Chemical Engineering Journal 429 (2022) 132176

Contents lists available at ScienceDirect

Chemical Engineering Journal

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

A self-sufficient electro-Fenton system with enhanced oxygen transfer for

decontamination of pharmaceutical wastewater
Yingshi Zhu, Fengxia Deng, Shan Qiu *, Fang Ma *, Yanshi Zheng, Lei Gao
State Key Laboratory of Urban Water Resource and Environment, School of Environment, Harbin Institute of Technology, Harbin 150090, PR China

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

Keywords: The need for extra iron catalyst, limited selectivity of the oxygen reduction reaction (ORR) to produce H2O2, and
Self-sufficient electro-Fenton harsh pH applicability still plague the electro-Fenton (EF) process. To alleviate these concerns simultaneously, a
Fe, N and S co-doped electrode novel self-sufficient EF system was constructed, in which the newly fabricated Fe, N and S co-doped cathode was
H2O2 selectivity
aerated via microporous aeration and a greater number of cathodes was utilized. Results indicated that the
O2 microbubble
Auto-adjusted pH
prepared cathode achieved a high H2O2 accumulation (15.60 ± 0.16 mg L− 1 cm− 2) coupled with nearly 90%
H2O2 selectivity. The microporous aeration tube, which delivered microbubbles and changed the diffusion di­
rection to supply more O2 to the electrode active sites, is beneficial to the H2O2 formation. Apart from its high
H2O2 accumulation, the proposed system was set up to automatically reduce the bulk pH by regulating different
current densities on electrodes. The proposed EF system was employed to treat real pharmaceutical wastewater,
in which ⋅OH, 1O2, ⋅O2− and Fe(IV) were the main active species. Finally, an improvement of BOD5/COD values
after EF treatment was a good indicator to confirm the feasibility of the developed system as a pretreatment prior
to the subsequent engineering applications.

1. Introduction selectivity of the cathode, various cathode modification strategies have

been reported [8,9]. Among which, carbon materials are most
Antibiotics are detected in aqueous environment and gaining atten­ commonly used cathode and doping heteroatoms (N, B, P, S and F, etc.)
tion [1]. Because it can increase bacterial resistance, which in turn into carbon skeleton can induce Lewis acid base centers which is the
render an adverse impact on human health [2]. Therefore, it is urgent to active sites of ORR [10]. Moreover, the catalytic activity of carbon
seek for an effective and economical technology for its disposal. Among materials can be further boosted by the synergistic electronic in­
a variety of methods, electro-Fenton (EF) exerts a tremendous fascina­ teractions between co-doped heteroatoms and adjacent C atoms in sys­
tion on researchers [3–5]. Because EF generates Fenton’s reagent in-situ tems co-doped with two or more heteroatoms [11]. It was reported that
(Eqs. (1) and (2)), not only solving the risk of H2O2 storage and dosage, the ability to retain the O–O bond determines the selectivity for H2O2
but also easing the accumulation of iron sludge [6]. The outstanding [12]. Our previous work reported co-doping of N and S in carbon
decontamination ability of EF technique can be explained by the for­ framework can both facilitate oxygen adsorption and the retention of
mation of OH (Eq. (3)) [7] O–O bonds, improving the selectivity of the 2e− transfer ORR [13].
Interestingly, the single-atom catalysts (such as Fe, Co etc.) embed into
O2 + 2H+ + 2e− → H2O2 (1) carbon skeleton regulated the mobility of electrons and circumvented
Fe 3+ −
+ e → Fe 2+
(2) the undesirable leaching by wrapping the metal species in graphene
lattice [14]. To our knowledge, few studies have investigated the cata­
H2O2 + Fe 2+
→ Fe 3+
+ ⋅OH + OH −
(3) lytic H2O2 generation by Fe, N and S co-doped carbon material cathodes
in the EF system.
According to the Eqs. (1)–(3), providing sufficient Fenton’s reagents
Apart from reactivity/selectivity promotion, oxygen mass transfer
is the key to boost EF performance. Both oxygen reduction reaction
also plays a significant role in H2O2 accumulation during 2e− transfer
(ORR) reactivity/selectivity and oxygen mass transfer efficiency are
ORR. Therefore, Pérez et al. designed a reactor based on venturi jet
crucial for H2O2 promotion. To alleviate the issue of low reactivity and

* Corresponding authors.
E-mail addresses: qiushan@hit.edu.cn (S. Qiu), mafang@hit.edu.cn (F. Ma).

https://doi.org/10.1016/j.cej.2021.132176
Received 5 July 2021; Received in revised form 23 August 2021; Accepted 29 August 2021
Available online 3 September 2021
1385-8947/© 2021 Elsevier B.V. All rights reserved.
Y. Zhu et al. Chemical Engineering Journal 429 (2022) 132176

aeration to improve the oxygen transfer efficiency at the gas–liquid 2. Materials and methods
interface [15]. Considering gaseous oxygen concentration (101 kPa, 45
mM) is 45-fold than dissolved one (1 mM), direct use gaseous one is a 2.1. Preparation of cathodes
wise method to increase the yield of H2O2 [16]. With this regard, Zhou
et al. proposed a floating cathode instead of the commonly submerged All regents used in this study were described in S1 (SI). The process of
one and found that H2O2 yield increased from 14.4 mg L− 1 to 61.7 mg electrodes preparation was shown in Fig. 1. 10 g of melamine (nitrogen-
L− 1 [17]. Despite seeing an enhancement in H2O2 accumulation, venturi containing precursor, CN), 10 g of dibenzyl disulfide (sulfur-containing
jet aeration reactor required additional operating costs along with precursor, CS) and 0 g (3 g or 6 g or 9 g) of ferrous sulfate were weighed
floating cathodes only elevated oxygen utilization of the cathode. Beside and grinded. Then the mixture was placed in a crucible with a lid and
the abovementioned strategies, regulation of oxygen bubble size into pyrolyzed into a tube furnace (600 ◦ C for 3 h with N2 protection). After
micro- or nanoscale is an effective method to rise oxygen concentration natural cooling down, the powder was ground again and passed through
based on Young–Laplace equation and the smaller bubble favors the a 200-mesh sieve. In the following investigation, catalyst doped with
following utilization [18]. Nevertheless, there is a lack of relevant in­ only N and S atoms was denoted as CNS, while the one with Fe was
vestigations about refined bubble on H2O2 production. marked as CNS-Fe. Among which, CNS-Fe-1, CNS-Fe-2, CNS-Fe-3 were
Generally, conventional EF process can only work efficiently at harsh samples containing ferrous sulfate with 3 g, 6 g and 9 g, respectively.
pH conditions (pH = 2.8–3), and the catalytic effect is extremely low The pressure-rolling approach was adopted to transfer powder catalyst
when the pH is higher than 5 [19]. Therefore, a variety of heterogeneous into a sheet electrode and the details was illustrated below: I) 40 mL of
catalysts coupled with bifunctional cathode have been prepared to ethanol, 1.5 g of graphite powder and 1 g of different catalysts were
broaden the applicable pH range of EF. The bifunctional cathode not mixed in a beaker and sonicated for 20 min to form a homogeneous
only generates H2O2 in situ, but also participates in the heterogeneous mixture, II) 0.3 mL of PTFE (an adhesive) was added and continued to
and heterogeneous reaction through the surficial iron and the dissolved sonicate for another 20 min, III) the mixture was incubated at 80 ◦ C in
Fe2+/Fe3+, respectively [20]. In addition, it is also possible to auto­ water bath to obtain the paste, Ⅳ) The paste containing the catalyst was
matically adjust the pH in the absence of adding acid by using the semi- applied evenly on top of a stainless-steel mesh and was pressed with a
reaction of H+ ions generated from the anodic oxidation of H2O [21,22]. roller press to make a 0.3 mm thickness. Finally, it was placed in an oven
Yuan et al. proposed to establish a three-electrode EF system with one at 80 ◦ C for 24 h and then calcined at 340 ◦ C for 1 h under N2 conditions.
anode and two cathodes to achieve pH control by adjusting the total
current and current distribution [23]. However, to our knowledge, no 2.2. EF treatment of pharmaceutical wastewater
one has yet established a dual-function cathode and automatically
adjusted acid pH-worthy EF system and explored its specific mechanism Pharmaceutical wastewater was collected from the secondary sedi­
of action. mentation tank effluent of the pharmaceutical plant located in Yin­
According to above analysis, numbers of the current investigations chuan, China, and the key parameters for the wastewater was
about EF are limited to alleviating a single issue, including ORR reac­ characterized in Table S1. A working volume of 200 mL pharmaceutical
tivity/selectivity, oxygen mass transfer and acidic working pH range. wastewater was degraded by the EF with two anodes and three cathodes
However, simply solving one challenge and leaving others alone is not under a constant current mode (Fig. S1). A boron doped diamond (BDD)
enough, when it comes to bringing EF to a large-scale application. With of size 2 cm × 8 cm was used as the anode and the same size of cathode
this regard, a novel dual-function cathode and automatically adjusted developed in this study was working as the cathode. The gap between
acid pH-EF system was developed at the first time to attempt to simul­ cathode and anode was set to 1 cm. The desired pH of electrolyte was
taneously solve problems in EF. adjusted by 1 M H2SO4 or NaOH. Air with a flow rate of 0.05 m3 h− 1 was
Therefore, the objective of this study was to combine H2O2 promo­ supplied during the whole EF process.
tion with pH broadening to establish a self-sufficient EF system. In this
work, the Fe, N and S co-doped cathode was fabricated and the active 2.3. Analytical procedure
sites that catalyze 2e− transfer ORR and promote the Fenton reaction
were explored. Oxygen mass transfer promotion through microporous X-ray diffraction (XRD) patterns were recorded with a Bruker D8
aeration was supported by computational fluid dynamics (CFD) simu­ advance. Transmission electron microscopy (TEM) images were ob­
lation. Moreover, the active species in the EF system was examined and tained from a FEI Talos F200X. X-ray photoelectron spectroscopy (XPS)
the possible mechanisms of its generation were proposed. Furthermore, was carried out with a ThermoFisher ESCALAB 250Xi. Scanning electron
the effects of the initial pH and applied current density on the removal of microscopy (SEM) was performed with a ZEISS Sigma-500 equipped
treating pharmaceutical wastewater which containing antibiotics were with an energy dispersive spectrometer (EDS). Electron spin resonance
investigated. Meanwhile, the effect of high anode current on the pH (ESR) spectra was recorded with a Bruker A200 S-9.5/12 spectrometer.
adjustment of the system was also explored. Finally, the feasibility of the The electrochemical measurements were conducted with a Chenhua CHI
self-sufficient EF system as a pretreatment process for actual pharma­ 760E workstation and details were given in S2 (SI). Chemical oxygen
ceutical wastewater was verified. dioxide (COD) and biological oxygen demand (BOD5) were determined
according to standards of HJ/T 399-2007 and HJ505-2009 in China
[24]. Dissolved oxygen (DO) was investigated by a DZS-708-B multi-

Fig. 1. Schematic diagram of method for the proposed electrodes.

2
Y. Zhu et al. Chemical Engineering Journal 429 (2022) 132176

parameter analyzer. The concentration of H2O2 was detected by an io­ which, Fe3N can promote the ORR, while Fe3O4 and FeS can enhance the
dide method with a Purkinje TU-1810 spectrophotometer at 351 nm Fenton reaction [29–31].
[13]. The current efficiency (CE) and energy consumption (ECH2O2) for To further investigate the effect of Fe doping on the elemental
H2O2 production were shown in S3 (SI). Computational fluid dynamics composition, XPS tests were performed (Fig. S2–S5). The atomic per­
(CFD) simulation was performed by ANSYS FLUENT 18.0 software and centage of S and O raised with Fe enhancement since Fe, S and O ele­
details were shown in S4 (SI). The concentration of leached Fe was ments are all contained in the FeSO4⋅7H2O precursor (Table S3).
determined with a Perkin Elmer NexION 300Q, while the concentration However, the percentage of N atoms decreased, which is opposite to the
of ⋅OH was determined using TA as a trapper by a Hitachi F-2700 rise of Fe, S and O. This is because these atoms replaced N atoms in
spectrophotometer [25]. The excitation emission matrix (EEM) fluo­ catalyst to form stable state, which is consistent with the analysis results
rescence spectroscopy was measured by a Hitachi F-2700 spectropho­ of XRD. Moreover, Fig. S6 shows CNS-Fe-x catalysts have a high pro­
tometer. The 1O2 was determined by using FFA as a probe [26]. The ⋅O2− portion of pyridinic N, Fe-Nx and pyrrolic N, which were reported as
was examined by using XTT as a probe and detected with a Purkinje TU- active sites to ORR. Because it boosts spin density, current density, hy­
1810 spectrophotometer at 470 nm [27]. The detail conditions for STZ, drophilicity of catalysts [8,32].
SDZ, SDM, FFA, PMSO and PMSO2 were shown in S5 (SI) and Table S2.
The calculation of energy consumption per unit of TOC mass (ECTOC) 3.2. Electrocatalytic performance of Fe, N and S co-doped electrodes
was shown in S6 (SI). The detail conditions for STZ degradation in­
termediates were given in S7 (SI). LSV experiments were performed to compare electrocatalytic activ­
ity of various Fe, N and S co-doped electrodes. Fig. 3a shows that there is
3. Results and discussion lack of any redox peaks in the N2-saturated electrolyte when different
Fe-doped electrodes were used as working electrodes. However, the
3.1. Characterizations of the Fe, N and S co-doped catalysts appearance of reduction peaks ranged from − 0.6 V to − 0.8 V coupled
with an increase in the current response indicated the existence of ORR
XRD characterization was carried out to explore the phase compo­ when oxygen is purged in the bulk [33]. In addition, the conductivity of
sition of different catalysts. Fig. 2a shows the XRD diffraction of N, S co- the electrode increased with the improvement of Fe doping in a certain
doped CNS catalyst has two peaks at 12.8◦ and 27.5◦ , corresponding to range and the conductivity of CNS-Fe-2 was the best. This was attributed
(001) and (002) crystal planes of g-C3N4 (JCPDS 87-1526) [28]. How­ to that the degree of graphitization could be enhanced by transition
ever, with an increase of Fe concentration, the diffraction peaks of g- metals doping, like Fe, as indicated in Fig. S7 [34].
C3N4 gradually weakened with new peaks appeared at other positions, EIS measurements were then carried out to explore the kinetic dif­
indicating that some new phase was formed with Fe introduction. And ferences of various electrodes (Fig. 3b). Detail EIS fitting results were
the area of this new peak increased with the rise of Fe concentration, shown in Table S4. In Fig. 3c and d, Rs represents the solution resistance,
confirming the formation of Fe related crystal structure. This was further Rct and Rads reflect mass transfer resistance and adsorption resistance.
proved by comparing this peak with PDF card, where the peaks at 29.9◦ And Cct and Cads are the capacitance of electrocatalytic reduction for
and 34.0◦ belong to (220) and (311) planes of Fe3O4 (JCPDS 89-0951), oxygen and W reflects the Warburg diffusion element [35]. The equiv­
and the peak at 43.9◦ was indexed as (-1–1 1) lattice plane of Fe3N alent circuit of Fe-doped electrodes (Fig. 3d) was different from the
(JCPDS 76-0091). As for Fe, N and S co-doped catalysts are not undoped ones (Fig. 3c), suggesting that the doping of Fe could change
composed of one phase, there will be overlapping of diffraction peaks. the electrode dynamics. It is worth noting that Fe doping significantly
Therefore, the peaks at 29.9◦ , 34.0◦ , 43.9◦ and 53.2◦ can also match with reduced Rct and Rads, proving Fe doping was conducive to lower charge
(110), (112), (114) and (213) planes of FeS (JCPDS 76-0962), respec­ transfer resistance. In addition, the lowest level of Rct (2.46 Ω) and Rads
tively. The above XRD analysis demonstrated that Fe doping can destroy (3.98 Ω) at CNS-Fe-2 suggested that proper Fe doping was beneficial to
the bond between C and N, resulting in the formation of Fe3O4, Fe3N and the adsorption of electrolyte to the electrode surface and might further
FeS. interface reaction. Furthermore, W in the low frequency region reflects
Fig. 2b presents a low-magnification TEM image of CNS-Fe-2, in soluble Fe of the electrode interface diffused into the bulk solution [36].
which a two-dimensional sheet structure was seen. Moreover, the HR- In a word, CNS-Fe-2 witnessed the lowest Faraday resistance and pro­
TEM image in Fig. 2c verified the polycrystalline nature of the pre­ moted ORR. And it was in good agreement with LSV curves.
pared material. The fingerprints with lattice spacing of 0.289, 0.265, The electrochemical surface area (ECSA) was estimated by
0.205 and 0.175 nm in Fig. 2c correspond to the (110), (112), (114) and comparing the electrochemical double layer capacitance (EDLC) and it
(213) crystal planes of FeS. And the fingerprint with a lattice spacing of was calculated based on CV curves at a sweep speed of 5–30 mV s− 1
0.205 nm is ascribed to the (-1–1 1) crystal plane of Fe3N. In addition, (Fig. S8) [37]. As shown in Fig. 3e, the EDLC for CNS was only 1.67 mF
the presence of Fe3O4 was confirmed by the fingerprints with lattice cm− 2, locating at the lowest level of these electrodes. The order for other
spacing of 0.289 and 0.265 nm, which are consistent with the (220) and Fe-doped electrodes was given as follows: CNS-Fe-2 > CNS-Fe-1 > CNS-
(311) crystal planes of Fe3O4. According to the analysis, the main con­ Fe-3. Among them, CNS-Fe-2 electrode with the addition of Fe reached
tent of the CNS-Fe-2 catalyst includes Fe3N, Fe3O4 and FeS. Among 19.02 mF cm− 2, which is 11.39-fold prior to the modification. It

Fig. 2. (a) XRD patterns of different catalysts. (b) TEM and (c) high resolution (HR)-TEM image of CNS-Fe-2.

3
Y. Zhu et al. Chemical Engineering Journal 429 (2022) 132176

Fig. 3. (a) LSV curves and (b) Nyquist plots of different electrodes. Equivalent circuit of (c) CNS, (d) CNS-Fe-x (x = 1, 2, 3). (e) Current density-scan rates curves of
four electrodes. (f) RRDE results in O2-saturated solution at 1600 rpm rotation speed. (g) H2O2 selectivity and electron transfer number. Conditions: 0.05 M Na2SO4,
pH = 3.

Fig. 4. (a) The concentrations of DO in different systems. (b–e) CFD simulation of molar concentration of O2 in different aeration systems. (f) The concentration of
H2O2 with different aeration methods. Conditions: 0.05 M Na2SO4, pH = 3, j = 10 mA cm− 2, 0.05 m3 h− 1 aeration rate and CNS-Fe-2 cathode. (g–j) Schematic
diagram of bubble transmission with different aeration methods.

4
Y. Zhu et al. Chemical Engineering Journal 429 (2022) 132176

suggested that CNS-Fe-2 contained numerous electrocatalytic active 3.4. Oxidizing species in EF system
sites for ORR [38].
RRDE measurements were performed to get insight into the ORR Although ⋅OH is the most frequently reported active species in EF,
activity and H2O2 selectivity (Fig. 3f). As indicated in Fig. 3 g, H2O2 other strong oxidizing substances could also present as well, like ⋅SO4− ,
yields and electron transfer numbers were calculated based on RRDE Fe(IV) and 1O2, etc. Therefore, scavengers including ethanol, TBA, FFA,
results. Among them, CNS-Fe-2 had the highest H2O2 yields approaching PBQ was externally added to evaluate the corresponding active species,
90% and its electron transfer number was around 2.2, indicating its where ethanol capture ⋅SO4− , Fe(IV) and ⋅OH, TBA capture ⋅OH, PBQ
tendency for the 2e− transfer ORR [39]. exclusively quench ⋅O2− and FFA consume 1O2 and ⋅OH [13]. Fig. 5a
shows 98.21 ± 0.37% of STZ (a target pollutant and a common anti­
3.3. Enhancement of oxygen mass transfer biotic, used as a short-acting sulfa drug, but it must be removed from
wastewater [42]) can be removed in 60-min reaction in the absence of
Although 2e− transfer ORR reactivity and selectivity has been capture agents. However, the addition of ethanol (51.81 ± 2.06%) had a
boosted by developing the CNS-Fe-2 electrode, the issue of low O2 mass higher inhibition on STZ removal than TBA (39.84 ± 4.52%), indicating
transfer caused by its limited solubility is still existing. To alleviate this there may exist ⋅SO4− or Fe(IV) except for ⋅OH [43]. FFA significantly
concern, various aeration methods were compared in terms of accu­ inhibited STZ removal (68.32 ± 2.16%), confirming the formation of
1
mulated H2O2 since it provides flow to the electrolyte and enhances O2 [44]. Moreover, a lower STZ decay was seen with PBQ addition, and
mass transfer [40]. As shown in Fig. 4a, the equilibrium DO concen­ it proved the existence of ⋅O2− [44]. ESR experiments were carried out to
tration in the systems using aeration was higher compared to the equi­ further prove the presence of these active species in the system directly.
librium DO concentration in the non-aeration system (4.49 ± 0.01 mg Fig. 5b depicts the 4-fold peaks of DMPO-⋅OH adducts with an intensity
L− 1). The DO with coarse and fine pipes was lower than that with sand ratio of 1:2:2:1 was observed, indicating the generation of ⋅OH [45]. And
core aerator and microporous aeration tubes. This is because, as shown the concentration of ⋅OH reached 174.74 ± 3.93 μM in 60 min
in Fig. S9a–d, the air bubble size in the pipe aeration was large, resulting (Fig. S11a). Moreover, Fig. 5c presents the three typical peaks of
in diffusing back to the atmosphere immediately and contacting elec­ TEMP-1O2 adducts with a 1:1:1 triplet and it again proved existence of
1
trode in a short time. However, sand core aeration offered small size of O2 [45]. Furthermore, 1O2 was examined by using FFA as a probe. The
air bubbles and thus increased the DO concentration to 7.80 ± 0.01 mg ethanol was also added in 1O2 determination system to exclude ⋅OH and
L− 1 (Fig. S9c). It is noteworthy the highest equilibrium DO concentra­ Fe(IV) role in FFA degradation (Fig. S11b). Results showed that with the
tion (8.35 ± 0.08 mg L− 1) was observed in microporous aeration tube addition of the trapping agent, FFA removal rate could reach 49.36 ±
since it provides microbubbles (Fig. S9d). According to Fick’s law and 2.45% in 60 min, revealing 1O2 was produced in the system [26].
double film theory, microporous aeration tube provides small size However, as for ⋅O2− , only weak six characteristics of DMPO-⋅O2− ad­
bubbles, even microbubbles, leading to the enhancement of oxygen ducts was observed (Fig. 5d). One explanation is ⋅O2− is unstable in
transfer rate in the liquid phase [41]. water and its concentration is still low even captured in the combined
A CFD simulation was carried out to visually express the influence of methanol and water solution [33]. Another possibility is that the
microbubbles on mass transfer. Fig. 4b–e compares the DO distribution generated ⋅O2− was immediately converted to 1O2 through the Haber-
in the four aeration systems when the oxygen concentration in the Weiss reaction, recombination and ⋅OH-mediated pathways [27].
microporous aeration tube system almost reached saturated DO. Only Therefore, XTT-an indicator for ⋅O2− by forming XTT-formazan-was
half of the sand core aeration system was saturated with DO, and even used to further check the generation of ⋅O2− [27]. Fig. S11c shows the
less of the coarse and fine tubes were saturated with it. This is because absorbance of XTT-formazan gradually increases, confirming its pres­
according to the Young-Laplace equation, the presence of microbubbles ence of ⋅O2− in this study.
in the system causes a high Laplace pressure, which accelerates the Fe(IV) (a redox potential of 2.0 V vs. SHE for Fe(IV)/Fe3+) can attack
dissolution of oxygen [18]. In addition, it shows that the DO gradually target pollutants selectively [46]. To investigate the existence of Fe(IV),
increases from bottom to top, indicating that the DO in the system was PMSO was used as a probe. And considering pH can influence the Fenton
generated when bubbles rising or bursting at the gas–liquid interface. reaction to generate ⋅OH or Fe(IV) species, different initial pH was
In this part, CNS-Fe-2 was chosen as the working electrode since it adjusted [47]. PMSO was chosen as the Fe(IV) indicator because its
has the highest H2O2 selectivity in the RRDE experiment⋅H2O2 accu­ byproducts oxidized by Fe(IV) is different from ⋅OH [46]. Fig. 6a shows
mulation was directly measured in Fig. 4f, where it increased using sand more PMSO could be oxidized at the conditions of acidic or neutral so­
core aeration (12.07 ± 0.04 mg L− 1 cm− 2) compared to fine pipe aera­ lution than alkaline one. PMSO dropped from initial 249.77 ± 1.32 μM
tion (10.32 ± 0.04 mg L− 1 cm− 2). It can be explained that microbubbles to 76.32 ± 1.94 μM at pH of 3, it could only drop to 170.78 ± 4.88 μM
provided by sand core aeration could form a three-phase interface on the when pH rose to 9. Fig. 6b shows the generation of PMSO2-the
cathode and thus increase the H2O2 accumulation (Fig. 4i). In addition, byproduct of PMSO oxidation by Fe(IV) can reach 50.45 ± 0.05 μM at
H2O2 accumulation was much higher with microporous aeration tube pH = 9, which is much higher than at acidic or neutral conditions. To
aeration (15.60 ± 0.16 mg L− 1 cm− 2) than sand-core aeration. Aside investigate the role of Fe(IV), ηPMSO2 was calculated by the ratio of
from the improvement of DO, microporous aeration was facing cathode PMSO2 production to PMSO consumption [43]. As deduced in Fig. 6c,
directly, in this way, the microbubbles could go through the cathodic ηPMSO2 was higher than 60% at pH = 9, proving Fe(IV) tends to form in
active sites easily and then it could be used in the three-phase interface high pH value.
(Fig. 4j). Moreover, H2O2 accumulation in the system was compared According to the above results, it can be seen that the main active
with literatures to highlight the superiority of the modified electrode. species in the system were ⋅OH, 1O2, ⋅O2− and Fe(IV). To clarify the
Table S5 shows CNS-Fe-2 cathode could accumulate the highest level of mechanism of pollutant removal, the contribution ratio of these active
H2O2, indicating its high selectivity towards 2e− transfer ORR catalytic species in the system was obtained by calculating the inhibition rate of
performance. Furthermore, the system using microporous aeration tube adding a trapping agent to the removal of STZ. Fig. S12 shows the
exhibited the highest CE (98.36 ± 1.00%) and the lowest ECH2O2 (6.41 contribution of ⋅OH and ⋅O2− to the removal of STZ exceeds 30%, while
± 0.07 Wh g− 1 H2O2) among different aeration methods due to the the effect of Fe(IV) is only 9.64%.
improvement in oxygen utilization efficiency via microbubbles Based on the detected active species including ⋅OH, 1O2, ⋅O2− and Fe
(Fig. S10). (IV), a feasible mechanisms of these active species were proposed
(Fig. 7). ⋅OH is one of the main radicals and its generation paths can be
divided into: I) first, O2 is reduced to form H2O2 (Eq. (1)) at the cathode
interface via 2e− transfer ORR and then it is catalyzed by surficial or

5
Y. Zhu et al. Chemical Engineering Journal 429 (2022) 132176

2
Fig. 5. (a) Effect of scavengers on STZ removal using CNS-Fe-2 cathode. Conditions: 0.05 M Na2SO4, 50 mM scavengers, pH = 3, j = 10 mA cm− and t = 60 min. ESR
spectrum of (b) DMPO-⋅OH, (c) TEMP-1O2 and (d) DMPO-⋅O2− adducts. Conditions: 100 mM DMPO and 50 mM TEMP.

2
Fig. 6. The concentrations of (a) PMSO and (b) PMSO2. (c) the values of ηPMSO2. Conditions: 0.05 M Na2SO4, j = 10 mA cm− and initial CPMSO = 250 μM.

dissolved iron (discussed in Fig. S13) to produce ⋅OH (Eq. (3)), II) O2
obtains an electron to generate ⋅O2− at the cathode interface (Eq. (4)), ⋅O2− + H2O2 → O2 + OH− + ⋅OH (6)
and it combines with H+ to strengthen H2O2 production (Eq. (5));
finally, H2O2 is decomposed into ⋅OH (Eq. (6)) [45]. Moreover, pH H2O2 + ≡Fe/Fe 2+
→ H2O + Fe OIV 2+
(7)
regulates the formation of Fe(IV) and ⋅OH (Eq. (7)), where basic pH ⋅O2− + ⋅OH → 1O2 + OH− (8)
favors Fe(IV) and acid condition benefits ⋅OH [43]. Regarding 1O2 for­
mation pathway, there are three approaches: I) the photosensitive sub­ 1
4⋅OH → O2 + 2H2O (9)
stance like g-C3N4 on the cathode surface can transfer energy from the
excited triplet state exciton to the ground state oxygen molecule to
produce 1O2 with the help of visible light [45], or II) ⋅O2− can react with
⋅OH to form 1O2 through Eq. (8) [48]; III) ⋅OH undergoes a dispropor­
3.5. Optimization of key parameters in EF by treating real wastewater
tionation reaction could also produce 1O2 (Eq. (9)) [49].

O2 + e− → ⋅O2− (4) Considering pH is the core parameter, because it affects not only the
iron leaching but also the reactivity/precipitation of iron [50]. Hence, in
− +
2⋅O2 + 2H → H2O2 + O2 (5) this part, pH variance on wastewater degradation was investigated.

6
Y. Zhu et al. Chemical Engineering Journal 429 (2022) 132176

Fig. 7. Possible production mechanisms of active species.

Fig. 8. Effect of different initial pH on (a) STZ and (b) TOC removal. (c) k values for STZ and TOC removal with different pH. (d) EEM contour of samples with
different electrolysis time. Conditions: j = 10 mA cm− 2.

First, STZ decay along with TOC at a series of pH was investigated in exceeded the environmental standard set by the European Union (2 mg
Fig. 8a and b, where acidic conditions favored the removal of STZ and L− 1) (Fig. S15) [51].
TOC. The variance of pH during the whole electrolysis was monitored, in Fluorescence spectrum is an effective tool to evaluate the complexity
which pH dropped at the first 30 min and kept stable at 60 min of real wastewater, in this way, the effectiveness of the proposed EF
(Fig. S14a). It suggests that there was an excessive formation of H+ ions system was tested by the change of fluorescence [52]. Fig. 8d shows the
within 30 min because of the novel electrode arrangement: two parallel raw wastewater exhibits a humic-like fluorescence peak at Ex/Em of
anodes and three parallel cathodes. In this way, the current applied to a 350/440 nm [53]. This fluorescence was removed after 180 min treat­
single anode was higher than cathode, resulting in much more H+ ions ment at different pH values. It means the remaining organic matter was
generated on the anode than that of OH− ions from the cathode [23]. To almost non-fluorescent small molecules although the organic pollutants
prove this assumption, a conventional system with two parallel anodes were not completely mineralized as indicated in TOC results. It is worth
and two parallel cathodes were designed and the variance of pH was noting that both STZ and TOC removal rate at pH 5 was not as effective
compared to this proposed system. As expected, pH gradually increased, as that at pH 3, but it can still degrade most of macromolecular organics
as shown in Fig. S14b, because the consumption rate of H+ at cathode as shown in the fluorogram. In a word, pH value of 5 was selected in this
like ORR was much larger than its production rate. Hence, pH increased system for real wastewater treatment to avoid the addition of acid agent
by 1.45 after 180 min and it was not helpful for EF due to the iron and alleviate electrode corrosion.
precipitation. As concluded, the strategy to use more anode than cath­ Aside from pH estimation, the influence of current density was also
ode is a wise way to auto-adjust the pH of the bulk. Fig. 8c shows the investigated since it could regulate electron transfer [54]. Fig. 9a-b
first-order rate constant (k) of STZ and TOC removal at pH 3 are 0.0126 presents higher current density promotes both STZ and TOC degradation
± 0.0004 and 0.0069 ± 0.0002 min− 1, respectively, while it drops to when it ranged from 2 to15 mA cm− 2. Fig. 9c shows the kSTZ and kTOC
0.0046 ± 0.0001 and 0.0038 ± 0.0001 min− 1 at pH 9. Despite a higher reaches the highest value of 0.0129 ± 0.0006 and 0.0069 ± 0.0002
oxidation ability at pH 3, one concern is too much iron was released and min− 1 when current density was 15 mA cm− 2. One thing needed to pay

7
Y. Zhu et al. Chemical Engineering Journal 429 (2022) 132176

Fig. 9. Effect of current density on (a) STZ and (b) TOC removal. (c) k values for STZ and TOC removal and ECTOC with different current densities. (d) EEM contour of
samples. Conditions: pH = 5.

attention is that the enhancement of ECTOC (1.44-fold) was larger than the possible STZ pathway was proposed (Table S6). Fig. 10 shows STZ
kTOC (1.30-fold) as current density increased from 10 mA cm− 2 to 15 mA was degraded mainly through two pathways: I) the break of C-S bond, II)
cm− 2, showing kTOC value increased non-proportionally. It could be the cleavage of S-N bond. It is noteworthy that the degradation was also
explained in the following reasons: I) high current density promotes side accompanied by the hydroxylation and amino oxidation. The detected
reaction like the hydrogen evolution [55] and the cathodic polarization intermediates confirmed STZ was decayed into the smaller molecules,
[56]; II) high current density intensifies the Joule heat and then elec­ which is consistent with the EEM fluorescence results.
trical energy is wasted in the form of heat (Fig. S16) [57]. Moreover,
Fig. 9d represents the EEM fluorescence maps after 180 min of treatment
at different current densities. Current densities of 2 mA cm− 2 and 5 mA 3.6. Assessment of suitability on real wastewater treatment
cm− 2 failed to completely remove the fluorescent substances from the
wastewater, which is in accordance with the low STZ and TOC removal The strong oxidizing substances generated in reactor would attack
rate. Most of the fluorescent substances can be removed and converted the electrode, lowering the selectivity for H2O2 accumulation and
to non-fluorescent small molecules of organic matter when current shortening the lifespan of cathode. According to the analysis from sec­
density beyond 10 mA cm− 2. tion 3.5, the following key parameters were chosen to evaluate the
In addition, based on the intermediates detected by UPLC-QTOF-MS, durability of the proposed EF system: pH of 5 and a current density of 10
mA cm− 2. Fig. 11a shows negligible decrease in the removal rates of STZ

Fig. 10. Possible degradation pathway of STZ in the EF process.

8
Y. Zhu et al. Chemical Engineering Journal 429 (2022) 132176

Fig. 11. (a) The stability and Fe leaching using CNS-Fe-2 cathode. (b) the variance of BOD5/COD ratio during wastewater treatment. Conditions: pH = 5 and j = 10
mA cm− 2.

and TOC after six consecutives cycles, proving its excellent stability. In more cathodes than anodes. The fabricated CNS-Fe-2 ensured a high
addition, the dissolved Fe concentrations were always below 2 mg L− 1, H2O2 accumulation since its 2e− transfer ORR selectivity reached 90%
which meets the environmental standards set by the European Union along with its special microporous aeration improved DO concentration.
[57]. Moreover, the reusability of the electrode was confirmed through The main oxidizing species in this system were ⋅OH, 1O2, ⋅O2− and Fe
the stable surface roughness and mapping results of the electrode before (IV). Then the real pharmaceutical wastewater was used as an indicator
and after six cycles (Fig. S17). And the XPS analysis shows the content of to check its feasibility and the BOD5/COD ratio could be improved from
S and O slightly increased after electrolysis (Table S7). Aside from the 0.146 ± 0.002 to 0.350 ± 0.005 after 180 min. Therefore, it provides an
oxidation of the cathode by active species, it might be the introduction efficient and feasible method to further promote the application of EF as
of Na2SO4 electrolyte. In addition, besides STZ, other antibiotics like a pretreatment technique in practical engineering.
SDZ and SDM were detected in the wastewater. To confirm the potential
application for real pharmaceutical wastewater treatment, the removal
Declaration of Competing Interest
rate of SDZ and SDM were also investigated. Fig. S18 shows the SDZ and
SDM removal rate could reach 87.33 ± 1.95% and 70.43 ± 1.19% after
The authors declare that they have no known competing financial
EF treatment, illustrating its feasibility for removing antibiotics in
interests or personal relationships that could have appeared to influence
wastewater treatment.
the work reported in this paper.
The developed EF system was used as a pre-treatment means prior to
biological treatment for the actual wastewater treatment to save the cost
Acknowledgments
of the whole treatment. The BOD5/COD ratio is a crucial indicator for
biological treatment, and a value higher than 0.3 indicates biodegrad­
The research was supported by the National Natural Science Foun­
able [58]. Therefore, the variance of BOD5/COD during EF treatment
dation of China (No. 52070056, No. 52000052 and No. 62004143),
was measured. As shown in Fig. 11b, the refractory nature of this
National Major Science and Technology Projects of China
wastewater was proved by the low BOD5/COD ratio (0.146 ± 0.002).
(No.2019YFD1100204), State Key Laboratory of Urban Water Resource
However, it rose to 0.350 ± 0.005 after EF treatment. Thereafter, the
and Environment (No.2020TS04), Heilongjiang Province Post-doctoral
electrodes were reused five times and showed ignorable effect on
Grants (No.LBH-Z19063) and Natural Science Foundation of Hei­
biodegradable properties. In conclusion, the system is superior stable for
longjiang Province (No. LH2020E054). Moreover, the authors would
practical wastewater treatment as a pre-treatment technique.
like to thank the Shiyanjia Lab (www.shiyanjia.com) for the measure­
Operating costs are the key to the practical application of a tech­
ment of TEM.
nology. Therefore, the specific technical conditions and operating costs
for treating real pharmaceutical wastewater were listed in Table S8.
Electricity bill for electrolysis and aeration is the main cost in the elec­ Appendix A. Supplementary data
trochemical technology during operation. As a consequence, cost from
power supply and aeration was calculated in this work. In this paper, the Supplementary data to this article can be found online at https://doi.
electricity bill was calculated based on the Chinese industrial electricity org/10.1016/j.cej.2021.132176.
rate of 0.104 USD kWh− 1. As indicated in Table S8, a minimum oper­
ating cost of 1.82 US$ m− 3 (0.01 US$ g− 1 TOC) was seen in this proposed References
EF system, evidencing its great potential for practical applications in
[1] Y. Pan, Y. Zhang, M. Zhou, J. Cai, Y. Tian, Enhanced removal of antibiotics from
cost. secondary wastewater effluents by novel UV/pre-magnetized Fe0/H2O2 process,
Water Res. 153 (2019) 144–159, https://doi.org/10.1016/j.watres.2018.12.063.
4. Conclusions [2] Y. Ben, M. Hu, X. Zhang, S. Wu, M.H. Wong, M. Wang, C.B. Andrews, C. Zheng,
Efficient detection and assessment of human exposure to trace antibiotic residues
in drinking water, Water Res. 175 (2020) 115699, https://doi.org/10.1016/j.
A novel self-contained EF system was developed as a pretreatment watres.2020.115699.
process for actual pharmaceutical wastewater treatment with three ad­ [3] F. Deng, S. Li, Y. Cao, M.A. Fang, J. Qu, Z. Chen, S. Qiu, A dual-cathode pulsed
current electro-Fenton system: improvement for H2O2 accumulation and Fe3+
vantages: I) the reactivity/selectivity of ORR was boosted along with its reduction, J. Power Sources 466 (2020) 228342, https://doi.org/10.1016/j.
oxygen mass transfer by fabricating a new CNS-Fe-2 cathode and using jpowsour.2020.228342.
microporous aeration for micro-bubble supply, II) avoided the external [4] Z. Ye, J.A. Padilla, E. Xuriguera, J.L. Beltran, F. Alcaide, E. Brillas, I. Sirés, A highly
stable metal-organic framework-engineered FeS2/C nanocatalyst for heterogeneous
additional of iron catalyst and, III) automatic pH adjustment by using
electro-Fenton treatment: validation in wastewater at mild pH, Environ. Sci.

9
Y. Zhu et al. Chemical Engineering Journal 429 (2022) 132176

Technol. 54 (7) (2020) 4664–4674, https://doi.org/10.1021/acs. [27] Y. Zhao, M. Sun, X. Wang, C. Wang, D. Lu, W. Ma, S.A. Kube, J. Ma, M. Elimelech,
est.9b0760410.1021/acs.est.9b07604.s001. Janus electrocatalytic flow-through membrane enables highly selective singlet
[5] J. Qu, M.S. Akindolie, Y. Feng, Z. Jiang, G. Zhang, Q. Jiang, F. Deng, B.o. Cao, oxygen production, Nat. Commun. 11 (1) (2020), https://doi.org/10.1038/
Y. Zhang, One-pot hydrothermal synthesis of NaLa(CO3)2 decorated magnetic s41467-020-20071-w.
biochar for efficient phosphate removal from water: kinetics, isotherms, [28] J. Lim, H. Kim, J. Park, G.-H. Moon, J.J.M. Vequizo, A. Yamakata, J. Lee, W. Choi,
thermodynamics, mechanisms and reusability exploration, Chem. Eng. J. 394 How g-C3N4 works and is different from TiO2 as an environmental photocatalyst:
(2020) 124915, https://doi.org/10.1016/j.cej.2020.124915. mechanistic view, Environ. Sci. Technol. 54 (1) (2020) 497–506, https://doi.org/
[6] N. Barhoumi, N. Oturan, H. Olvera-Vargas, E. Brillas, A. Gadri, S. Ammar, M. 10.1021/acs.est.9b0504410.1021/acs.est.9b05044.s001.
A. Oturan, Pyrite as a sustainable catalyst in electro-Fenton process for improving [29] M. Park, J. Lee, K. Hembram, K.-R. Lee, S. Han, C. Yoon, S.-W. Nam, J. Kim,
oxidation of sulfamethazine kinetics, mechanism and toxicity assessment, Water Oxygen reduction electrocatalysts based on coupled iron nitride nanoparticles with
Res. 94 (2016) 52–61, https://doi.org/10.1016/j.watres.2016.02.042. nitrogen-doped carbon, Catalysts 6 (6) (2016) 86, https://doi.org/10.3390/
[7] H. Olvera-Vargas, X. Zheng, O. Garcia-Rodriguez, O. Lefebvre, Sequential catal6060086.
“electrochemical peroxidation–Electro-Fenton” process for anaerobic sludge [30] K. Liu, J.-C. Yu, H. Dong, J.C.S. Wu, M.R. Hoffmann, Degradation and
treatment, Water Res. 154 (2019) 277–286, https://doi.org/10.1016/j. mineralization of carbamazepine using an electro-Fenton reaction catalyzed by
watres.2019.01.063. magnetite nanoparticles fixed on an electrocatalytic carbon fiber textile cathode,
[8] L. Li, C. Tang, Y. Zheng, B. Xia, X. Zhou, H. Xu, S.-Z. Qiao, Tailoring selectivity of Environ. Sci. Technol. 52 (21) (2018) 12667–12674, https://doi.org/10.1021/acs.
electrochemical hydrogen peroxide generation by tunable pyrrolic-nitrogen- est.8b0391610.1021/acs.est.8b03916.s001.
carbon, Adv. Energy Mater. 10 (21) (2020) 2000789, https://doi.org/10.1002/ [31] D. Cheng, A. Neumann, S. Yuan, W. Liao, A.o. Qian, Oxidative degradation of
aenm.v10.2110.1002/aenm.202000789. organic contaminants by FeS in the presence of O2, Environ. Sci. Technol. 54 (7)
[9] E. Jung, H. Shin, B.-H. Lee, V. Efremov, S. Lee, H.S. Lee, J. Kim, W. Hooch Antink, (2020) 4091–4101, https://doi.org/10.1021/acs.est.9b07012.
S. Park, K.-S. Lee, S.-P. Cho, J.S. Yoo, Y.-E. Sung, T. Hyeon, Atomic-level tuning of [32] P. Yan, D. Kong, W. Yuan, In situ synthesis and electrocatalytic performance of Fe/
Co-N-C catalyst for high-performance electrochemical H2O2 production, Nat. Fe2.5C/Fe3N/nitrogen-doped carbon nanotubes for the oxygen reduction reaction,
Mater. 19 (4) (2020) 436–442, https://doi.org/10.1038/s41563-019-0571-5. Chemelectrochem 6 (2019) 3030–3038, https://doi.org/10.1002/celc.201900716.
[10] D. Guo, R. Shibuya, C. Akiba, S. Saji, T. Kondo, J. Nakamura, Active sites of [33] Y. Zhu, S. Qiu, F. Deng, Y. Zheng, K. Li, F. Ma, D. Liang, Enhanced degradation of
nitrogen-doped carbon materials for oxygen reduction reaction clarified using sulfathiazole by electro-Fenton process using a novel carbon nitride modified
model catalysts, Science 351 (6271) (2016) 361–365, https://doi.org/10.1126/ electrode, Carbon 145 (2019) 321–332, https://doi.org/10.1016/j.
science:aad0832. carbon.2019.01.032.
[11] O.Y. Bisen, R. Nandan, K.K. Nanda, Unique one-step strategy for nonmetallic and [34] Y. Yang, F. He, Y. Shen, X. Chen, H. Mei, S. Liu, Y. Zhang, A biomass derived N/C-
metallic heteroatom doped carbonaceous materials, ACS Omega 5 (51) (2020) catalyst for the electrochemical production of hydrogen peroxide, Chem. Commun.
32852–32860, https://doi.org/10.1021/acsomega.0c04432. 53 (72) (2017) 9994–9997, https://doi.org/10.1039/C7CC04819J.
[12] X. Zhang, Y. Xia, C. Xia, H. Wang, Insights into practical-scale electrochemical [35] Y. Yang, The properties of PdRu/C with respect to the electro-oxidation of
H2O2 synthesis, Trends Chem. 2 (10) (2020) 942–953, https://doi.org/10.1016/j. methanol and ethanol, Int. J. Electrochem. Sc 14 (2019) 1270–1282, https://doi.
trechm.2020.07.007. org/10.20964/2019.02.50.
[13] Y. Zhu, F. Deng, S. Qiu, F. Ma, Y. Zheng, R. Lian, Enhanced electro-Fenton [36] M. Li, X. Xi, Q. Liu, Z. Nie, L. Ma, Direct electrolytic separation of tungsten and
degradation of sulfonamides using the N, S co-doped cathode: Mechanism for H2O2 cobalt from waste cemented carbide and electrochemical behavior of tungsten and
formation and pollutants decay, J. Hazard. Mater. 403 (2021) 123950, https://doi. cobalt ions in NaF-KF molten salts, J. Electroanal. Chem. 833 (2019) 480–489,
org/10.1016/j.jhazmat.2020.123950. https://doi.org/10.1016/j.jelechem.2018.12.032.
[14] Y. Wang, X. Duan, Y. Xie, H. Sun, S. Wang, Nanocarbon-based catalytic ozonation [37] J. Yu, Y. Zhong, W. Zhou, Z. Shao, Facile synthesis of nitrogen-doped carbon
for aqueous oxidation: engineering defects for active sites and tunable reaction nanotubes encapsulating nickel cobalt alloys 3D networks for oxygen evolution
pathways, ACS Catal. 10 (22) (2020) 13383–13414, https://doi.org/10.1021/ reaction in an alkaline solution, J. Power Sources 338 (2017) 26–33, https://doi.
acscatal.0c04232. org/10.1016/j.jpowsour.2016.11.023.
[15] J.F. Pérez, J. Llanos, C. Sáez, C. López, P. Cañizares, M.A. Rodrigo, Electrochemical [38] X. Zhou, D. Xu, Y. Chen, Y. Hu, Enhanced degradation of triclosan in
jet-cell for the in-situ generation of hydrogen peroxide, Electrochem. Commun. 71 heterogeneous E-Fenton process with MOF-derived hierarchical Mn/Fe@PC
(2016) 65–68, https://doi.org/10.1016/j.elecom.2016.08.007. modified cathode, Chem. Eng. J. 384 (2020) 123324, https://doi.org/10.1016/j.
[16] I. Yamanaka, T. Onizawa, S. Takenaka, K. Otsuka, Direct and continuous cej.2019.123324.
production of hydrogen peroxide with 93 % selectivity using a fuel-cell system, [39] X. Shen, F. Xiao, H. Zhao, Y. Chen, C. Fang, R. Xiao, W. Chu, G. Zhao, In situ-
Angew. Chem. 115 (31) (2003) 3781–3783, https://doi.org/10.1002/(ISSN)1521- formed PdFe nanoalloy and carbon defects in cathode for synergic reduction-
375710.1002/ange.v115:3110.1002/ange.200351343. oxidation of chlorinated pollutants in electro-Fenton process, Environ. Sci.
[17] W. Zhou, X. Meng, L. Rajic, Y. Xue, S. Chen, Y. Ding, K. Kou, Y. Wang, J. Gao, Technol. 54 (7) (2020) 4564–4572, https://doi.org/10.1021/acs.est.9b05896.
Y. Qin, A.N. Alshawabkeh, “Floating” cathode for efficient H2O2 electrogeneration [40] D. Li, T. Zheng, Y. Liu, D. Hou, K.K. Yao, W. Zhang, H. Song, H. He, W. Shi, L.
applied to degradation of ibuprofen as a model pollutant, Electrochem. Commun. u. Wang, J. Ma, A novel Electro-Fenton process characterized by aeration from
96 (2018) 37–41. inside a graphite felt electrode with enhanced electrogeneration of H2O2 and cycle
[18] D. Ohde, B. Thomas, S. Matthes, S. Tanaka, P. Bubenheim, K. Terasaka, of Fe3+/Fe2+, J. Hazard. Mater. 396 (2020) 122591, https://doi.org/10.1016/j.
M. Schlüter, A. Liese, Microbubble enhanced mass transfer efficiency of CO2 jhazmat.2020.122591.
capture utilizing aqueous triethanolamine for enzymatic resorcinol carboxylation, [41] K.G. Clarke, in: Bioprocess engineering, Publishing, Woodhead, 2013,
RSC Adv. 11 (7) (2021) 4087–4096, https://doi.org/10.1039/D0RA08690H. pp. 147–170.
[19] I. Sirés, E. Brillas, in: Electro-Fenton process: New trends and scale-up, Springer [42] A.C. Velosa, C.A.O. Nascimento, Evaluation of sulfathiazole degradation by
Singapore, Singapore, 2018, pp. 1–28. persulfate in Milli-Q water and in effluent of a sewage treatment plant, Environ.
[20] S.O. Ganiyu, T.X. Huong Le, M. Bechelany, G. Esposito, E.D. van Hullebusch, M. Sci. Pollut. R. 24 (7) (2017) 6270–6277, https://doi.org/10.1007/s11356-016-
A. Oturan, M. Cretin, A hierarchical CoFe-layered double hydroxide modified 7036-z.
carbon-felt cathode for heterogeneous electro-Fenton process, J. Mater. Chem. a 5 [43] Z. Wang, J. Jiang, S. Pang, Y. Zhou, C. Guan, Y. Gao, J. Li, Y.i. Yang, W. Qiu,
5 (7) (2017) 3655–3666, https://doi.org/10.1039/C6TA09100H. C. Jiang, Is sulfate radical really generated from peroxydisulfate activated by iron
[21] S. Yuan, Y.e. Fan, Y. Zhang, M. Tong, P. Liao, Pd-catalytic in situ generation of (II) for environmental decontamination? Environ. Sci. Technol. 52 (19) (2018)
H2O2 from H2 and O2 produced by water electrolysis for the efficient electro- 11276–11284, https://doi.org/10.1021/acs.est.8b02266.
Fenton degradation of rhodamine B, Environ. Sci. Technol. 45 (19) (2011) [44] X. Li, C. Xiao, X. Ruan, Y. Hu, C. Zhang, J. Cheng, Y. Chen, Enrofloxacin
8514–8520, https://doi.org/10.1021/es2022939. degradation in a heterogeneous electro-Fenton system using a tri-metal-carbon
[22] C. Zhang, M. Zhou, G. Ren, X. Yu, L. Ma, J. Yang, F. Yu, Heterogeneous electro- nanofibers composite cathode, Chem. Eng. J. 427 (2022) 130927, https://doi.org/
Fenton using modified iron–carbon as catalyst for 2,4-dichlorophenol degradation: 10.1016/j.cej.2021.130927.
influence factors, mechanism and degradation pathway, Water Res. 70 (2015) [45] Y. Zhu, S. Qiu, F. Deng, F. Ma, Y. Zheng, Degradation of sulfathiazole by electro-
414–424, https://doi.org/10.1016/j.watres.2014.12.022. Fenton using a nitrogen-doped cathode and a BDD anode: Insight into the H2O2
[23] S. Yuan, M. Chen, X. Mao, A.N. Alshawabkeh, A three-electrode column for Pd- generation and radical oxidation, Sci. Total Environ. 722 (2020) 137853, https://
catalytic oxidation of TCE in groundwater with automatic pH-regulation and doi.org/10.1016/j.scitotenv.2020.137853.
resistance to reduced sulfur compound foiling, Water Res. 47 (1) (2013) 269–278, [46] S. Liang, L. Zhu, J. Hua, W. Duan, P.-T. Yang, S.-L. Wang, C. Wei, C. Liu, C. Feng,
https://doi.org/10.1016/j.watres.2012.10.009. Fe2+/HClO reaction produces FeIVO2+: an enhanced advanced oxidation process,
[24] Y. Zheng, S. Qiu, F. Deng, Y. Zhu, G. Li, F. Ma, Three-dimensional electro-Fenton Environ. Sci. Technol. 54 (10) (2020) 6406–6414, https://doi.org/10.1021/acs.
system with iron foam as particle electrode for folic acid wastewater pretreatment, est.0c00218.
Sep. Purif. Technol. 224 (2019) 463–474, https://doi.org/10.1016/j. [47] K. Cheaib, M.Q.E. Mubarak, K. Sénéchal-David, C. Herrero, R. Guillot,
seppur.2019.05.054. M. Clémancey, J.-M. Latour, S.P. deVisser, J.-P. Mahy, F. Banse, F. Avenier,
[25] I. Hua, M.R. Hoffmann, Optimization of ultrasonic irradiation as an advanced Selective formation of an FeIVO or an FeIIIOOH intermediate from iron(II) and
oxidation technology, Environ. Sci. Technol. 31 (8) (1997) 2237–2243, https:// H2O2: Controlled heterolytic versus homolytic oxygen-oxygen bond cleavage by
doi.org/10.1021/es960717f. the second coordination sphere, Angew. Chem. Int. Edit. 58 (3) (2019) 854–858,
[26] Y.i. Yang, G. Banerjee, G.W. Brudvig, J.-H. Kim, J.J. Pignatello, Oxidation of https://doi.org/10.1002/anie.201812724.
organic compounds in water by unactivated peroxymonosulfate, Environ. Sci. [48] X. Li, J. Liu, A.I. Rykov, H. Han, C. Jin, X. Liu, J. Wang, Excellent photo-Fenton
Technol. 52 (10) (2018) 5911–5919, https://doi.org/10.1021/acs. catalysts of Fe-Co Prussian blue analogues and their reaction mechanism study,
est.8b0073510.1021/acs.est.8b00735.s001. Appl. Catal. B-Environ. 179 (2015) 196–205, https://doi.org/10.1016/j.
apcatb.2015.05.033.

10
Y. Zhu et al. Chemical Engineering Journal 429 (2022) 132176

[49] Q. Yi, J. Ji, B. Shen, C. Dong, J. Liu, J. Zhang, M. Xing, Singlet oxygen triggered by [54] Q. Zhang, M. Zhou, G. Ren, Y. Li, Y. Li, X. Du, Highly efficient electrosynthesis of
superoxide radicals in a molybdenum cocatalytic Fenton reaction with enhanced hydrogen peroxide on a superhydrophobic three-phase interface by natural air
REDOX activity in the environment, Environ. Sci. Technol. 53 (16) (2019) diffusion, Nat. Commun. 11 (1) (2020), https://doi.org/10.1038/s41467-020-
9725–9733, https://doi.org/10.1021/acs.est.9b01676. 15597-y.
[50] J. He, X. Yang, B. Men, D. Wang, Interfacial mechanisms of heterogeneous Fenton [55] F. Sopaj, N. Oturan, J. Pinson, F. Podvorica, M.A. Oturan, Effect of the anode
reactions catalyzed by iron-based materials: a review, J. Environ. Sci.-China 39 materials on the efficiency of the electro-Fenton process for the mineralization of
(2016) 97–109, https://doi.org/10.1016/j.jes.2015.12.003. the antibiotic sulfamethazine, Appl. Catal. B-Environ. 199 (2016) 331–341,
[51] P. Gao, Y. Song, M. Hao, A. Zhu, H. Yang, S. Yang, An effective and magnetic https://doi.org/10.1016/j.apcatb.2016.06.035.
Fe2O3-ZrO2 catalyst for phenol degradation under neutral pH in the heterogeneous [56] Y. Sheng, Y. Zhao, X. Wang, R. Wang, T. Tang, Electrogeneration of H2O2 on a
Fenton-like reaction, Sep. Purif. Technol. 201 (2018) 238–243, https://doi.org/ composite acetylene black–PTFE cathode consisting of a sheet active core and a
10.1016/j.seppur.2018.03.017. dampproof coating, Electrochim. Acta 133 (2014) 414–421, https://doi.org/
[52] Y. Zheng, S. Qiu, F. Deng, Y. Zhu, F. Ma, G. Li, A charcoal-shaped catalyst NiFe2O4/ 10.1016/j.electacta.2014.04.071.
Fe2O3 in electro-Fenton: high activity, wide pH range and catalytic mechanism, [57] S. Pei, J. Teng, N. Ren, S. You, Low-temperature removal of refractory organic
Environ. Technol. 42 (13) (2021) 1996–2008, https://doi.org/10.1080/ pollutants by electrochemical oxidation: role of interfacial joule heating effect,
09593330.2019.1687586. Environ. Sci. Technol. 54 (7) (2020) 4573–4582, https://doi.org/10.1021/acs.
[53] Z. Ye, H. Zhang, X. Zhang, D. Zhou, Treatment of landfill leachate using est.9b05929.
electrochemically assisted UV/chlorine process: effect of operating conditions, [58] S. SUAREZ, J. LEMA, F. OMIL, Pre-treatment of hospital wastewater by
molecular weight distribution and fluorescence EEM-PARAFAC analysis, Chem. coagulation-flocculation and flotation, Bioresource Technol. 100 (7) (2009)
Eng. J. 286 (2016) 508–516, https://doi.org/10.1016/j.cej.2015.10.017. 2138–2146, https://doi.org/10.1016/j.biortech.2008.11.015.

11