Electro-Fenton and Induced Electro-Fenton as Versatile Wastewater Treatment Processes for Decontamination and Nutrient Removal without Byproduct Formation

It is a long-pursued goal to develop electrified water treatment technology that can remove contaminants without byproduct formation. This study unveiled the overlooked multifunctionality of electro-Fenton (EF) and induced EF (I-EF) processes to remove organics, pathogens, and phosphate in one step without halogenated byproduct formation. The EF and I-EF processes used a sacrificial anode or an induced electrode to generate Fe2+ to activate H2O2 produced from a gas diffusion cathode fed by naturally diffused air. We used experimental and kinetic modeling approaches to illustrate that the •OH generation and radical speciation during EF were not impacted by chloride. More importantly, reactive chlorine species were quenched by H2O2, which eliminated the formation of halogenated byproducts. When applied in treating septic wastewater, the EF process removed >80% COD, >50% carbamazepine (as representative trace organics), and >99% phosphate at a low energy consumption of 0.37 Wh/L. The EF process also demonstrated broad-spectrum disinfection activities in removing and inactivating Escherichia coli, Enterococcus durans, and model viruses MS2 and Phi6. In contrast to electrochemical oxidation (EO) that yielded mg/L level byproducts to achieve the same degree of treatment, EF did not generate byproducts (chlorate, perchlorate, trihalomethanes, and haloacetic acids). The I-EF carried over all the advantages of EF and exhibited even faster kinetics in disinfection and carbamazepine removal with 50–80% less sludge production. Last, using septic wastewater treatment as a technical niche, we demonstrated that iron sludge formation is predictable and manageable, clearing roadblocks toward on-site water treatment applications.


Text S1. The calculation for COD removal by coagulation and oxidation
The COD removed by coagulation and oxidation was calculated based on the procedure and the equations below (Eq.S1 and S2), developed by Han et  where the mixed liquor is defined as the Fe(OH) x -containing treated water.The supernatant was taken after centrifugal separation of Fe(OH) x .

Text S2. Benzoic acid analysis
The benzoic acid analysis was performed using an ultra-high performance liquid chromatography (UHPLC) system (ExionLC 2.0+) coupled with a quadrupole time-of-flight mass spectrometer (QToF-MS) SCIEX 5600 Model X500B, USA).Separation was performed with an Atlantis® HILIC Silica column (3 µm, 2.1 × 100 mm).A linear gradient solvent program was operated with 5 mM ammonium acetate in LC-MS grade water (A) and acetonitrile (B).The gradient started from 90% B for a 1-min hold, then decreased to 50% over 1 min, and further decreased to 10% B from 2 to 2.4 min.The electrospray ionization (ESI) of the mass spectrometer was operated in the negative ion mode with a spray voltage of +4.5 kV and a capillary temperature of 350 °C.The collision energy was at 35 V, the declustering potential was at 60 V, and the default values were used for other parameters.Quantification was performed using multiple reaction monitoring (MRM), with MRM transition at m/z 50-1000 Da targeting the ionized compound (121.12Da).The injection volume was set at 10 µL, and the oven temperature was fixed at 40 ℃.
Data were acquired from Analyst 1.7 software and analyzed by PeakViews (AB SCIEX).

Text S3. Carbamazepine analysis
The analysis was performed using a UHPLC system (ExionLC 2.0+) coupled with a QToF-MS (SCIEX 5600 Model X500B, USA) operated in the electrospray ionization mode.Separation was performed with a C18 column (Gemini® 3 µm C18, 110 Å, 50 x 2 mm) purchased from Phenomenex (USA) at 40 ºC.An 11-min gradient elution, with a flow rate of 0.6 mL/min and an injection volume of 10 µL.The mobile phases of methanol and water were amended with 0.1% formic acid.

S3
The data was acquired from Analyst 1.7 software using the high-resolution multiple reaction monitoring (HR-MRM) mode.The electrospray ionization (ESI) source conditions are as follows: spray voltage at + 5.5 kV, the capillary temperature at 350 °C, probe heater temperature at 430 °C, nebulize gas at 55 (GS2, arbitrary units), sheath gas at 50 (GS2, arbitrary units), curtain gas at 40 (arbitrary units), and sweep gas at 1 (arbitrary units), collision energy at 35 V, declustering potential at 100 V, other parameters were set by default.

Preparation of bacterial stocks
Escherichia coli (E. coli K-12 strain, ATCC 10798) and Enterococcus durans (E.durans, ATCC 6056) were selected as model gram-negative and gram-positive bacteria, respectively.The bacterial cultures were incubated overnight in Luria-Bertani (LB) Broth for E. coli and Brain Heart Infusion (BHI) Broth for E. durans at 37 °C in a shaking incubator at 180 rpm.Bacteria in the stationary phase were harvested by centrifugation at 5000 g for 10 minutes, and the bacterial × pellet was resuspended in phosphate-buffered saline (1 PBS).The fresh bacterial stock was × prepared for every experiment.The target concentration was 10 5 -10 6 CFU/mL based on the culturable bacteria in the raw septic water (Table S1).The detection limit of 10 CFU/mL was determined following Sutton's method (2011). 2

Preparation of viral stocks
Two bacteriophages were used as surrogates of pathogenic non-enveloped and enveloped human viruses: MS2 (ssRNA, non-enveloped, ATCC 15597-B1) and Phi6 (dsRNA, enveloped, provided by Dr. Ye Y. from University at Buffalo).LB agar and broth were used for MS2 propagation and its host (Escherichia coli C3000 [ATCC 15597]) cultivation at 37 °C.Tryptic soy agar (TSA) and broth were used for Phi6 propagation and its host (Pseudomonas psyringae) cultivation at 25 °C.
Liquid virus propagation and purification were performed for both viruses following Ye's protocols. 3The supernatant, after centrifugation at 5000 g for 10 min, was recovered and × filtered.The filtrate was diluted 1000-fold after purification with a 0.2 µm syringe filter to obtain the final viral stock using 1 PBS.The seeding level of both microorganisms in sterilized septic water × was 10 5 -10 6 PFU/mL, a typical value adopted by other disinfection studies. 4cterial and viral inactivation and removal

S4
Bacteria and viruses were inoculated individually in the sterilized water in the electrochemical reactors.They were constantly mixed by a magnetic stirrer for 30 min before treatment.
The colony-counting and plaque assay were used to quantify viable bacteria and viruses, respectively, after overnight culture (16 ± 2 h).
The inactivation of bacteria and viruses from the treated septic water via coagulation or oxidation mechanisms is expressed in log reduction calculated by equations S3 and S4.Mixed liquor refers to iron sludge-containing treated water without separation; the supernatant was obtained after 1 min centrifugation at 3500 rpm.

)
The results were expressed by Log 10 (N/N 0 ).The number of colonies is expressed on colony forming units (CFU) per mL of sample.As for virus analysis, the number of plaques is expressed in plaque-forming units (PFU) per mL.

Text S5. Theoretical mass loss of iron from LCS and Faradaic efficiency calculation
We assume that iron on LCS was oxidized and released as Fe 2+ (Fe → Fe 2+ + 2e -).The theoretical mass loss of sacrificial LCS anode (m Fe ) in the EF processes was calculated based on Faraday's laws of electrolysis: (Eq.S5) m Fe, theoretical = where I is the applied current (A), t is the reaction time (s), M is the molar mass of the iron metal

×
-a : Culturable bacteria in raw septic tank wastewater were evaluated by cultivation in LB broth followed by colony counting.The procedure is the same as E.coli quantification described in Text S4.
b :"<" indicates the readings were below the detection limits.
Table S2.Reactions and rate constants involved in the EF reactions.

Rxn
No.  Electrochemical chlorine evolution

ROS generation
Fitted value at 30 mA b a Rxn S27 stands for the unknown HO• quenching pathways, in addition to the pathways in which •OH reacts with HO•, H 2 O 2 , chlorine, and Cl -.The rate constant of rxn 27 was arbitrarily set as a value equal to or smaller than rxn 43, as we believe the unknown quenching reactions should not be faster than the homogeneous reaction between HO• and H 2 O 2 .
b Rxn S47 refers to the electrochemical oxidation of Cl -to OCl -by IrOx anode in the HPP mode following the higher oxide oxidation pathway. 20The rate constant was obtained by fitting the experimental data in Figure S3a.The data set denoted by "FIT" means the data were fed to the kinetic models to calibrate specific rate constants; those tagged as "SIM" are results predicted by the calibrated kinetic models without manual intervention.Lines of "EF_SIM" and "EF w/Cl -_SIM" overlapped because the model simulation concluded that Cl -did not impact H 2 O 2 evolution.

(
55.8 g/mol), and z = 2 is the number of electrons to oxidize Fe to Fe 2+ .The Faraday constant (F) is 96,485 C/mol.The Faradaic efficiency can then be calculated below (Eq.S6) Faradaic efficiency = mFe, measured mFe, theoretical × 100% where m Fe, measured is the mass loss of LCS determined by the weight difference before and after EF reactions.Text S6.Iron sludge quantification 10 mL of treated mixed liquor generated from EF and I-EF in 10 mM NaClO 4 electrolyte was collected and filtrated using PTFE 0.45 μm membranes.The initial weight of the filters was then recorded.The filters with wet sludge were placed in the oven at 90 ºC overnight until the water was evaporated.The dry weight of the filter, together with the dry sludge, was recorded.The evaluation was performed at least in duplicates.The total sludge production was calculated as follows:(Eq.S7)

Figure S4 .
Figure S4.Hydroxyl radical concentrations as a function of EF reaction time predicted by the model simulation at (a) 30 mA and (b) 60 mA.

Figure S5 .
Figure S5.The radical speciation when 1.8 mM Cl -in the EF reaction predicted by the model excluding Rxn S44.

Figure S6 .Figure S7 .Figure S8 .
Figure S6.COD removal by EO using a NATO anode paired with a SS cathode.A current of 30 mA was applied on 6 cm 2 electrodes to treat 60 mL septic wastewater.

Table S1 .
Septic wastewater composition before and after EF or EC treatment.