Soluble chemical oxygen demand removal from bypass wastewater using iron electrocoagulation

Soluble chemical oxygen demand removal from bypass wastewater using iron electrocoagulation

Journal Pre-proof Soluble chemical oxygen demand removal from bypass wastewater using iron electrocoagulation Haitham Elnakar, Ian Buchanan PII: S00...

5MB Sizes 0 Downloads 102 Views

Journal Pre-proof Soluble chemical oxygen demand removal from bypass wastewater using iron electrocoagulation

Haitham Elnakar, Ian Buchanan PII:

S0048-9697(19)36072-3

DOI:

https://doi.org/10.1016/j.scitotenv.2019.136076

Reference:

STOTEN 136076

To appear in:

Science of the Total Environment

Received date:

23 August 2019

Revised date:

25 November 2019

Accepted date:

9 December 2019

Please cite this article as: H. Elnakar and I. Buchanan, Soluble chemical oxygen demand removal from bypass wastewater using iron electrocoagulation, Science of the Total Environment (2019), https://doi.org/10.1016/j.scitotenv.2019.136076

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

© 2019 Published by Elsevier.

Journal Pre-proof

Soluble chemical oxygen demand removal from bypass wastewater using iron electrocoagulation

Jo

ur

na

lP

re

-p

ro

of

Haitham Elnakar1* and Ian Buchanan1

1

Department of Civil and Environmental Engineering, University of Alberta, 9211 116 St. NW Edmonton, Alberta, Canada, T6G 1H9.

* Corresponding author: 1-780-952-8739. [email protected]

1

Journal Pre-proof Abstract In-plant wastewater treatment strategies to handle bypass wastewater exceeding design capacity are insufficiently investigated in the scientific literature notwithstanding their importance in ensuring sustainable wastewater management. In this study, the effectiveness of iron electrocoagulation was investigated, for the first time, to enhance primary treatment capability in removing soluble chemical oxygen demand (sCOD) from bypass wastewater. In addition, the appropriate assumptions and

of

experimental protocols for the application of adsorption isotherm models, widely used to describe the

ro

electrocoagulation process, were discussed in light of experimental results. Under neutral pH conditions, the bypass wastewater treatment was performed to test the effects of three preselected variables

-p

(electrolysis duration, current density, and temperature) on sCOD removal. Using a 15 mA/cm2 current

re

density, an average 52% sCOD removal efficiency was achieved after 15 minutes at 23 oC while approximately 40 minutes were needed to attain comparable removal efficiency at 8 oC. sCOD removals

lP

of 74% and 87% were achieved after 40 minutes treatment using a 22 mA/cm2 current density at 8 oC and

na

23 oC, respectively. Experimental results and theory show that adsorption equilibrium was not reached in the electrocoagulation cell; consequently, variable-order-kinetic (VOK) models derived from Langmuir

ur

and Langmuir-Freundlich adsorption expressions were adapted to describe the process. These models

Jo

were modified to account for the de facto estimation of ferric hydroxide (adsorbent) mass that accounts for the conversion of ferrous ion to particulate end products. The Langmuir-based VOK model was found to better describe sCOD removal under all the operating conditions tested and showed the sCOD removal mechanism to be consistent with chemisorption. This research shows the promising ability of iron electrocoagulation to achieve superior removal of sCOD as compared to established and emerging standalone bypass wastewater treatment technologies. KEYWORDS: bypass wastewater; enhanced primary treatment; iron electrocoagulation; COD removal; adsorption isotherms; VOK model

2

Journal Pre-proof Nomenclature

Akaike Information Criterion

Co

Initial concentration of the sCOD

Ct

Aqueous adsorbate concentration achieved after time t of iron electrocoagulation

F

Faraday’s constant

I

Current

KL

Langmuir constant

kLF

Langmuir–Freundlich constant

n

Langmuir–Freundlich exponent

qLmax

Maximum adsorption capacity per unit mass of adsorbent

qt

Amount of sCOD adsorbed per gram of the Fe(OH)3(s) precipitate at time t

sCOD

Soluble chemical oxygen demand

t

Time

V

Batch reactor volume

VOK

Variable-order-kinetic

Wt

Mass of the Fe(OH)3(s) precipitate at time t

Z

Charge transfer number

Φ

Efficiency of iron electrocoagulation the process in forming Fe(OH)3

Jo

ur

na

lP

re

-p

ro

of

AICc

3

Journal Pre-proof 1. Introduction It is estimated that globally 74% of existing urban and 66% of existing rural wastewater services do not completely prevent human contact with untreated excreta (UNESCO, 2017). Several of those urban and rural communities are served by combined sewer systems which convey both sanitary sewage and stormwater runoff in the same sewer pipes to treatment plants. During heavy wet weather events caused by rainfall or snowmelt, the flow may exceed the hydraulic design capacity of the collection system or

of

wastewater treatment plant. This results in sewer overflows or treatment plant bypasses, both of which

ro

release untreated combined sewerage to surface waters (Risch et al., 2018; U.S. EPA, 2013, 2008). Efforts to manage the risks of combined sewer overflows have evolved over the past several decades (Risch et

-p

al., 2018; U.S. EPA, 1999a, 2013). The most straightforward way to eliminate the existing combined

re

sewer system problem would be to install an entirely separate sewer system (U.S. EPA, 1999b). However,

lP

this solution is usually not economically feasible. As a result, gradual sewer separation along with treatment of excess flows is the most viable solution to deal with this problem.

na

There is an immediate need for technological development to leverage the existing wastewater treatment infrastructure and to provide more effective options to the newly built treatment systems to handle bypass

ur

wastewaters (Hart and Halden, 2019; Tram VO et al., 2014). Among these technological options is

Jo

physically and/or chemically enhanced primary sedimentation which can be used as stand-alone auxiliary treatment for bypass wastewater flows that exceed treatment plant capacity (City of Edmonton, 2000; Elnakar and Buchanan, 2019a, 2019b). Primary sedimentation tanks can be enhanced using physical means such as lamella plate settlers, or by chemical means such as chemical coagulation. Electrochemical processes such as electrolysis, electrocoagulation, and electroflotation are also attractive alternative enhancement techniques for the primary sedimentation process. Notably, iron electrocoagulation has been used to treat a wide range of pollutants in industrial wastewater (Ahmadzadeh and Dolatabadi, 2018a; GilPavas et al., 2019) and groundwater (Ahmadzadeh and Dolatabadi, 2018b; Müller et al., 2019); however, little attention has been given to its use in municipal or bypass wastewater treatment (Nguyen et

4

Journal Pre-proof al., 2016). Such enhancements would allow primary sedimentation tanks to be designed or retrofitted to increase removal efficiencies to meet stricter standards or to accommodate higher overflow rates at reduced capital cost (Harleman and Murcott, 1999; Metcalf & Eddy Inc., 2003). In addition, very few reports exist in the literature concerning the enhanced primary treatment of low-temperature municipal or bypass wastewater in general or of its treatment by iron electrocoagulation. Several technologies have been tested for the treatment of combined sewer overflows or bypass

of

wastewater. These include the use of wetlands, hydraulic separators, high rate ballasted clarification, and

ro

enhanced primary treatment using metal salts (Chhetri et al., 2016; Masi et al., 2017; U.S. EPA, 2013). Most of these technologies achieved good suspended solids removal but little removal of soluble organic

-p

compounds (Chhetri et al., 2016; Masi et al., 2017; U.S. EPA, 2013). Iron electrocoagulation, on the other

re

hand, has been shown to achieve excellent removals of both particulates and soluble organic compounds

lP

from various wastewater matrices (Brar et al., 2009; Yılmaz Nayır and Kara, 2018; Yoosefian et al., 2017). Aqueous iron species generated electrochemically during iron electrocoagulation form precipitates

na

which become involved in such physio-chemical processes as: 1) inclusion, where a contaminant occupies a cavity within the precipitated iron species floc; 2) occlusion, where a floc particle entirely surrounds a

ur

contaminant, trapping it in the floc structure, so it cannot return to the solution; or 3) adsorption, where

Jo

contaminants adhere to the surface of a floc particle (Mollah et al., 2001). Some researchers have applied adsorption isotherms for iron electrocoagulation by assuming all dissolved iron, predicted by Faraday’s law, is immediately converted to Fe(OH)3(s) precipitate capable of adsorbing the organic molecules (Ahmadzadeh and Dolatabadi, 2018c; Kalyani et al., 2009; Yoosefian et al., 2017). During the electrocoagulation process, there is a continuous production of ferrous ions (Fe2+) and their ongoing reaction to form precipitates (Müller et al., 2019). Consequently, it is an inherently nonequilibrium process and data used to calibrate isotherm models must be collected in a manner that is consistent with the dynamic equilibrium underpinnings of the various models. A prudent approach would be to perform adsorption isotherm data collection in several discrete steps similar to the approach

5

Journal Pre-proof described by Essadki et al. (2010). In this approach, the adsorbent (hydroxide precipitate) is first produced from the electrocoagulation process in the absence of adsorbate. The adsorbent is then recovered, dried, and finally used in a separate set of isotherm data collection experiments. Many isotherm studies reported in the literature suffer from one or more deficiencies; the most basic being a lack of well-specified experimental procedure. In some studies, samples for isotherm modeling have been collected from electrocoagulation cells while the DC was applied, as one would collect data during a kinetics

of

investigation. Even if samples were collected after the DC supply was stopped, the adsorbent concentration may continue to increase if aqueous Fe2+ continues to be converted to ferric (Fe3+) and then

ro

to iron hydroxides (Fe(OH)3(s)). These approaches cannot ensure that equilibrium conditions are reached.

-p

Therefore, a multi-step approach that separates the production of adsorbent from its use in isotherm data

re

collection is needed. Also, while isotherm analyses may reveal valuable information regarding the interactions between adsorbent and adsorbate that occur within these cells, models that take into account

lP

the process kinetics should be used to model the removal process itself.

na

The two primary goals of this research are to (1) test, for the first time, the use of iron electrocoagulation as an enhancement to primary sedimentation for the removal of soluble chemical oxygen demand (sCOD)

ur

from bypass wastewaters; and (2) discuss and develop heuristic procedures to obtain data for the

Jo

calibration of adsorption kinetics models for sCOD removal based on more appropriate data collection methods and science-based estimation of the adsorbent concentrations produced during iron electrocoagulation.

2. Materials and methods 2.1 Iron electrocoagulation treatment Bypass wastewater samples were collected from a grit removal process effluent channel at a wastewater treatment plant in central Alberta, Canada. Samples were specifically collected during wet weather conditions. At this time, wastewater in excess of the plant capacity was only receiving primary treatment before being discharged to the river, bypassing the downstream biological treatment processes. Samples 6

Journal Pre-proof were stored at 4 oC. Two sets of experiments were conducted in a temperature-controlled laboratory; one at 23.0 oC ± 0.5 oC to represent the highest temperature that could be reasonably anticipated for raw influent wastewater, and the other at 8.0 oC ± 0.2 oC to represent the typical winter raw influent temperature in central Alberta. A 100-mL glass beaker was used as the electrochemical cell with a 5-cm long and 1.2-cm in diameter iron anode (99.99% as Fe, VWR). In order to prevent the possibility of further dissolution from the cathode, a

of

303 – stainless-steel cathode was used. A 50-mL sample size was used in all experiments. The effective

ro

submerged area of the electrodes immersed in the solution was 7.5 cm2. A 1.5 cm interelectrode distance was used in all experiments. A magnetic stirrer was placed below the middle of the cell. Samples were

-p

stirred for 5 minutes without electric current to ensure the homogeneity of the sample. The velocity of the

re

stirrer in all experiments was approximately 500 rpm, which provided vigorous mixing in the cell. The

lP

electric current was supplied by a DC-regulated power source (BK Precision 1685B 1-50 V DC and 5 A). A digital multimeter (DMiOTECH‐ UA92015N, China) was used to measure the voltage and current.

na

Three current densities (i) of 8, 15 and 22 mA/cm2 were selected in this study and were applied for various durations ranging from 5 to 40 minutes, where each treatment duration represents an independent

ur

experiment. The DC-regulated power source and stirring were stopped at the designated treatment

Jo

duration, and a 30-minute flotation period started. Samples were then collected for measurements from the clear layer under the surface froth layer. The iron electrode was cleaned after each experiment in a diluted HCl solution (5% v/v).

2.2 Analytical methods The bypass wastewater samples’ characteristics were measured before each experiment. Total suspended solids (TSS) was analyzed using standard methods 2540-D (APHA AWWA WEF, 1998). Total and soluble chemical oxygen demand (tCOD and sCOD) was measured colorimetrically using the HACH method 8000 based on dichromate oxidation standard method number 5220 D (APHA AWWA WEF, 1998). sCOD was measured after filtering the sample through 0.45 µm GF/C Whatman filter paper. The 7

Journal Pre-proof Zeta potential of bypass wastewater samples was measured using a Malvern Zetasizer instrument (Malvern Instruments, Worcestershire, UK). The actual iron electrode loss was measured by gravimetric analysis of the dry anode before and after each treatment run. Fe2+ ion concentration was measured using the 1,10-phenanthroline method (APHA AWWA WEF, 1998). In addition, the following parameters were also measured before and after iron electrocoagulation: pH (Accumet electrode and Accumet Excel; Fisher Scientific, Ottawa, ON, Canada)), redox potential (Accumet Platinum Pin Ag/AgCl Combination

of

electrode and Accumet Excel; Fisher Scientific, Ottawa, ON, Canada), and dissolved oxygen (Model 52,

ro

YSI, Yellow Springs, Ohio, US).

-p

3. Results and discussion 3.1 Bypass wastewater quality

re

The characteristics of the bypass wastewater sample, shown in Table 1, indicate it to be relatively low

lP

strength in terms of tCOD, sCOD, and TSS (Henze and Comeau, 2008). This is consistent with the dilution caused by stormwater mixing with the sanitary sewage in the combined sewers that serve

na

approximately 15% of the community’s drainage area. The net surface charge of the bypass wastewater

ur

particles is negative, with a median zeta potential of -24.2 mV, indicating a stable suspension. The characteristics of bypass wastewater can be compared to combined sewer overflows which are noted to

Jo

vary considerably from one location to another, and even from one event to another at the same outfall. For instance, event mean concentrations (EMC) for COD in combined sewers were found to range from 135 to 1,873 mg/L during 10 overflow events in three cities in Spain (the maximum COD EMC in each city reached more than 1200 mg/L during at least one the events studied) (Suárez and Puertas, 2005). Additionally, the EMC of road runoff in various locations in China showed the very high variability of measured COD values from as low as 1.8–11.0 mg/L in Shenyang to as high as 245–1,902 mg/L in Xi’an (Li et al., 2014).

8

Journal Pre-proof 3.2 Overall process performance 3.2.1 Final pH and dissolved Fe2+ The effect of applied current density and two temperatures (a) 23 oC and (b) 8 oC on final pH after various electrocoagulation reaction durations is shown in Figure 1. The initial pH for all experiments was pH = 7.0 ± 0.1. pH values were observed to increase during each test run. Over the range of treatment durations, the pH stabilized somewhere in the range of 7.9 to 8.6 for 23 oC tests and 7.3 to 8.0 for tests

of

conducted at 8 oC. It is evident from Figures 1 that the average pH increase during treatment at 8 oC was

ro

approximately 0.5 pH units less than that at 23 oC. In an iron electrocoagulation batch reactor, Fe2+ is

-p

generated at the anode as shown in Reaction 1, and hydroxide (OH-) ions are generated at the cathode according to Reaction 2 which causes pH values to increase.

re

Anode: Fe(s)→Fe2+ (aq) +2e-

lP

Cathode: 2H2 O(l) + 2e- → H2 (g) ↑ + 2OH- (aq)

(1) (2)

na

Confirmation that Fe2+ was not entirely reacted with OH- was obtained by measurement at the end of each test run during this study. Figure 2 shows the effect of the applied current density on the percentage of

ur

total iron (Fet) generated at the anode that remained as Fe2+ after various electrocoagulation reaction

Jo

durations at 23 oC and 8 oC. None of the test conditions resulted in the conversion of all the anode generated Fe2+ to Fe(OH)3(s). At the lowest current density tested in this study (8 mA/cm2), Fe2+ was found to be the predominant form of aqueous iron with the average Fe2+ to Fet ratios being 70% and 90% for all reaction durations at 23 oC and 8 oC, respectively. Some of the Fe2+ to Fet ratio measurements after 5 minutes of electrocoagulation time at 8 mA/cm2 current density and 8 oC temperature were observed to be higher than 100%. In a study that used synthetic wastewater and stainless steel electrodes, it was found that some of the experiments conducted at lower currents (<0.2 A) yielded current efficiencies higher than 100% (Lee and Gagnon, 2015). This generation of aqueous iron in excess of that predicted by Faraday’s Law was attributed to electrochemical side-reactions (Lee and Gagnon, 2015).

9

Journal Pre-proof For the tests conducted at 23 oC, increasing the current density beyond 8 mA/cm2 resulted in a decrease in the ratio of Fe2+ to Fet from 70% to an average of 32% for all electrolysis durations using 15 mA/cm2, followed by a slight increase to an average of 38% for all electrolysis durations at 22 mA/cm2 current density. During the experiments at 8 oC, the increase of current density beyond 8 mA/cm2 resulted in an overall decrease in the ratio of Fe2+ to Fet with average ratios of 74% and 56% being reached for all electrolysis durations using 15 mA/cm2 and 22 mA/cm2 current densities, respectively. Increasing

of

electrolysis duration for a given applied current density had no significant impact on Fe2+ to Fet ratio at the tested temperatures (p > 0.05).

ro

The decrease of Fe2+ to Fet ratio with increasing applied current density is consistent with the increased

-p

OH- production, according to Reaction 2, and subsequent increase in Fe(OH)3(s) production, according to

re

Reaction 3, that an increase in applied current would cause.

(3)

lP

− 4[Fe2+ (aq) + 2OH(aq) ] + O2 + 2H2 O(l) → 4Fe(OH)3 (𝑠)

One would expect the conversion of Fe2+ to Fe(OH)3(s) to continue until the equilibrium conditions

na

specified by the Pourbaix diagrams shown in Figure 3 were reached. In fact, the final pH had not changed

ur

when measured 30 minutes after the current had been discontinued (data not shown), suggesting that the

Jo

progress of Reaction 3 was also limited by the availability of dissolved oxygen. A sensitivity analysis of Fe2+ oxidation was performed with respect to matrix pH using the Fe2+ oxidation model of Millero et al. (1987) to estimate the degree to which Fe2+ would be oxidized under the dissolved oxygen and salinity conditions of this study. The dissolved oxygen concentration measured in the electrocoagulation cell at the beginning of each test was approximately 3 mg/L on average and was observed to change (decrease) by no more than 0.3 mg/L during any of the tests. Figure 4 shows the estimated proportion of Fet remaining as Fe2+ at the 3 mg/L dissolved oxygen concentration and an assumed salinity of 0.1 PSU (Hoffman and Meighan, 1984) after 40 minutes reaction time at 23 oC and 8 o

C.

10

Journal Pre-proof These calculations show that Fe2+ oxidation is not completed within a 40-minute reaction time under these conditions and that Fe(OH)3(s) would still be forming. Thus, adsorption equilibrium could not have been established within the electrocoagulation cell during the tests that ranged from 5 to 40 minutes in duration. As was observed experimentally, Figure 4 shows that an increase in temperature from 8 oC to 23 o

C reduces the amount of Fet remaining as Fe2+, but a considerable proportion of the Fet still remains as

Fe2+ after 40 minutes within the pH range measured in the electrocoagulation cell during this study.

of

Indeed, pH has a great effect on the rate of Fe2+ oxidation, with an increase of one pH unit resulting in a 100-fold increase in the Fe2+ oxidation rate. The estimated Fe2+ to Fet ratios shown in Figure 4 are higher

ro

than those observed in the present study and shown in Figure 2 for 40-minute reaction times. This may be

-p

due to some aqueous Fe2+ forming complexes with the organic matter present in the wastewater sample.

re

Fe2+ bound in these complexes would not have been measured by the assay used in this study (Bagga et

3.2.2 sCOD removal efficiencies

lP

al., 2008; Tanneru and Chellam, 2012; Theis and Singer, 1974).

na

sCOD removal efficiencies increased with the electrolysis duration at each of the three current densities and two temperatures tested, as shown in Figure 5. During the experiments conducted at 23 oC as depicted

ur

in Figure 5-a, the percentage of sCOD removal increased significantly when the current density was

Jo

increased from 8 mA/cm2 to 15 mA/cm2; however, increasing the current density beyond 15 mA/cm2 did not show any significant improvement in sCOD removal efficiency (p > 0.05). Data shown in Figure 5-b indicate that sCOD removal improved considerably with increasing current density during the treatment at 8 oC. While increasing current densities increase the amount of dissolved iron, these increases may lead to the creation of conditions favorable to the formation of other soluble species such as ferric hydroxo complexes with hydroxide ions and polymeric species that do not necessarily convert fast enough to form the insoluble Fe(OH)3(s) and may not contribute to sCOD removal (Kalyani et al., 2009; Kobya et al., 2003). This might explain the meagre improvement in sCOD removal efficiency as the current density 11

Journal Pre-proof was increased beyond 15 mA/cm2 at 23 oC. The results of the experiments conducted at 8 oC, contained in Figure 5-b, show on the other hand that the percentage sCOD removal continued to increase as the current density was increased from 8 mA/cm2 to 15 mA/cm2 and then to 22 mA/cm2. This shows that an increase in current density at 8 oC resulted in a very beneficial decrease in the proportion of Fet remaining as Fe2+ as shown in Figure 2, and so relatively more precipitate was available to adsorb sCOD. Very few reports exist in the literature concerning the effect of temperature on the removal of sCOD

of

using electrocoagulation technology (Vepsäläinen et al., 2009; Wang et al., 2010). Comparison of the sCOD removal results at each temperature shows that removal efficiencies increased considerably for a

ro

given current density when the temperature was increased from 8 to 23 oC. Or for a given current density,

-p

raising the temperature from 8 oC to 23 oC considerably reduced the treatment time required to achieve a

re

given sCOD removal. These increases in sCOD removal efficiencies due to increased temperature can be ascribed to: (1) increased amount of Fe(OH)3(s) precipitate formed at 23 oC to that formed at 8 oC as

lP

indicated by the lesser amount of dissolved Fe2+ measured in solution at 23 oC as compared to 8 oC

na

(Figure 2); and (2) faster de-passivation of the anode. The iron electrocoagulation technique employed in this study can achieve remarkable removal of sCOD

ur

as compared to other established and emerging technologies employed as a single step and standalone

Jo

treatment of bypass or municipal wastewater. In this study, iron electrocoagulation achieved sCOD removals of more than 70% at 8 oC and more than 80% at 23 oC after 40 minutes and at 22 mA/cm2. The conventional primary sedimentation process, if used alone, is usually effective in removing particles larger than 50 µm within a reasonable time (Levine et al., 1985). However, Lessard and Beck (1988) found that only 12% of sCOD was removed from sewage after 3 hours of conventional primary treatment. Henrichs et al. (2007) applied combined sewer overflow to wetlands and reported less than 30% sCOD removal after a 20-hour residence time. High rate ballasted clarification that combines coagulation, flocculation, and settling processes in compact footprint has achieved sCOD removal between 50% and 60% under optimal operating conditions following a 30 minute startup period (Jolis and Ahmad, 2004).

12

Journal Pre-proof Potassium ferrate(VI), identified as an emerging technology, showed less than 20% removal of sCOD after 80 minutes when used as the sole process for the treatment of municipal wastewater as reported elsewhere (Jiang et al., 2007; Li et al., 2016; Wang et al., 2018).

3.3 Modeling the iron electrocoagulation process 3.3.1 Modeling assumptions For the purpose of modeling iron electrocoagulation for the removal of soluble material, clear

of

assumptions should be stated especially with regard to defining end products and the degree to which the

ro

iron ionized from the electrode is converted to Fe(OH)3(s). The reaction mechanisms and the end products

-p

surveyed in the literature are summarized in Table S1 in the supplementary information. Depending on the operational conditions, the reported primary end products summarized in Table S1 varied from being

re

predominantly Fe(OH)3(s) to being a mixture of Fe(OH)2 and Fe(OH)3(s), or a mixture of Fe(OH)3(s) and

lP

other polymeric hydroxy complexes similar to ferric chloride products, or a mixture of soluble Fe2+ and insoluble Fe(OH)3/FeOOH or a mixture of Fe(OH)3(s) and FeOOH. It can be seen that there is a conflict

na

between these reports and the assumption that iron electrocoagulation end products are entirely

ur

Fe(OH)3(s). Changing operating parameters such as pH, dissolved oxygen, actual iron released, redox potential and treatment duration can lead to a significant impact on the ionic iron reactions (Lakshmanan

calculated values.

Jo

et al., 2009); thus potentially yielding lower concentrations of precipitates than the theoretically

Other researchers sought to provide insights into the predominant species resulting from iron electrocoagulation under the conditions tested in their studies by the use of the Pourbaix diagram (Moreno-Casillas et al., 2007). Figure 3 shows the regions of the current study experimental conditions superimposed upon an iron Pourbaix diagram. The superimposed highlighted regions show the area in which the reactions occur starting with the initial neutral pH measured in this study. At 23 oC, the process begins and under favorable conditions would proceed to form the desirable insoluble Fe(OH)3(s). At 8 oC, the initial equilibrium position lies at the boundary between Fe2+ and Fe(OH)3(s), and as the pH increases 13

Journal Pre-proof during the treatment, the equilibrium position shifts further into the precipitate region in which Fe(OH)3(s) is formed. It is postulated that the successful iron electrocoagulation process may not relate entirely to the redox potential and initial pH, but the rate of oxidation of Fe2+ should be equally considered. Iron Pourbaix diagrams describe equilibrium conditions according to thermodynamic considerations, from which reaction tendencies may be inferred, but no kinetic information can be deduced. They provide no information regarding the rate at which hydrated Fe2+ ions formed by the

of

external DC will be oxidized to Fe(OH)3(s) as per Reaction 3. In addition, the theoretical estimate shown in Figure 4 demonstrates that the oxidation of Fe2+ to Fe3+ would not reach equilibrium (the point

-p

tests, mainly because of oxygen and hydroxide limitations.

ro

expected according to the Pourbaix diagram) within the reaction times allowed in the present study batch

re

Since equilibrium conditions are not attained during iron electrocoagulation of environmental samples, an

lP

appropriate modeling approach to the electrocoagulation system should consider the system kinetics. The development of adsorption kinetics theory for iron electrocoagulation is slow despite its importance to

na

practical applications. Variable-order-kinetic (VOK) models have been developed for aluminum electrocoagulation systems (Essadki et al., 2010; Hu et al., 2007). These VOK models are based on either

ur

the Langmuir (Hu et al., 2007) or the Langmuir-Freundlich (Essadki et al., 2010) expressions and are

Jo

formulated to describe the adsorption rate. VOK models are able, with minimal complexity, to identify the effects of operating conditions such as duration and current, and to predict electrocoagulation performance in removing a target contaminant.

3.3.2 Kinetic adsorption modeling For the purpose of modeling adsorbate-coagulant interactions, four assumptions are made in this study. These are: (1) adsorption takes place on the precipitate species generated from the iron electrocoagulation process; (2) precipitate species are considered to be Fe(OH)3(s) whose concentration is calculated by deducting the Fe2+ measured in the effluent from the Fe2+ produced theoretically according to Faraday’s law and then considering this resultant Fe2+ to be converted to Fe(OH)3(s); (3) once this precipitate is 14

Journal Pre-proof formed, its surface area is large enough to allow immediate adsorption of sCOD; and (4) sCOD removal by flotation is negligible. The present study treats the iron electrocoagulation batch reactor sCOD results in a manner analogous to that found extremely suitable in aluminum-based electrocoagulation (Essadki et al., 2010; Hu et al., 2007), but with some modifications. The VOK form of a Langmuir adsorption process before attaining equilibrium may be represented by Equation 1 (Hu et al., 2007). dCt I k L Ct = −ϕ. . qLmax . dt zFV 1 + k L Ct

of ro

[Fe2+ ] ) [Fet ]

(2)

-p

ϕ = (1 −

(1)

where: ϕ accounts for the efficiency of the process in forming Fe(OH)3 precipitate and is calculated

re

according to Equation 2 in which [Fet] represents the total iron liberated from the electrode according to

lP

Faraday’s Law and [Fe2+] is the measured aqueous ferrous ion concentration. The term, ϕ, is treated as a constant and calculated as an average value evaluated and applied for each experimental run. F is

na

Faraday’s constant (96485 C(mol e-)-1). The charge transfer number (z) in Equation 1, has been previously

ur

investigated, and it has been conclusively shown that z = 2 for iron (Ben Sasson et al., 2009; Lakshmanan et al., 2009). qLmax in mg/g is a constant reflecting the maximum adsorption capacity per unit mass of

Jo

adsorbent; KL in min-1 is a constant representing the adsorption intensity; V is the batch reactor volume (L); and Ct in mg/L is the aqueous adsorbate concentration achieved after time t of iron electrocoagulation. Unfortunately, the integrated form of Equation 1 is an implicit function of aqueous adsorbate concentration, and so the mathematical expression for the treatment time required to achieve a specific aqueous adsorbate concentration using Langmuir VOK model is represented by Equation 3. t=

zFV 1 Co [(Co − Ct ) + ln ( )] ϕ. I. qLmax kL Ct

(3)

15

Journal Pre-proof As sorbable material exists in two phases: in a sorbed phase on Fe(OH)3(s) precipitates and the aqueous phase of the bulk liquid outside precipitates, the "conservation equation" expressed in Equation 4 can be applied to Equation 3 and Equation 5 can be produced. Co . V = Ct . V + qt . Wt

(4)

(5)

of

zFV qt . Wt 1 1 t= + ln ( [ )] q .W ϕ. I. qLmax V kL 1 − V.t C t o

where Co in mg / L is the initial concentration of the sCOD, V is the volume of solution, Wt in g is the

ro

mass of the Fe(OH)3(s) precipitate at time t; and qt in mg/g is the amount of sCOD adsorbed per gram of

-p

the Fe(OH)3(s) precipitate at time t.

re

The other model tested in the current study is the VOK form of a Langmuir-Freundlich adsorption process

(6)

na

dCt I k LF Ct 𝑛 = −ϕ. .q . dt zFV Lmax 1 + k LF Ct 𝑛

lP

represented by Equation 6 (Essadki et al., 2010).

The mathematical expression for the treatment time required to achieve specific treatment efficiency of

ur

sCOD using the Langmuir-Freundlich VOK model is represented by Equation 7. (1−n)

zFV Co − Ct (1−n) t= [C − Ct + ] ϕ I qLFmax o k LF (1 − n)

Jo

(7)

Utilizing Equation 4, Equation 7 can be also rewritten as shown in Equation 8 to describe the same model.

zFV qt . Wt t= + [ ϕ. I. qLFmax V

(1−n) q .W Co (1−n) + ( t V t − Co )

k LF (1 − n)

(8) ]

In the current study, calibration of the models was conducted using all the data collected at a given temperature based on the assumption that current density would not affect the precipitate characteristics in terms of its adsorption kinetics. This assumption is supported by the results of the mathematical models of the studies that investigated the use of Langmuir and Langmuir-Freundlich VOK models which found

16

Journal Pre-proof that the adsorption capacity and intensity did not vary when changing the initial current or current density (Essadki et al., 2010; Hu et al., 2007). The estimates of the isotherm model parameter values and fitting statistics of the two tested models are shown in Table 2. Figures 6 and 7 show the simulation results of sCOD by Langmuir and Langmuir-Freundlich VOK models, respectively, at different current densities. The second-order Akaike Information Criterion (AICc) was used in this study to evaluate the models’ relative abilities to describe the adsorption kinetics (Sugiura, 1978). As can be seen from Table 2, the

of

AICc values associated with the Langmuir VOK model are lower compared to those of the LangmuirFreundlich VOK model for all current densities and temperatures. This indicates that Langmuir VOK

ro

provides a better description of the process under these conditions.

-p

As shown in Table 2, increasing temperature from 8 to 23 oC drastically increased the adsorption capacity

re

of the insoluble Fe(OH)3(s) for removing sCOD from bypass wastewater. This observation might give

lP

insights into the adsorption category i.e., physisorption or chemisorption. Although the adsorption process is exothermic in nature as the residual forces on the surface of the adsorbent decrease with the increase of

na

the adsorbate occupying the adsorption sites according to Le Chatelier’s principle, physisorption or chemisorption respond differently. On one hand, physisorption decreases with increasing temperature due

ur

to the weak forces between the adsorbate and adsorbent. On the other hand, chemisorption initially

Jo

increases with the increase in temperature as activation energy is required for chemical reactions to take place then decreases reflecting the desorption effect. The results in this study reveal that the first phase of chemisorption prevails as the adsorption capacity of the insoluble Fe(OH)3(s) for removing sCOD from bypass wastewater increased with increasing temperature from 8 to 23 oC. Similar conclusions regarding the chemisorption nature of the iron electrocoagulation were drawn in a previous study using secondorder kinetics representing the removal of ciprofloxacin from hospital wastewater (Yoosefian et al., 2017).

17

Journal Pre-proof 3.4 Challenges for iron electrocoagulation bypass wastewater treatment Retrofitting existing primary sedimentation tanks using iron electrocoagulation can increase their current capacity without increasing their footprint. This will greatly help in intensifying existing infrastructure. While the batch-mode bench-scale experiments employed in this study demonstrated that iron electrocoagulation can be an efficient treatment strategy for bypass wastewater, there are several issues that still need to be addressed before its full-scale implementation. This will include flow-through

of

experiments to test operational strategies to deal with variable flow and organic loads, examining

ro

different iron electrode configurations and influent flow modes through the reactor, identifying reactor scale-up parameters, and investigating iron electrode cleaning requirements. Testing of alternative reactor

-p

designs should also consider a dosing system that involves electrolytic production of iron species in a

re

clean water side stream in order to reduce electrode fouling. Moreover, potential reactor designs should

lP

consider the possibility of reusing scrap iron in the treatment of municipal wastewater and compare its efficiency with higher purity iron electrodes. This latter recommendation may ultimately minimize the

ur

4. Conclusions

na

overall cost of maintaining iron electrocoagulation systems.

Iron electrocoagulation was tested as an enhancement option for the primary treatment of bypass

Jo

wastewater. Assessment of the removal of sCOD at near-neutral pH conditions by changing electrolysis duration, current density, and temperature was conducted. sCOD removal efficiencies increased with the increase of electrolysis duration, current density, and temperature. Particularly, the temperature effect was proven in this study, for the first time, for the treatment of bypass wastewater using iron electrocoagulation. It was found that at 23 oC, it took 15 minutes to reach an average 52% sCOD removal efficiency while it took around 40 minutes to achieve comparable removal efficiency at 8 oC; both tested at 15 mA/cm2 current density. Perspectives on using adsorption isotherms to model the iron electrocoagulation process that considered both experimental protocols and modeling assumptions were included in this study. Under the operational conditions tested in this study, equilibrium could not have 18

Journal Pre-proof been established. This conclusion underscores the need for researchers to exercise caution when applying adsorption isotherms to iron electrocoagulation systems. Alternatively, kinetic based models such as the Langmuir and Langmuir-Freundlich VOK models would be more appropriate in modeling adsorption results from iron electrocoagulation. These models were examined in this study with suitable assumptions and consideration of the de-facto estimation of Fe(OH)3(s) (adsorbent) that accounts for the contribution of Fe2+ to treatment end products. The Langmuir VOK model was found to be the better model to

of

describe sCOD removal in the electrocoagulation cell under all test conditions with chemisorption being

ro

considered the dominant sCOD removal mechanism. 5. Acknowledgments

-p

This research was supported by the Natural Sciences and Engineering Research Council of Canada

re

(NSERC). The first author, Haitham Elnakar, was supported by NSERC Vanier Canada doctoral

lP

scholarship, Alberta Innovates-Technology Futures scholarship and top-up award, and University of

6. References

na

Alberta President’s Doctoral Prize of Distinction.

ur

Ahmadzadeh, S., Dolatabadi, M., 2018a. Removal of acetaminophen from hospital wastewater using

Jo

electro-Fenton process. Environ. Earth Sci. 77, 1–11. doi:10.1007/s12665-017-7203-7 Ahmadzadeh, S., Dolatabadi, M., 2018b. Modeling and kinetics study of electrochemical peroxidation process for mineralization of bisphenol A; a new paradigm for groundwater treatment. J. Mol. Liq. 254, 76–82. doi:10.1016/j.molliq.2018.01.080 Ahmadzadeh, S., Dolatabadi, M., 2018c. Electrochemical treatment of pharmaceutical wastewater through electrosynthesis of iron hydroxides for practical removal of metronidazole. Chemosphere 212, 533–539. doi:10.1016/j.chemosphere.2018.08.107 APHA AWWA WEF, 1998. Standard methods for the examination of water and wastewater, 20th ed. APHA-AWWA-WEF, Washington, D.C. 19

Journal Pre-proof Bagga, A., Chellam, S., Clifford, D.A., 2008. Evaluation of iron chemical coagulation and electrocoagulation pretreatment for surface water microfiltration. J. Memb. Sci. 309, 82–93. doi:10.1016/j.memsci.2007.10.009 Ben Sasson, M., Calmano, W., Adin, A., 2009. Iron-oxidation processes in an electroflocculation (electrocoagulation) cell. J. Hazard. Mater. 171, 704–709. doi:10.1016/j.jhazmat.2009.06.057 Brar, S.K., Blais, J.-F., Benmoussa, H., Asselin, M., Drogui, P., 2009. Electrochemical removal of

of

organics and oil from sawmill and ship effluentsA paper submitted to the Journal of Environmental

ro

Engineering and Science. Can. J. Civ. Eng. 36, 529–539. doi:10.1139/l09-003

-p

Chhetri, R.K., Bonnerup, A., Andersen, H.R., 2016. Combined Sewer Overflow pretreatment with chemical coagulation and a particle settler for improved peracetic acid disinfection. J. Ind. Eng.

re

Chem. 37, 372–379. doi:10.1016/j.jiec.2016.03.049

na

Implementation Plan. Edmonton.

lP

City of Edmonton, 2000. The City of Edmonton Combined Sewer Overflow (CSO) Control Strategy -

Elnakar, H., Buchanan, I., 2019a. The role of mixing in potassium ferrate(VI) consumption kinetics and

ur

disinfection of bypass wastewater. J. Environ. Manage. 231, 515–523.

Jo

doi:10.1016/j.jenvman.2018.10.076 Elnakar, H., Buchanan, I., 2019b. Treatment of bypass wastewater using potassium ferrate(VI): assessing the role of mixing. Environ. Technol. (United Kingdom). doi:10.1080/09593330.2019.1609589 Essadki, A.H., Gourich, B., Azzi, M., Vial, C., Delmas, H., 2010. Kinetic study of defluoridation of drinking water by electrocoagulation/electroflotation in a stirred tank reactor and in an external-loop airlift reactor. Chem. Eng. J. 164, 106–114. doi:10.1016/j.cej.2010.08.037 GilPavas, E., Dobrosz-Gómez, I., Gómez-García, M.Á., 2019. Optimization and toxicity assessment of a combined electrocoagulation, H2O2/Fe2+/UV and activated carbon adsorption for textile wastewater treatment. Sci. Total Environ. 651, 551–560. doi:10.1016/j.scitotenv.2018.09.125 20

Journal Pre-proof Harleman, D., Murcott, S., 1999. The role of physical-chemical wastewater treatment in the mega-cities of the developing world. Water Sci. Technol. 40, 75–80. doi:10.1016/S0273-1223(99)00487-4 Hart, O.E., Halden, R.U., 2019. On the need to integrate uncertainty into U.S. water resource planning. Sci. Total Environ. 691, 1262–1270. doi:10.1016/j.scitotenv.2019.07.164 Henrichs, M., Langergraber, G., Uhl, M., 2007. Modelling of organic matter degradation in constructed wetlands for treatment of combined sewer overflow. Sci. Total Environ. 380, 196–209.

of

doi:10.1016/j.scitotenv.2006.11.044

ro

Henze, M., Comeau, Y., 2008. Wastewater characterization, in: Henze, M., Loosdrecht, M.C.M. van,

-p

Ekama, G.A., Brdjanovic, D. (Eds.), Biological Wastewater Treatment: Principles Modelling and

re

Design. IWA Publishing, London, UK, p. 21.

Hoffman, R.W., Meighan, R.B., 1984. The Impact of Combined Sewer Overflows from San Francisco on

lP

the Western Shore ofCentral San Francisco Bay. Water Pollut. Control Fed. 56, 1277–1285.

na

doi:10.1016/S0262-1762(99)80122-9

Hu, C.Y., Lo, S.L., Kuan, W.H., 2007. Simulation the kinetics of fluoride removal by electrocoagulation

ur

(EC) process using aluminum electrodes. J. Hazard. Mater. 145, 180–185.

Jo

doi:10.1016/j.jhazmat.2006.11.010 Jiang, J.-Q., Wang, S., Panagoulopoulos, A., 2007. The role of potassium ferrate(VI) in the inactivation of Escherichia coli and in the reduction of COD for water remediation. Desalination 210, 266–273. doi:10.1016/j.desal.2006.05.051 Jolis, D., Ahmad, M., 2004. Evaluation of high-rate clarification for wet-weather-only treatment facilities. Water Environ. Res. 76, 474–480. doi:10.2175/106143004X151563 Kalyani, K.S.P., Balasubramanian, N., Srinivasakannan, C., 2009. Decolorization and COD reduction of paper industrial effluent using electro-coagulation. Chem. Eng. J. 151, 97–104. doi:10.1016/j.cej.2009.01.050 21

Journal Pre-proof Kobya, M., Can, O.T., Bayramoglu, M., 2003. Treatment of textile wastewaters by electrocoagulation using iron and aluminum electrodes. J. Hazard. Mater. 100, 163–178. doi:10.1016/S03043894(03)00102-X Lakshmanan, D., Clifford, D.A., Samanta, G., 2009. Ferrous and ferric ion generation during iron electrocoagulation. Environ. Sci. Technol. 43, 3853–3859. doi:10.1021/es8036669 Lee, S.Y., Gagnon, G.A., 2015. The rate and efficiency of iron generation in an electrocoagulation

of

system. Environ. Technol. (United Kingdom) 36, 2419–2427. doi:10.1080/09593330.2015.1032367

ro

Lessard, P., Beck, M.B., 1988. Dynamic Modeling of Primary Sedimentation. J. Environ. Eng. 114, 753–

-p

769. doi:10.1061/(ASCE)0733-9372(1988)114:4(753)

re

Levine, A.D., Tchobanoglous, G., Asano, T., Journal, S., Pollution, W., Federation, C., Jul, N., 1985. Characterization of the Size Distribution of Contaminants in Wastewater: Treatment and Reuse

lP

Implications. Water Pollut. Control Fed. 57, 805–816. doi:10.2307/25042701

na

Li, C., Liu, M., Hu, Y., Gong, J., Sun, F., Xu, Y., 2014. Characterization and first flush analysis in road and roof runoff in Shenyang, China. Water Sci. Technol. 70, 397–406. doi:10.2166/wst.2014.203

ur

Li, N., Deng, Y., Sarkar, D., 2016. Ferrate(VI) Reaction with Effluent Organic Matter (EfOM) in

1238.ch015

Jo

Secondary Effluent for Water Reuse. ACS Symp. Ser. 1238, 411–420. doi:10.1021/bk-2016-

Masi, F., Rizzo, A., Bresciani, R., Conte, G., 2017. Constructed wetlands for combined sewer overflow treatment: Ecosystem services at Gorla Maggiore, Italy. Ecol. Eng. 98, 427–438. doi:10.1016/j.ecoleng.2016.03.043 Metcalf & Eddy Inc., 2003. Wastewater engineering: Treatment and reuse, 4th ed. McGraw-Hill, New York. Mollah, M.Y., Schennach, R., Parga, J.R., Cocke, D.L., 2001. Electrocoagulation (EC)-Science and

22

Journal Pre-proof Applications. J. Hazard. Mater. 84, 29–41. doi:10.1016/S0304-3894(01)00176-5 Moreno-Casillas, H.A., Cocke, D.L., Gomes, J.A.G., Morkovsky, P., Parga, J.R., Peterson, E., 2007. Electrocoagulation mechanism for COD removal. Sep. Purif. Technol. 56, 204–211. doi:10.1016/j.seppur.2007.01.031 Müller, S., Behrends, T., van Genuchten, C.M., 2019. Sustaining efficient production of aqueous iron during repeated operation of Fe(0)-electrocoagulation. Water Res. 155, 455–464.

of

doi:10.1016/j.watres.2018.11.060

ro

Nguyen, D.D., Ngo, H.H., Guo, W., Nguyen, T.T., Chang, S.W., Jang, A., Yoon, Y.S., 2016. Can

-p

electrocoagulation process be an appropriate technology for phosphorus removal from municipal

re

wastewater? Sci. Total Environ. 563–564, 549–556. doi:10.1016/j.scitotenv.2016.04.045 Risch, E., Gasperi, J., Gromaire, M.C., Chebbo, G., Azimi, S., Rocher, V., Roux, P., Rosenbaum, R.K.,

lP

Sinfort, C., 2018. Impacts from urban water systems on receiving waters – How to account for

na

severe wet-weather events in LCA? Water Res. 128, 412–423. doi:10.1016/j.watres.2017.10.039 Suárez, J., Puertas, J., 2005. Determination of COD, BOD, and suspended solids loads during combined

ur

sewer overflow (CSO) events in some combined catchments in Spain. Ecol. Eng. 24, 201–219.

Jo

doi:10.1016/j.ecoleng.2004.11.005

Sugiura, N., 1978. Further analysts of the data by akaike’ s information criterion and the finite corrections. Commun. Stat. - Theory Methods 7, 13–26. doi:10.1080/03610927808827599 Tanneru, C.T., Chellam, S., 2012. Mechanisms of virus control during iron electrocoagulation Microfiltration of surface water. Water Res. 46, 2111–2120. doi:10.1016/j.watres.2012.01.032 Theis, T.L., Singer, P.C., 1974. Complexation of Iron(II) by Organic Matter and Its Effect on Iron(II) Oxygenation. Environ. Sci. Technol. 8, 569–573. doi:10.1021/es60091a008 Tram VO, P., Ngo, H.H., Guo, W., Zhou, J.L., Nguyen, P.D., Listowski, A., Wang, X.C., 2014. A mini-

23

Journal Pre-proof review on the impacts of climate change on wastewater reclamation and reuse. Sci. Total Environ. 494–495, 9–17. doi:10.1016/j.scitotenv.2014.06.090 U.S. EPA, 2013. Emerging technologies for wastewater treatment and in-plant wet weather management. Virginia. U.S. EPA, 2008. A Screening Assessment of the Potential Impacts of Climate Change on Combined Sewer Overflow ( CSO ) Mitigation in the Great Lakes and New England Regions. Washington,

of

DC.

ro

U.S. EPA, 1999a. Combined sewer overflow technology fact sheet: Alternative disinfection methods.

-p

Washington, D.C. doi:EPA 832-F-99-041

re

U.S. EPA, 1999b. Combined Sewer Overflow Technology Fact Sheet: Sewer Separation, United States

lP

Environmental Protection Agency (U.S. EPA). Washington, D.C. doi:EPA 832-F-99-041 UNESCO, 2017. The United Nations World Water Development Report 2017: Wastewater - The

na

Untapped Resource. UNESCO on behalf of UN-Water, France. Vepsäläinen, M., Ghiasvand, M., Selin, J., Pienimaa, J., Repo, E., Pulliainen, M., Sillanpää, M., 2009.

ur

Investigations of the effects of temperature and initial sample pH on natural organic matter (NOM)

Jo

removal with electrocoagulation using response surface method (RSM). Sep. Purif. Technol. 69, 255–261. doi:10.1016/j.seppur.2009.08.001 Wang, C.T., Chou, W.L., Huang, K.Y., 2010. Treatment of polyvinyl alcohol from aqueous solution via electrocoagulation. Sep. Sci. Technol. 45, 212–220. doi:10.1080/01496390903423808 Wang, H., Li, H., Ding, N., Li, M., Wang, N., 2018. Using potassium ferrate as advanced treatment for municipal wastewater. Desalin. Water Treat. 106, 90–97. doi:10.5004/dwt.2018.22064 Yılmaz Nayır, T., Kara, S., 2018. Container washing wastewater treatment by combined electrocoagulation–electrooxidation. Sep. Sci. Technol. 53, 1592–1603.

24

Journal Pre-proof doi:10.1080/01496395.2017.1411365 Yoosefian, M., Ahmadzadeh, S., Aghasi, M., Dolatabadi, M., 2017. Optimization of electrocoagulation process for efficient removal of ciprofloxacin antibiotic using iron electrode; kinetic and isotherm

Jo

ur

na

lP

re

-p

ro

of

studies of adsorption. J. Mol. Liq. 225, 544–553. doi:10.1016/j.molliq.2016.11.093

25

Journal Pre-proof List of Figures Figure 1 Final pH following treatment at different current densities and electrocoagulation reaction times at (a) 23 oC and (b) 8 oC. Initial pH for all experiments was 7.0 ± 0.1. Figure 2 The percentage of total iron (Fet) remaining as ferrous (Fe2+) at different current densities and after various electrocoagulation reaction times, at (a) 23 oC and (b) 8 oC.

of

Figure 3 Iron Pourbaix diagram showing the area (highlighted in green) in which the iron

ro

electrocoagulation process occurs at 23 oC and 8 oC (after Pourbaix (1966)).

-p

Figure 4 Estimated proportion of total iron (Fet) remaining as ferrous (Fe2+) versus pH after 40 minutes electrocoagulation reaction time at 23 oC and 8 oC using the ferrous oxidation model

re

described in Millero et al. (1987) (Dissolved Oxygen = 3 mg/L and Salinity = 0.1 PSU).

lP

Figure 5 Removal efficiencies of soluble chemical oxygen demand (sCOD) at different current

ur

temperatures.

na

densities and after various electrocoagulation reaction times, at (a) 23 oC and (b) 8 oC

Figure 6 Simulation results of soluble chemical oxygen demand (sCOD) by Langmuir VOK

Jo

model at different current densities and at (a) 23 oC and (b) 8 oC temperatures. Figure 7 Simulation results of soluble chemical oxygen demand (sCOD) by LangmuirFreundlich VOK model at different current densities and at (a) 23 oC and (b) 8 oC temperatures.

26

Journal Pre-proof List of Tables Table 1 Characterization of bypass wastewater used for experiments

Jo

ur

na

lP

re

-p

ro

of

Table 2 Estimates of the batch adsorption kinetic model parameter values and fitting statistics

27

Journal Pre-proof

Table 1 Characterization of bypass wastewater used for experiments Parameter

Units

pH

Mean ± Standard Deviation

mV

-24.2 ± 2.8

Total Chemical Oxygen Demand

mg/L

350.0 ± 4.4

Soluble Chemical Oxygen Demand

mg/L

197.6 ± 2.5

Total Suspended Solids

mg/L

212.4 ± 29.4

Jo

ur

na

lP

re

-p

ro

of

Zeta Potential

7.0 ± 0.1

28

Journal Pre-proof

Table 2 Estimates of the batch adsorption kinetic model parameter values and fitting statistics 23 °C Current Density (mA/cm2) Parameters and Model Statistics 8 15 22 Langmuir Variable Order Kinetic Model qLmax 54.47 kL 1.03 x 10-05 Standard Error 5.38 4.22 4.14 2

0.89

8.29

1.18 6.90 x 10-04 1.62

3.02

0.73

0.98

0.96

14.82

18.49

8.14

0.12 1.02 x 10-07 3.70 4.41

2.91

0.89

0.83

0.95

0.98

44.30

46.38

40.26

36.12

0.93

0.93

of

R

8 °C Current Density (mA/cm2) 8 15 22

28.60

R

0.94

41.43

42.94

Jo

ur

na

lP

AICc

0.91

re

2

-p

ro

AICc 24.28 21.85 21.66 Langmuir-Freundlich Variable Order Kinetic Model qLFmax 16.21 kLF 1.24 x 10-06 n 1.88 Standard Error 4.96 5.77 6.61

29

Journal Pre-proof CONFLICT OF INTEREST STATEMENT

Jo

ur

na

lP

re

-p

ro

of

There are no conflicts of interest to declare.

30

Journal Pre-proof

Highlights

ur

na

lP

re

-p

ro

of

Primary wastewater treatment can be enhanced using iron electrocoagulation Temperature showed a significant effect on soluble chemical oxygen demand removal Final ferrous ion could be considerable, and process equilibrium can’t be assumed De-facto estimation of ferric hydroxide (adsorbent) should account for ferrous ion Suggested mechanism of soluble chemical oxygen demand removal was chemisorption

Jo

    

31

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7