O)OOH → CH3C(O)OH + CH3C(O)O + O2 (19)

− − 3.3.4. Water constituents
The capacity of typical PAA-based processes to degrade micro­
CH3C(O)OOH + H2O → CH3C(O)OH + H2O2 (20) pollutants in water has been demonstrated to be impacted by water
constituents, including anions and natural organic matter (NOM) found

9
R. Zou et al. Water Research 261 (2024) 122065

in water (Ao et al. 2021). Thus, the effects of several common anions (e. H2PO−4 , and HA severely hindering the process, while CO2−
3 and HCO3
−

g., Cl− , NO−3 , and PO3− 4 ) and HA on the removal of SMX with the have a relatively minor impact (Fig. 6, Discussion S4). Similarly, the
BES-PAA process were studied. Therefore, the effects of several common BES-PAA process using CEM also exhibited similar phenomena (the
anions (e.g., Cl− , NO−3 , CO2− − − 3−
3 , HCO3 , H2PO4 and PO4 ) and HA on the impact of CO2−3 and HCO3 was not explored due to the initial pH being
−

removal of SMX by the BES-PAA process (using BPM and CEM, respec­ 2) (Fig. S20, Discussion S5).
tively) were evaluated. Observations indicate that all selected water
constituents negatively affect the removal of SMX via bio­
electrochemical activation of PAA (using BPM at pH 6), with Cl− , NO−3 ,

Fig. 6. Effects of Cl− (a), NO−3 (b), CO−3 (c), HCO−3 (d), H2PO−4 (e) and HA concentration (f) on the removal of SMX during the BES-PAA process using BPM under
nearly neutral conditions. System operational parameters: initial SMX of 2 µM, PAA of 100 µM, Na2SO4 of 50 mM, and pH of 6.

10
R. Zou et al. Water Research 261 (2024) 122065

3.4. Environmental application of BES-PAA technique pharmaceutical was approximately 10–20 % lower compared to acidic
conditions, primarily due to the relatively lower system current as the
To further verify the feasibility of practical applications, we further results of the potential loss over the BPM (Fig.S21) (Zou et al. 2022). The
tested the residual recalcitrant micropollutants found in the secondary observed disparities in removal efficiencies of the 21 chosen pharma­
effluent of WWTPs. Therefore, 21 pharmaceuticals with the highest ceuticals are likely attributable to the significant selectivity of the
detection frequency in the secondary effluent of EU WWTPs were dominant reactive radicals R-O• produced in the BES-PAA process (Lin
selected and spiked into WWTP secondary effluent (Lundtofte WWTP, et al. 2021; Zong et al. 2021). It is noteworthy that despite the variation
Lyngby, Denmark). In addition, the physical properties of the 21 phar­ in removal efficiency, the lowest removal efficiency was still close to 50
maceuticals are shown in Table S3. Note that we investigated the %, which was much higher than the efficiency of other previously re­
removal of 21 pharmaceuticals by the BES-PAA process under acidic ported PAA-based processes like Co3O4/PAA and hydroxylamin/Fe
(using CEM, pH was adjusted to 2) and unadjusted initial pH (using (II)/PAA (Lin et al. 2021; Wu et al. 2020). For example, the removal of
BPM, pH was 7.2 after adding 100 µM PAA), respectively. The operating carbamazepine was generally below 10 % in such processes. This was
conditions with the actual water samples were consistent with the because in these R-O•, a radical with low selectivity for pharmaceuticals
optimal conditions obtained with the laboratory experiments except that such as carbamazepine, is the main active substance without the pres­
50 mM Na2SO4 was not added as a supporting electrolyte. As shown in ence of the broad-spectrum radical •OH (Cl•, k•OH/carbamazepine = 8.8 ×
Fig. 7, the BES-PAA process demonstrated exceptional removal of 109 M-1 s-1) (Lin et al. 2021; Wu et al. 2020). These findings again
mycophenolic acid, propranolol, sertraline, 5-methyl-1-h-benzotriazole, confirm our earlier conclusion (Section 3.2) that in the BES-PAA process,
citalopram, venlafaxine, 1-h-benzotriazole, 5-chlorobenzotriazole, a certain quantity of •OH was also produced and jointly responsible with
atenolol, and metoprolol within 2 h, with corresponding degradation R-O• for effectively removing the selected 21 pharmaceuticals. Since
efficiencies of 100 %, 98 %, 97 %, 96 %, 95 %, 94 %, 92 %, 92 %, 91 %, R-O• exhibits selectivity for eliminating micropollutants with
and 91 % at pH 2. In addition, the removal efficiency of diclofenac and electron-rich moieties like sulfur, nitrogen and hydroxyl groups con­
bezafibrate also exceeded 75 %, reaching 79 % and 77 % at pH 2, taining compounds, aromatic compounds and alkenes, and demon­
respectively. In contrast, the BES-PAA process performed mediocrely on strates a positive correlation, the BES-PAA approach could gain more
removing the other 9 pharmaceuticals. Specifically, only around 63 % of popularity for addressing such wastewaters (Song et al. 2023; Wu et al.
hydrochlorothiazide, 62 % of carbamazepine, 55 % of iohexol, 52 % of 2020). As exhibited in Table S3, the molecular structures of the 21
ketoprofen, 51 % of mefenamic acid, 48 % of clofobric acid, 48 % of selected pharmaceuticals all incorporate the electron-rich groups. This
iomeprol, 45 % of iopromide, and 42 % of bicalutamide, were removed elucidates why the BES-PAA process demonstrates a certain removal
within 2 h at pH 2. When BPM was used, similar pollutant removal efficiency for all these 21 pharmaceuticals. Furthermore, it has been
trends were found in acidic conditions, but overall removal of each observed that the ionization potential (IP) of micropollutants closely

Fig. 7. Elimination of different recalcitrant pharmaceuticals from real wastewater during the BES-PAA process. PAA of 100 µM, initial pharmaceuticals of 10 − 30 µg
l-1, and without applied voltage.

11
R. Zou et al. Water Research 261 (2024) 122065

mirrors their reactivity in PAA-based AOPs (Zhang et al. 2022). Gener­ conductivity and stability will be selected and optimized to ensure
ally, a higher IP value signifies a greater electron-donating capacity, efficient operation at an industrial scale; (iii) Advanced automation
rendering the compound more vulnerable to electrophilic attacks such control systems will be introduced to monitor and adjust reaction con­
as from the electrophilic radicals R-O• and •OH in the BES-PAA process ditions in real-time, enhancing system stability and efficiency. These
(Kim et al. 2020). Additionally, it has been noted that micropollutants measures are expected to facilitate the successful implementation of the
with IP values less than 9.0 eV were more susceptible to oxidation by BES-PAA process at an industrial scale. Additionally, studies have re­
CH3C(O)OO•. Table S3 shows that the IP values of the selected 20 ported scaling up MFC and MEC from batch experiments to pilot-scale
pharmaceuticals (the remaining 5-chlorobenzotriazole with an un­ for wastewater treatment in the anode chamber, along with scaling
known IP value) were all less than 9 eV, which again explains why the MFC and MEC derived bio-electro-Fenton processes for micropollutant
BES-PAA process, dominated by R-O• and •OH, demonstrated a high treatment in the cathode chamber based on MFC and MEC, offering
removal efficiency on the selected pharmaceuticals (Zhang et al. 2022). valuable references for this research (Li et al. 2018; Zou et al. 2020a).
Finally, due to the absence of accurate k values between R-O• and these Third, exploring the potential activation of additional oxidants using
chosen micropollutants, more work is required to comprehend the se­ this approach offers an exciting direction for further research, as un­
lective removal capabilities of the BES-PAA process. Furthermore, we derstanding the fundamental mechanisms of the BES-PAA process could
detected the presence of 6.50 mM of Cl− , 0.16 mM of NOX-N, 0.02 mM of identify oxidants with compatible properties for synergistic activation.
PO4-P and 25.48 mg l-1 of TOC in the selected actual secondary effluent Performic acid (PFA) and peroxypropionic acid (PPA) share structural
(Table S1), which undoubtedly would also deplete the PAA and the free similarities with PAA, as all three contain a peroxy bond (O–O) and an
radicals including R-O• and •OH generated after activation and thus acyl group (Luukkonen and Pehkonen 2017, Wang et al., 2023). These
have a negative effect on the performance of the BES-PAA process. The structural features make it possible for them to exhibit similar chemical
results obtained indicate that increased dosages of PAA would be properties and, in turn, be able to undergo similar activation mecha­
necessary for the treatment of micropollutants in practical water/­ nisms. In a BES system, biogenic electrons derived from electroactive
wastewater treatment processes. bacteria can facilitate the reduction of these peroxy acids at the cathode,
In addition, previous studies have reported some ecotoxicity of leading to the cleavage of the O–O bond and the generation of reactive
pharmaceuticals, including carbamazepine and diclofenac (Wang et al. species. This structural similarity suggests that, like PAA, both PFA and
2022). Hence, an ecotoxicity evaluation was undertaken to evaluate the PPA can be effectively activated in the BES-PAA system to degrade
safety of this technology by using the Vibrio Fischeri technique for micropollutants
luminescence inhibition studies on untreated and treated wastewater
samples (Fig. S22). The findings indicate that the pharmaceuticals un­ 4. Conclusion
derwent detoxification in the BES-PAA process, and the treated effluent
from the BES-PAA process exhibited negligible toxic effects (Discussion A novel energy- and catalyst-free bioelectronically activated PAA
S6). process was studied for the effective elimination of pharmaceuticals in
wastewater. The innovative BES-PAA technique activates PAA at the
3.5. Environmental implications cathode using renewable electrons from anodic electroactive bacteria,
generating reactive radicals like R-O• (mainly as CH3C(O)OO• and
It’s crucial to explore and validate eco-friendly PAA activation CH3C(O)O•) and •OH through a radical pathway, as confirmed by
methods, as current techniques rely heavily on UV irradiation and metal quenching experiments, radical probes, ESR tests, and DFT calculations.
catalysis, carrying environmental risks such as high energy consumption Besides, with an increase in input voltage, PAA dosage, or pH, the rate of
and potential metal ion leakage. This research proposes a novel bio­ SMX elimination was clearly reduced. Furthermore, the identification of
electrochemical approach for the activation of PAA, which has been transformation products using HPLC-MS/MS in conjunction with rele­
shown to be effective in overcoming the above-mentioned drawbacks. In vant DFT calculations revealed that S-N and S-C bond cleavage, amino
the BES-PAA process, the bioanode efficiently eliminated the biode­ oxidation, and methyl substitution of isoxazole ring are the four primary
gradable organic matter (67 % removal from 1 mg l-1 of COD in 12 h) in degradation pathways of SMX in the BES-PAA process. In addition, the
the wastewater. Meanwhile, the abiotic cathode efficiently removed the BES-PAA process demonstrated high efficiency in removing 21 phar­
refractory pharmaceutical SMX without requiring energy input and maceuticals widely detected in EU water environments from the actual
catalyst addition. PAA was shown to be efficiently activated by biogenic secondary effluent of WWTPs under neutral conditions, further high­
electrons derived from organic matter (e.g., acetic acid) oxidation at the lighting the broad applicability of this technology. This research will
bio-anode. The R-O• and •OH generated through bioelectrochemical shed light on developing cost-efficient, ecologically friendly, effective,
activation of PAA via the radical route were the main reactive sub­ and catalyst- and energy-free PAA activation approaches for water
stances responsible for SMX abatement at the cathode. Furthermore, it treatment.
has been demonstrated that the BES-PAA process exhibits outstanding
capability for removing a diverse range of typical pharmaceutical spe­ CRediT authorship contribution statement
cies from real water matrix. The variation in the removal efficiency also
thoroughly validated the existence and contribution of R-O• and •OH. Rusen Zou: Writing – review & editing, Supervision, Resources,
Though promising, there are still some challenges that need to be Project administration, Funding acquisition, Conceptualization. Wen­
addressed before large-scale applications can be implemented. First, qiang Yang: Writing – review & editing, Validation, Resources, Formal
long-term BES operation may result in membrane contamination. analysis. Babak Rezaei: Writing – review & editing, Validation, Re­
Therefore, in the future, regular cleaning or the development of new sources, Methodology, Investigation, Formal analysis. Kai Tang:
anti-fouling membranes will be necessary to mitigate this issue. Second, Writing – review & editing, Methodology, Formal analysis. Kuangxin
given the ultimate objective of implementing the innovative BES-PAA Guo: Writing – review & editing, Validation, Methodology, Investiga­
technique for real-world applications, additional investigation into tion. Pingping Zhang: Methodology, Formal analysis. Stephan Sylvest
scaling up this system will be essential to evaluate its commercial Keller: Writing – review & editing, Supervision, Methodology, Funding
feasibility moving forward. Therefore, for the scaling-up the above acquisition. Henrik Rasmus Andersen: Writing – review & editing,
system, the following specific measures are proposed: (i) A modular Resources, Methodology, Formal analysis. Yifeng Zhang: Writing – re­
design for the reactors will be adopted, allowing multiple units to view & editing, Supervision, Resources, Project administration, Funding
operate in parallel or series, thereby increasing treatment capacity and acquisition, Conceptualization.
simplifying the scaling process; (ii) Electrode materials with high

12
R. Zou et al. Water Research 261 (2024) 122065

Declaration of competing interest Kim, J., Du, P., Liu, W., Luo, C., Zhao, H., Huang, C.-H., 2020. Cobalt/peracetic acid:
advanced oxidation of aromatic organic compounds by acetylperoxyl radicals.
Environ. Sci. Technol. 54 (8), 5268–5278.
The authors declare that they have no known competing financial Kim, J., Huang, C.-H., 2020. Reactivity of peracetic acid with organic compounds: a
interests or personal relationships that could have appeared to influence critical review. ACS ES&T Water 1 (1), 15–33.
the work reported in this paper. Kim, J., Zhang, T., Liu, W., Du, P., Dobson, J.T., Huang, C.-H., 2019. Advanced oxidation
process with peracetic acid and Fe (II) for contaminant degradation. Environ. Sci.
Technol. 53 (22), 13312–13322.
Data availability Li, B., Cheng, X., Zou, R., Yong, X., Pang, C., Su, Y., Zhang, Y., 2022a. Simple modulation
of Fe-based single atoms/clusters catalyst with acidic microenvironment for ultrafast
Fenton-like reaction. Appl. Catal. B Environ. 304, 121009.
Data will be made available on request. Li, R., Manoli, K., Kim, J., Feng, M., Huang, C.-H., Sharma, V.K., 2021. Peracetic
acid–ruthenium (III) oxidation process for the degradation of micropollutants in
water. Environ. Sci. Technol. 55 (13), 9150–9160.
Li, S., Yang, Y., Zheng, H., Zheng, Y., He, C.-S., Lai, B., Ma, J., Nan, J., 2022b.
Acknowledgments Introduction of oxygen vacancy to manganese ferrite by Co substitution for
enhanced peracetic acid activation and 1O2 dominated tetracycline hydrochloride
The authors thank The Carlsberg Foundation (Distinguished Fel­ degradation under microwave irradiation. Water Res. 225, 119176.
Li, W., Ren, R., Liu, Y., Li, J., Lv, Y., 2019. Improved bioelectricity production using
lowships, CF18-0084, Denmark), The Ministry of Foreign Affairs of potassium monopersulfate as cathode electron acceptor by novel bio-electrochemical
Denmark (no. 21-08-DTU, Denmark), VILLUM FONDEN (no. 40828, activation in microbial fuel cell. Sci. Total Environ. 690, 654–666.
Denmark), Independent Research Fund Denmark (Project 1, No. Li, X., Chen, S., Angelidaki, I., Zhang, Y., 2018. Bio-electro-Fenton processes for
wastewater treatment: advances and prospects. Chem. Eng. J. 354, 492–506.
171114, Denmark), and Novo Nordisk Foundation (NNF16OC0021568) Lin, J., Zou, J., Cai, H., Huang, Y., Li, J., Xiao, J., Yuan, B., Ma, J., 2021. Hydroxylamine
for partly funding the research. enhanced Fe (II)-activated peracetic acid process for diclofenac degradation:
efficiency, mechanism and effects of various parameters. Water Res. 207, 117796.
Liu, Z., Ding, H., Zhao, C., Wang, T., Wang, P., Dionysiou, D.D., 2019. Electrochemical
Supplementary materials activation of peroxymonosulfate with ACF cathode: kinetics, influencing factors,
mechanism, and application potential. Water Res. 159, 111–121.
Supplementary material associated with this article can be found, in Long, Y., Yang, P., Wang, C., Wu, W., Chen, X., Liu, W., Cao, Z., Zhan, X., Liu, D.,
Huang, W., 2023. Peroxymonosulfate activation by iron-carbon composite derived
the online version, at doi:10.1016/j.watres.2024.122065. from coal gasification slag for sulfamethoxazole removal: performance evaluation
and mechanism insight. Chem. Eng. J. 456, 140996.
References Lu, J., Zhang, P., Li, J., 2021. Electrocoagulation technology for water purification: an
update review on reactor design and some newly concerned pollutants removal.
J. Environ. Manag. 296, 113259.
Ao, X.-w., Eloranta, J., Huang, C.-H., Santoro, D., Sun, W.-j., Lu, Z.-d., Li, C., 2021.
Luo, J., Cheng, X., Su, Y., Zhang, L., Du, W., Bao, X., Huang, W., Feng, Q., Cao, J., Wu, Y.,
Peracetic acid-based advanced oxidation processes for decontamination and
2022. Metagenomic assembly deciphered the type-dependent effects of surfactants
disinfection of water: a review. Water Res. 188, 116479.
on the fates of antibiotics resistance genes during sludge fermentation and the
Bertagna Silva, D., Buttiglieri, G., Babić, S, 2021. State-of-the-art and current challenges
underlying mechanisms. Sci. Total Environ. 807, 150822.
for TiO2/UV-LED photocatalytic degradation of emerging organic micropollutants.
Luukkonen, T., Pehkonen, S.O., 2017. Peracids in water treatment: a critical review. Crit.
Environ. Sci. Pollut. Res. 28, 103–120.
Rev. Environ. Sci. Technol. 47 (1), 1–39.
Cai, M., Sun, P., Zhang, L., Huang, C.-H., 2017. UV/peracetic acid for degradation of
Matafonova, G., Batoev, V., 2018. Recent advances in application of UV light-emitting
pharmaceuticals and reactive species evaluation. Environ. Sci. Technol. 51 (24),
diodes for degrading organic pollutants in water through advanced oxidation
14217–14224.
processes: a review. Water Res. 132, 177–189.
Chaves, M.d.J.S., Barbosa, S.C., de Melo Malinowski, M., Volpato, D., Castro, Í.B., dos
Mirzaei, A., Eddah, M., Roualdes, S., Ma, D., Chaker, M., 2021. Multiple-homojunction
Santos Franco, T.C.R., Primel, E.G., 2020. Pharmaceuticals and personal care
gradient nitrogen doped TiO2 for photocatalytic degradation of sulfamethoxazole,
products in a Brazilian wetland of international importance: occurrence and
degradation mechanism, and toxicity assessment. Chem. Eng. J. 422, 130507.
environmental risk assessment. Sci. Total Environ. 734, 139374.
Nguyen, T.M.H., Suwan, P., Koottatep, T., Beck, S.E., 2019. Application of a novel,
Chhetri, R.K., Thornberg, D., Berner, J., Gramstad, R., Öjstedt, U., Sharma, A.K.,
continuous-feeding ultraviolet light emitting diode (UV-LED) system to disinfect
Andersen, H.R., 2014. Chemical disinfection of combined sewer overflow waters
domestic wastewater for discharge or agricultural reuse. Water Res. 153, 53–62.
using performic acid or peracetic acids. Sci. Total Environ. 490, 1065–1072.
Pan, Y., Zhou, Q., Meng, F., Cao, Q., Bai, J., Zhang, Z., Liu, Q., Zhang, H., 2024. (1 1 1)
da Silva Rodrigues, D.A., da Cunha, C.C.R.F., Freitas, M.G., de Barros, A.L.C., e Castro, P.
Facet-dominant peracetic acid activation by octahedral CoO anchored hollow carbon
B.N., Pereira, A.R., de Queiroz Silva, S., da Fonseca Santiago, A., Afonso, R.J.d.C.F.,
microspheres for tetracycline degradation. Chem. Eng. J., 152446
2020. Biodegradation of sulfamethoxazole by microalgae-bacteria consortium in
Pazda, M., Kumirska, J., Stepnowski, P., Mulkiewicz, E., 2019. Antibiotic resistance
wastewater treatment plant effluents. Sci. Total Environ. 749, 141441.
genes identified in wastewater treatment plant systems–a review. Sci. Total Environ.,
Dai, C., Li, S., Duan, Y., Leong, K.H., Liu, S., Zhang, Y., Zhou, L., Tu, Y., 2022.
134023
Mechanisms and product toxicity of activated carbon/peracetic acid for degradation
Rizzo, L., Lofrano, G., Gago, C., Bredneva, T., Iannece, P., Pazos, M., Krasnogorskaya, N.,
of sulfamethoxazole: implications for groundwater remediation. Water Res. 216,
Carotenuto, M., 2018. Antibiotic contaminated water treated by photo driven
118347.
advanced oxidation processes: ultraviolet/H2O2 vs ultraviolet/peracetic acid.
Dong, Y., He, C.-S., Sun, S., Liu, J., Xie, Z.-H., Li, J.-Y., Zhou, P., Zhang, H., Dong, F.,
J. Clean. Prod. 205, 67–75.
Lai, B., 2024. Mechanically treated Mn2O3 triggers peracetic acid activation for
Shen, P., Hou, K., Chen, F., Pi, Z., He, L., Chen, S., Li, X., Yang, Q., 2023. Ultra-rapid and
superior non-radical oxidation of micropollutants: identification of reactive
long-lasting activation of peracetic acid by Cu-Co spinel oxides for eliminating
complexes. Water Res. 255, 121486.
organic contamination: role of radical and non-radical catalytic oxidation. Chem.
Du, P., Wang, J., Sun, G., Chen, L., Liu, W., 2022. Hydrogen atom abstraction mechanism
Eng. J. 463, 142344.
for organic compound oxidation by acetylperoxyl radical in Co (II)/peracetic acid
Song, Z., Zhang, Y., Zhang, X., Zhou, X., Chen, Y., Duan, X., Ren, N., 2023. Kinetics study
activation system. Water Res. 212, 118113.
of chloride-activated peracetic acid for purifying bisphenol A: role of Cl2/HClO and
Duan, P., Liu, X., Liu, B., Akram, M., Li, Y., Pan, J., Yue, Q., Gao, B., Xu, X., 2021. Effect
carbon-centered radicals. Water Res., 120274
of phosphate on peroxymonosulfate activation: accelerating generation of sulfate
Wang, B., Wu, H., Zhang, G., Licht, S., 2012. STEP wastewater treatment: a solar thermal
radical and underlying mechanism. Appl. Catal. B Environ. 298, 120532.
electrochemical process for pollutant oxidation. ChemSusChem 5 (10), 2000–2010.
Fan, X., Gao, J., Li, W., Huang, J., Yu, G., 2020. Determination of 27 pharmaceuticals and
Wang, J., Chen, W., Wang, T., Reid, E., Krall, C., Kim, J., Zhang, T., Xie, X., Huang, C.-H.,
personal care products (PPCPs) in water: the benefit of isotope dilution. Front.
2023. Bacteria and virus inactivation: Relative efficacy and mechanisms of
Environ. Sci. Eng. 14 (1), 1–10.
peroxyacids and chlor (am) ine. Environ. Sci. Technol. 57 (47), 18710–18721.
Gao, L., Mao, Q., Luo, S., Cao, L., Xie, X., Yang, Y., Deng, Y., Wei, Z., 2020. Experimental
Wang, G., Hambly, A.C., Dou, Y., Wang, G., Tang, K., Andersen, H.R., 2022. Polishing
and theoretical insights into kinetics and mechanisms of hydroxyl and sulfate
micropollutants in municipal wastewater, using biogenic manganese oxides in a
radicals-mediated degradation of sulfamethoxazole: similarities and differences.
moving bed biofilm reactor (BioMn-MBBR). J. Hazard. Mater. 427, 127889.
Environ. Pollut. 259, 113795.
Wang, J., Wan, Y., Ding, J., Wang, Z., Ma, J., Xie, P., Wiesner, M.R., 2020. Thermal
He, M.-F., Li, W.-Q., Xie, Z.-H., Yang, S.-R., He, C.-S., Xiong, Z.-K., Du, Y., Liu, Y.,
activation of peracetic acid in aquatic solution: the mechanism and application to
Jiang, F., Mu, Y., 2022. Peracetic acid activation by mechanochemically sulfidated
degrade sulfamethoxazole. Environ. Sci. Technol. 54 (22), 14635–14645.
zero valent iron for micropollutants degradation: enhancement mechanism and
Wang, J., Xiong, B., Miao, L., Wang, S., Xie, P., Wang, Z., Ma, J., 2021. Applying a novel
strategy for extending applicability. Water Res. 222, 118887.
advanced oxidation process of activated peracetic acid by CoFe2O4 to efficiently
Henao, L.D., Turolla, A., Antonelli, M., 2018. Disinfection by-products formation and
degrade sulfamethoxazole. Appl. Catal. B Environ. 280, 119422.
ecotoxicological effects of effluents treated with peracetic acid: a review.
Wang, Z., Wang, J., Xiong, B., Bai, F., Wang, S., Wan, Y., Zhang, L., Xie, P., Wiesner, M.
Chemosphere 213, 25–40.
R., 2019. Application of cobalt/peracetic acid to degrade sulfamethoxazole at
Hu, P., Long, M., 2016. Cobalt-catalyzed sulfate radical-based advanced oxidation: a
neutral condition: efficiency and mechanisms. Environ. Sci. Technol. 54 (1),
review on heterogeneous catalysts and applications. Appl. Catal. B Environ. 181,
464–475.
103–117.

13
R. Zou et al. Water Research 261 (2024) 122065

Wu, J., Zou, J., Lin, J., Li, S., He, L., Wu, Z., Li, Q., Gong, C., Ma, J., 2024. Overlooked Zhang, T., Huang, C.-H., 2020. Modeling the kinetics of UV/peracetic acid advanced
role of coexistent hydrogen peroxide in activated peracetic acid by Cu (II) for oxidation process. Environ. Sci. Technol. 54 (12), 7579–7590.
enhanced oxidation of organic contaminants. Environ. Sci. Technol. https://doi.org/ Zhang, X., Sun, P., Wei, K., Huang, X., Zhang, X., 2020. Enhanced H2O2 activation and
10.1021/acs.est.3c09753. sulfamethoxazole degradation by Fe-impregnated biochar. Chem. Eng. J. 385,
Wu, K., Chen, R., Zhou, Z., Chen, X., Lv, Y., Ma, J., Shen, Y., Liu, S., Zhang, Y., 2022. 123921.
Elucidating Electrocatalytic Oxygen Reduction Kinetics via Intermediates by Time Zhao, Y.-M., Oliveira, M., Burgess, C.M., Cropotova, J., Rustad, T., Sun, D.-W., Tiwari, B.
Resolved Electrochemiluminescence. Angew. Chem. Int. Ed. 135 (12), e202217078. K., 2021. Combined effects of ultrasound, plasma-activated water, and peracetic acid
Wu, W., Tian, D., Liu, T., Chen, J., Huang, T., Zhou, X., Zhang, Y., 2020. Degradation of on decontamination of mackerel fillets. LWT 150, 111957.
organic compounds by peracetic acid activated with Co3O4: a novel advanced Zheng, H., Zhang, L., Zhang, G., Gan, Y., Xie, M., Zhang, S., 2021. UV-induced redox
oxidation process and organic radical contribution. Chem. Eng. J. 394, 124938. conversion of tellurite by biacetyl. Environ. Sci. Technol. 55 (24), 16646–16654.
Xu, W., Huang, D., Wang, G., Zhou, W., Li, R., Huang, H., Du, L., Xiao, R., Chen, S., 2024. Zhu, S., Jin, C., Duan, X., Wang, S., Ho, S.-H., 2020. Nonradical oxidation in persulfate
Doped Cu0 and sulfidation induced transition from RO• to• OH in peracetic acid activation by graphene-like nanosheets (GNS): differentiating the contributions of
activation by sulfidated nano zero-valent iron-copper. Water Res. 256, 121621. singlet oxygen (1O2) and sorption-dependent electron transfer. Chem. Eng. J. 393,
Yang, J., Jia, Y., Yao, J., Hu, F., Zhang, Y., 2024. Efficient degradation of propranolol 124725.
with electrochemically activated peracetic acid process: mechanisms, degradation Zong, Y., Shao, Y., Zeng, Y., Shao, B., Xu, L., Zhao, Z., Liu, W., Wu, D., 2021. Enhanced
pathway, and toxicity evaluation. Chem. Eng. J. 481, 148083. oxidation of organic contaminants by iron (II)-activated periodate: the significance
Yang, Y., Lu, X., Jiang, J., Ma, J., Liu, G., Cao, Y., Liu, W., Li, J., Pang, S., Kong, X., 2017. of high-valent iron–oxo species. Environ. Sci. Technol. 55 (11), 7634–7642.
Degradation of sulfamethoxazole by UV, UV/H2O2 and UV/persulfate (PDS): Zou, R., Hasanzadeh, A., Khataee, A., Meng, F., Angelidaki, I., Zhang, Y., 2021a.
formation of oxidation products and effect of bicarbonate. Water Res. 118, 196–207. Synergistic effect for efficient oxidization of refractory organics with high chroma by
Yao, J., Zhang, Y., Dong, Z., 2021. Enhanced degradation of contaminants of emerging an innovative persulfate assisted microbial electrolysis ultraviolet cell. Chem. Eng. J.
concern by electrochemically activated peroxymonosulfate: performance, 419, 129477.
mechanism, and influencing factors. Chem. Eng. J. 415, 128938. Zou, R., Tang, K., Angelidaki, I., Andersen, H.R., Zhang, Y., 2020. An innovative
Yuan, Z., Ni, Y., Van Heiningen, A., 1997. Kinetics of the peracetic acid decomposition: microbial electrochemical ultraviolet photolysis cell (MEUC) for efficient
part II: pH effect and alkaline hydrolysis. Can. J. Chem. Eng. 75 (1), 42–47. degradation of carbamazepine. Water Res. 187, 116451.
Zhang, L., Chen, J., Zhang, Y., Xu, Y., Zheng, T., Zhou, X., 2022. Highly efficient Zou, R., Tang, K., Hambly, A.C., Chhetri, R.K., Yang, X., Xu, M., Su, Y., Andersen, H.R.,
activation of peracetic acid by nano-CuO for carbamazepine degradation in Angelidaki, I., Zhang, Y., 2021b. A novel persulfate-photo-bioelectrochemical hybrid
wastewater: the significant role of H2O2 and evidence of acetylperoxy radical system promoting the degradation of refractory micropollutants at neutral pH.
contribution. Water Res. 216, 118322. J. Hazard. Mater. 416, 125905.
Zhang, L., Chen, J., Zhang, Y., Yu, Z., Ji, R., Zhou, X., 2021. Activation of peracetic acid Zou, R., Tang, K., Hambly, A.C., Wünsch, U.J., Andersen, H.R., Angelidaki, I., Zhang, Y.,
with cobalt anchored on 2D sandwich-like MXenes (Co@ MXenes) for organic 2022. When microbial electrochemistry meets UV: the applicability to high-strength
contaminant degradation: high efficiency and contribution of acetylperoxyl radicals. real pharmaceutical industry wastewater. J. Hazard. Mater. 423, 127151.
Appl. Catal. B Environ. 297, 120475. Zou, R., Yang, W., Rezaei, B., Tang, K., Zhang, P., Andersen, H.R., Keller, S.S., Zhang, Y.,
Zhang, L., Chen, J., Zheng, T., Xu, Y., Liu, T., Yin, W., Zhang, Y., Zhou, X., 2023. Co–Mn 2024. Sustainable bioelectric activation of periodate for highly efficient
spinel oxides trigger peracetic acid activation for ultrafast degradation of micropollutant abatement. Water Res., 121388
sulfonamide antibiotics: unveiling critical role of Mn species in boosting Co activity.
Water Res. 229, 119462.

14