Electrocoagulation-electroflotation for primary treatment of animal rendering wastewater to enable recovery of fats

Ao Xie, David A. Ladner, Sudeep C. Popat

Published in the Chemical Engineering Journal

Citation: Xie, A.; Ladner, D.A.; Popat, S.C. “Electrocoagulation-electroflotation for primary treatment of animal rendering wastewater to enable recovery of fats.” Chemical Engineering Journal 2022, 431, 133910.

https://doi.org/10.1016/j.cej.2021.133910

Abstract

Rendering wastewater (RW) contains high levels of fats from the rendering process that recovers fats and proteins from waste animal fatty tissue. Primary treatment to remove floatable fats is necessary to avoid issues for downstream RW treatment. Recovery of waste fats that can be sent back to the rendering process can be an additional gain from RW primary treatment. Dissolved air flotation (DAF) with the addition of polymer coagulants is commonly used for primary treatment of RW, but this has two disadvantages: polymer coagulants are undesirable contaminants in reused fats and the fat quality is degraded through oxidation by pressurized air during DAF treatment. In this work, RW was treated by electrocoagulation-electroflotation (EC-EF) in a novel horizontally-placed electrode pair design as an alternative to DAF. EC-EF performed better in COD removal with increasing current densities. In situ flotation within the electrochemical cell from hydrogen gas bubbles produced was able to concentrate and preserve long chain fatty acids (LCFA). Treatment at higher current densities within the electrochemical cell generated surplus coagulants and allowed further coagulation in a separate following step. Longer mixing duration in the post-EC-EF treatment step enhanced overall COD removal. The overall COD removal for EC-EF with applied current densities of 2 mA/cm² reached $69.5 \pm 4.8\%$ after 15 min of mixing at 100 rpm followed by 15 min of sedimentation. EC-EF thus provides competitive treatment performance and recovery of well-preserved rendering products.

1. Introduction

The rendering process extracts fats and proteins from received animal waste material to yield final products for both edible and inedible applications by “cooking”, which removes moisture and separates fats from protein-based solids [1]. Wastewater is intensively generated in the rendering industry from condensing cooking vapors and cleaning, with COD concentrations in the range of 4,000 to 10,000 mg/L [2], [3]. The cooker condensate, which contains a high concentration of volatile fats and oils, is a major contributor to rendering wastewater (RW) organic loadings [2]. Dissolved air flotation (DAF) is frequently used for primary treatment of RW [2]. In a DAF system, oversaturated air is released in the form of fine bubbles on which fats, protein and suspended solids attach and eventually float to the surface. The sludge collected from a DAF system at a rendering plant is typically sent back to the front of the rendering process for additional recovery; however, the exposure of floating fats to pressurized air increases their oxidation and deteriorates their quality [1], [4]. The high dosage of coagulants (metal salts and/or organic polymers) often supplemented for sufficient DAF treatment not only increases cost and labor, but also increases the possibility of secondary pollution in recycled products [5], [6], [7].

Electrocoagulation (EC) is an electrochemical coagulation-flocculation method that can be an alternative to DAF for rendering wastewater primary treatment [8]. An EC system consists of pairs of conductive electrodes in an electrolytic cell. When connected to an external power source, the oxidation of a metal electrode and water electrolysis takes place at the anode while the reduction of water to hydrogen (H2) gas occurs at the cathode. In solution, metal ions from the anode and hydroxide ions from the cathode form various hydroxomonomeric and polymeric hydroxy complexes, with eventual formation of amorphous M(OH)n(s) that provide a large surface area for adsorption. Contaminants can be separated via sedimentation after precipitation and floc formation with metal complexes [9]. The presence of an electric field in EC systems allows continuously neutralizing residual charges and thus removing the smallest colloidal particles to produce an effluent of higher quality [8]. The formation of metal hydroxide complex in situ avoids the potential of a high concentration of dissolved solids resulting from excessive chemical addition as in traditional coagulation-flocculation; this also eliminates the need for salts and polymers associated with coagulant chemicals [2], [10]. H2 gas generated at the cathode can provide inherent flotation treatment and also helps in forming stable flocs, thus decreasing the amount of sludge to be handled [11].

There are a few applications of EC for industrial wastewater treatment reported in literature. A majority of studies have focused on the choice of electrodes and operational parameters such as current densities and pH, using vertically-aligned electrodes [9], [12], [13]. Yet, the inherent flotation treatment has not been emphasized and taken advantage of beyond its contribution to overall EC treatment performance. Inherent flotation with H2 will be beneficial particularly for the rendering industry to increase profits by recycling well-preserved fats collected from primary treatment of RW. In this work, we thus determine the treatment efficiency of electrocoagulation-electroflotation (EC-EF) in a novel electrochemical cell design with horizontally assembled electrodes for a well-distributed flotation treatment (Fig. 1 (a)). In addition, we characterize the presence of long-chain fatty acids (LCFA) in all phases after EC-EF treatment: floatable solids collected from EC-EF cell, post-EC-EF effluent and sludge sediments combined from both EC-EF and post-EC-EF treatment. The goal of this work is to provide proof of concept for the application of EC-EF for RW treatment with a novel configuration and bring attention to the advantages of recovery of fats from separated solids.

Fig. 1. (a) Schematic of EC-EF electrochemical cell and post-EC-EF tank configurations and (b) diagram of EC-EF electrochemical cell construction.

2. Materials and methods

2.1. Electrocoagulation-electroflotation cell

The EC-EF cell used for this study was constructed with a series of plexiglass plates, with a working volume of 380 mL and height of 4 cm, as shown in Fig. 1 (b). Aluminum mesh (0.053′’ opening size, McMaster-Carr) and stainless-steel mesh (0.015′’ opening size, McMaster-Carr) were used as anode and cathode, respectively. Both anode and cathode had a projected area of 100 cm². Anode and cathode were placed horizontally, with the anode being 2 cm above the cathode. Both electrodes were set away from endplates by 1 cm. A vent was opened in the middle of the top plate to allow the release of H₂ gas.

2.2. Electrocoagulation-electroflotation experiments

RW was sampled before and after primary DAF treatment from an independent rendering plant. The effluent from the DAF treatment process was characterized to compare against the effluent obtained from EC-EF experiments. RW from before DAF treatment was used in this work for EC-EF treatment evaluation (Table S1). The pre-DAF RW contained $6{,}941 \pm 186 \, \text{mg}$ COD/L and 0.64 mg Al/L. Once the EC-EF cell was filled with RW, current density was applied at either 1 mA/cm², 2 mA/cm² or 3 mA/cm² for 15 min using a multi-channel potentiostat (BioLogic Science Instruments VMP3). All current densities are based on the projected area. No pH adjustment was made. All tests were performed in triplicate under room temperature. Samples were taken every 5 min from the aqueous level between the anode and cathode.

2.3. Post-electrocoagulation-electroflotation experiments

After 15 min of electrochemical treatment in the EC-EF cell, samples were collected and further processed by mixing with a magnetic stirrer at 100 rpm in a post-EC-EF sedimentation vessel (Fig. 1(a)). Four durations were tested for mixing: 1) no mixing, 2) 5 min, 3) 10 min, and 4) 15 min. When mixing was stopped, sedimentation was allowed subsequently for up to 24 h. Samples for analysis were taken immediately after mixing was stopped and at various intervals during sedimentation.

2.4. Control experiments

Open-circuit control experiments were performed to determine the treatment performance without electroflotation. For these experiments, RW was amended with aluminum potassium sulfate (KAl(SO₄)₂·12H₂O), prior to loading into the EC-EF cell without application of any current. Samples were collected for 15 min and then the effluent was processed in the post-EC-EF step as described above. Aluminum-amended RW with concentrations ranging from 0.64 mg Al/L (inherent RW concentration without amended aluminum) to 3.74 mg Al/L was freshly prepared right before each experiment and buffered with sodium bicarbonate (5 mM, pH 8.5) to yield neutral pH.

2.5. Analysis

Aqueous samples collected during EC-EF experiments were analyzed for pH, COD, and aluminum. Post-EC-EF aqueous samples were analyzed for COD, LCFA and volatile fatty acids (VFA). A layer of floated solids was observed during EC-EF experiments. This layer was trapped primarily on top of the anode. Samples of this layer were collected for analysis of LCFA. Sludge from post-EC-EF experiments was also collected for determination of LCFA.

COD, total VFA and aluminum were measured using Hach kits TNT 823, TNT 872 and TNT 848, respectively. pH was determined using a Thermo Scientific™ Orion™ Star A211 benchtop pH meter. LCFA analysis followed the protocol described by Ziels et al. [14] and Burja et al. [15] , by trans-esterification of LCFA with methanol to fatty acid methyl esters (FAME). Pentadecanoic acid was added as a recovery standard to determine the extraction efficiency, which was calculated as recovered pentadecanoic acid concentration divided by the initial concentration added. Pentadecane was added as an internal standard for GC signal calibration. FAME was extracted by hexane and determined on a gas chromatograph (Shimadzu GC-2010) equipped with a flame ionization detector (FID) and a Restek Rt-2560 column (100 m × 0.25 mm × 0.20 µm). Injector and FID temperatures were set at 240 °C. Column temperature was programmed as 100 °C for 5 min, followed by a ramp up to 240 °C at a rate of 3 °C/minute. Helium was used as the carrier gas. Concentrations were determined as mg/L of individual acids.

3. Results and discussion

3.1. Effect of current density in electrocoagulation-electroflotation experiments

Shown in Fig. 2 is the COD concentration as a function of time of treatment in the EC-EF cell at three different current densities. For these experiments, the pH and aluminum concentrations as a function of treatment time are shown in the Supporting Information (Fig. S1 and Fig. S2 respectively). From Fig. 2, it is apparent that some COD removal was observed at all current densities in the EC-EF cell, with increasing treatment time correlating with higher COD removal. Even so, COD removal across all current densities was low to moderate. The highest COD removal of $30.7 \pm 5.2\%$ was achieved at 3 mA/cm² by the end of the 15-minute test period. Final COD removal at the end of 15 min with applied current density of 1 mA/cm² and 2 mA/cm² was $16.8 \pm 5.9\%$ and $24.0 \pm 6.2\%$ , respectively.

Fig. 2. Effect of current density on COD removal in EC-EF experiments. Error bars show standard deviations as determined from three replicate experiments for each current density.

pH during the experiments with applied current densities of 2 mA/cm² and 3 mA/cm² increased with time, from $6.92 \pm 0.05$ to $8.17 \pm 0.34$ for 2 mA/cm² and to $9.21 \pm 0.16$ for 3 mA/cm². On the other hand, pH in the experiments with applied current densities of 1 mA/cm² remained steady throughout the 15-minute test period, with initial pH being $7.02 \pm 0.06$ and final pH $7.03 \pm 0.08$ , as summarized in Fig. S1. Aluminum concentration increased with both increasing treatment time and current density, as shown in Fig. S2. The highest concentrations of up to $3.24 \pm 0.65$ mg Al/L were measured in experiments with an applied current density of 3 mA/cm². Concentrations in experiments with applied current densities of 1 mA/cm² and 2 mA/cm² were $0.65 \pm 0.06$ mg Al/L and $1.56 \pm 0.2$ mg Al/L at the end of the 15-minute test period, respectively. In terms of energy usage, the three current densities resulted in electricity consumption of $0.059 \pm 0.001$ kWh/m³, $0.23 \pm 0.01$ kWh/m³ and $0.5 \pm 0.13$ kWh/m³, respectively. The energy usage for DAF is usually in the range of 0.05 – 0.075 kW/m³, suggesting that at the higher current densities, the energy usage here in EC-EF was higher. However, further optimization of the electrochemical cells could allow decreasing the overall energy usage, particularly by reducing distance between the anode and the cathode.

3.2. Effect of current density during EC-EF on post-EC-EF experiments

In post-EC-EF experiments, mixing conditions were tested to determine the effect of available aluminum species in RW that was not exhausted during the EC-EF experiments on coagulation-flocculation in the sedimentation vessel. COD was measured starting when mixing was stopped and over time during sedimentation. Shown in Fig. 3 is the decrease in COD concentration in post EC-EF treatment for the three different current densities applied during EC-EF. Shown in Fig. 3 (a) are results from sedimentation without mixing, as a function of time of sedimentation. Results after a mixing for 5 min at 100 rpm magnetic stirring, are shown as a function of time of sedimentation in Fig. 3 (b). From Fig. 3 (a), it is apparent that further COD removal in the post-EC-EF cell by sedimentation without mixing was minor, ranging from 3.1% to 8.2% within 15 min of sedimentation for all current densities applied during EC-EF. Longer sedimentation time correlated with improved COD removal, with 24 h of settling resulting in an overall COD removal of 50.1–54.7%.

Fig. 3. Effect of current density on COD removal in post-EC-EF experiments over time (a) with no mixing, (b) after 5 min of 100 rpm stirring. Error bars show standard deviations in all data points as determined from three replicate experiments for each current density.

When mixing was applied, formation of flocs and development of a supernatant layer was observed as soon as the mixing was stopped. Results from these experiments are shown in Fig. 3 (b), as a function of sedimentation time after mixing was stopped. It is apparent that COD removal in post-EC-EF increased with increasing applied current density during EC-EF experiments. Five minutes of mixing at 100 rpm magnetic stirring, with subsequent 15 min of sedimentation, enhanced overall COD removal to $52.9 \pm 4.5\%$ and $59.5 \pm 7.7\%$ for EC-EF experiments with applied current densities of 2 mA/cm² and 3 mA/cm², respectively. This indicates that the aluminum species generated in EC-EF experiments with higher applied current densities were not exhausted during the EC-EF but were carried over in the EC-EF treated effluent. The concentration of aluminum species was high enough for coagulation facilitated by mixing and further COD removal improvement in post-EC-EF experiments. Higher concentration of aluminum, which is related here to increasing current densities, likely facilitated particle destabilization and floc formation resulting in better phase separation [16]. Overall COD removal of $50.7 \pm 4.0\%$ and $59.3 \pm 8.7\%$ was achieved as soon as after 5 min of sedimentation after 5 min of mixing for EC-EF experiments with applied current densities of 2 mA/cm²and 3 mA/cm², respectively. COD removal was largely enhanced in early-stage sedimentation after mixing for the higher applied current densities due to a rapidly developed phase separation. Additional COD removal facilitated by mixing was less than 1% for EC-EF experiments with applied current densities of 1 mA/cm². After 24 h of sedimentation, overall COD removal reached $54.4 \pm 4.5\%$ , $66.3 \pm 4.6\%$ and $71.6 \pm 4.3\%$ for applied current densities during EC-EF experiments of 1 mA/cm², 2 mA/cm² and 3 mA/cm², respectively.

The effect of aluminum concentration on COD removal in open-circuit control experiments is shown in the Supporting Information (Fig. S3). Overall COD removal for open-circuit controls with the highest amended aluminum concentration ( $3.74 \pm 0.06$ mg Al/L) was $55.1 \pm 3.1\%$ and $70.4 \pm 1.2\%$ after 5 min and 24 h of sedimentation, respectively, after 5 min of mixing. Although aluminum was supplemented to RW from the beginning of control experiments, which resulted in a greater amount of aluminum species to start with in the EC-EF cell, COD removals were similar to those for experiments with applied current densities of 3 mA/cm², which generated $3.24 \pm 0.65$ mg Al/L by the end of the 15-minute period. In post-EC-EF experiments using effluent from open-circuit experiments with 1.56 mg Al/L, sedimentation of 5 min after 5 min of mixing improved COD removal by 6.2%. In comparison, 26.7% additional COD removal was achieved for effluent with the same aluminum concentration from EC-EF experiments with applied current densities of 2 mA/cm². The rapid COD removal in electrochemically-treated effluent suggests a favored flocculation in post EC-EF experiments, likely resulting from destabilization of particles in the EC-EF cell facilitated by the electric field and flotation turbulence [17].

3.3. Effect of mixing duration in post electrocoagulation-flotation experiments

A series of experiments was performed with 2 mA/cm² of applied current density in the EC-EF cell followed by varying durations of 100 rpm mixing in the post-EC-EF tank; the results from these experiments are summarized in Fig. 4. The estimated velocity gradient G value at stirring rate of 100 rpm was 78.5 s⁻¹. The resulting GT values for mixing durations of 5, 10, and 15 min were 23600, 47,100 and 70700, respectively (see Supporting Information for calculations), which were within the recommended range for coagulation-flocculation of 24000–84000 [18]. Results from post-EC-EF experiments with applied current densities during EC-EF of 1 mA/cm² and 3 mA/cm² are included in the Supporting Information (Fig. S4). Longer mixing duration correlated with higher COD removal for all applied current densities. For EC-EF experiments with applied current densities of 2 mA/cm², overall COD removal after post-EC-EF experiments was $69.5 \pm 4.8\%$ after 15 min of mixing at 100 rpm and a subsequent 15 min of sedimentation. Further sedimentation to up to 24 h allowed the overall COD removal to reach $72.2 \pm 5.3\%$ , which is the highest COD removal obtained in this study. No mixing resulted in the smallest COD removal at $32.3 \pm 8.4\%$ and $51.4 \pm 10.1\%$ after 15 min and 24 h of sedimentation. In comparison, DAF effluent sampled while the influent used in this study had $1,816 \pm 27.9$ mg COD/L, representing a 69.9% COD removal during DAF treatment. EC-EF experiments with applied current densities of 2 mA/cm² followed by post EC-EF experiments of 15 min of mixing at 100 rpm and a subsequent 15 min of sedimentation provided a similar COD removal with the advantages of no pH adjustment and no coagulant addition. The similarity in COD removals after 15 min and 24 h of sedimentation for EC-EF experiments with applied current densities of 2 mA/cm² followed by post EC-EF experiments of 15 min of mixing at 100 rpm also suggested a negligible COD removal from biological activity within the time course of the experiments.

Fig. 4. Effect of mixing duration on COD removal in post-EC-EF experiments with applied current densities during EC-EF experiments of 2 mA/cm2. Error bars show standard deviations in all data points as determined from three replicate experiments for each mixing duration condition.

Most of the COD removal during sedimentation occurred within the first 5 min. Extended sedimentation did not significantly further improve COD removal. For 15 min of 100 rpm magnetic stirring, sedimentation of 24 h only improved COD removal by 2.7% compared to sedimentation of 15 min. On the other hand, shorter mixing duration showed a slower improvement in COD removal with respect to sedimentation duration. After 5 min of 100 rpm magnetic stirring and a subsequent 15 min of sedimentation, $52.9 \pm 4.5\%$ of initial COD was removed. A further COD removal of 13.3% was observed by increasing sedimentation duration from 15 min to 24 h. The slow sedimentation for shorter mixing duration indicated a slow settling velocity, likely resulting from smaller sizes of flocs. Longer mixing duration allowed growth of flocs and rapid separation of sediments from aqueous phase once mixing was stopped [16].

3.4. Distribution and composition of LCFA

During all EC-EF experiments, a floating layer of white foam developed on top of the aqueous phase because of the inherent H₂ flotation treatment, which is as per expectation of separating fats from RW within the electrochemical cell itself, even though a majority of the remaining COD was removed in the sludge in the post-EC-EF cell. This layer was collected for LCFA analysis and is referred to as the “fats” collected from the EC-EF cell. Sediment generated during EC-EF and in post-EC-EF experiments with different mixing durations were analyzed separately and results were combined as the “sludge” for mass balance determination. “Effluent” is defined as the aqueous phase after post-EC-EF treatment. The “unknown” fraction of LCFA distribution represented the difference between initial RW LCFA mass and the sum of the mass in effluent, fat, and sludge after treatment. Several LCFAs were detected in EC-EF treated effluent, fats, and sludge samples, including myristic acid, palmitic acid, and stearic acid, which are saturated aliphatic fatty acids with 14, 16, and 18 carbon-atom chain lengths respectively, and elaidic acid, oleic acid, and linoleic acid, which are unsaturated 18-carbon chain fatty acids. Among unsaturated LCFA, elaidic acid and oleic acid are a pair of isomers with one unsaturated bond, with elaidic acid being trans oriented and oleic acid being cis oriented. Linoleic acid has two unsaturated bonds in cis geometry.

Shown in Fig. 5 is the distribution of individual LCFA among the different phases collected for experiments with applied current density during EC-EF of 2 mA/cm² and followed by post-EC-EF treatment with mixing at 100 rpm. To reflect the mass balance of initial LCFAs used in the EC-EF cell which were later distributed in post-EC-EF tests of different mixing durations, effluent and sludge fractions were estimated based on the combined results for all mixing durations. Abundant LCFAs were trapped in the fats collected from the EC-EF cell, ranging from 43% to 64% of individual LCFA mass that was initially provided in the influent RW. The most efficient removal among LCFAs by flotation treatment in the fats layer was 64% for trans unsaturated elaidic acid. Cis unsaturated oleic acid and linoleic acid were removed to a lower degree in the fats collected by flotation treatment. The high occurrence of LCFA in the fats collected from the EC-EF cell suggests that the reuse of those fats for further product recovery in the rendering process is possible. The RW effluent after post-EC-EF treatment accounted for 20–37% of individual LCFA leaving treatment. Concentrations of saturated LCFA in the effluent increased with increasing carbon chain length. Both cis oleic acid and linoleic acid preferably stayed in the aqueous phase compared to other LCFAs, while trans elaidic acid was the least prevalent in the effluent among 18-carbon LCFAs. The sludge was not a significant carrier of any LCFA, suggesting that most of the fat is removed by flotation during EC-EF and not coagulation during post-EC-EF treatment.

Fig. 5. Distribution of LCFA mass among EC-EF treated rendering wastewater fractions with applied current density of 2 mA/cm2 and mixing of 100 rpm, with effluent and sludge results being combined from all mixing durations. Each data point represents an average of measurements from two experiments.

An increasing amount of sludge was generated and more LCFAs were associated with sludge in EC-EF experiments with applied current density of 3 mA/cm², for which results are summarized in the Supporting Information (Fig. S5). From Fig. 5 and Fig. S5, it can also be concluded that the unknown fraction of LCFA was inversely correlated with the fraction in sludge. We hypothesize that uncertainty in the unknown fraction is predominantly associated with challenges of accurately determining LCFA mass in the sludge. A lower unknown fraction with increasing sludge production is possible evidence of underestimation of the contribution of LCFA in sludge. EC-EF experiments with lower applied current densities generated lower concentrations of aluminum species for formation of flocs, which likely resulted in a slower separation of sediments from aqueous phase. Compared to rapidly occurring flocculation in the higher current density experiments, the possibility of underestimating sludge volume by early sampling when floating flocs were not fully settled is higher and therefore also increases the uncertainty in the determined LCFA mass associated with sludge.

Aluminum was detected in fats collected from the EC-EF cell with concentrations of $9.48 \pm 0.21$ , $11.76 \pm 0.69$ and $12.00 \pm 0.27$ mg Al/L for experiments with applied current densities of 1 mA/cm², 2 mA/cm² and 3 mA/cm², respectively. The trend with increasing current density was not as clear as with aqueous aluminum concentration, which was positively correlated with current density, suggesting a likely minor contribution of aluminum in fats from electrochemically induced aluminum species. It is hypothesized that inherent aluminum in RW ( $0.64 \pm 0.06$ mg/L) itself, that adsorbed on free fatty acids prior to arriving EC-EF cell, was concentrated along with free fatty acids by electroflotation.

The composition of LCFA in each phase among the different streams is compared in Fig. 6 to determine any redistribution of LCFA by EC-EF treatment with applied current density of 2 mA/cm², and to compare with DAF treatment. EC-EF treated effluent had the same composition and similar proportion of LCFA as the raw RW. On the other hand, less saturated myristic acid, palmitic acid, and stearic acid represented a large fraction of total LCFA in DAF-treated effluent. Elaidic acid was not detected at all in the DAF-treated effluent. Compared to the cis isomers, trans unsaturated LCFA have better oriented molecular structure, which are similar to their corresponding straight-chain saturated LCFA. The ordered molecular structure favors the compactness by intermolecular forces and similarity in chemical properties [19], [20]. The disappearance of trans elaidic and a lower proportion of saturated LCFA after DAF treatment is likely due to the physical separation of the floating phase and/or decomposition in the presence of pressurized air. Contrarily, H₂ gas-based flotation in EC-EF preserved trans elaidic acid in the effluent. Cis oleic acid and linoleic acid were of higher proportion in both DAF-treated effluent and EC-EF-treated effluent, which agrees with the previous observation that cis LCFA showed higher affinity to aqueous phase compared to other LCFAs. It is hypothesized from the results that during DAF treatment when pressurized air is provided, cis unsaturated LCFAs are more resistant to abiotic oxidation than trans elaidic acid. Fats and sludge generated by EC-EF treatment were composed of the same distribution of LCFA as raw RW. In terms of rendering product recovery and utilization of concentrated fats, EC-EF treatment preserved LCFA composition in the fats collected and thus provides a better alternative to DAF to keep the returned fat quality consistent with that initially produced.

Fig. 6. Composition of LCFA in raw RW, DAF-treated effluent and the different phases from EC-EF treatment with applied current density of 2 mA/cm2 followed by mixing of 100 rpm during post-EC-EF. The proportion of LCFA was calculated based on the mass distribution of individual LCFA, with EC-EF effluent and sludge results being combined from all mixing durations during post-EC-EF treatment. Each data point represents an average of measurements from two experiments.

4. Conclusion

COD removal from RW by EC-EF was evaluated in an electrochemical cell design with horizontally-assembled electrodes. EC-EF was able to electrochemically produce aluminum species for solids removal. Inherent flotation treatment concentrated abundant LCFA in the floated fat layer. LCFA composition was similar to the raw RW in all EC-EF treated phases. Further coagulation was possible with surplus aluminum species generated in EC-EF with higher applied current densities. The highest overall COD removal ( $72.2 \pm 5.3\%$ ) was achieved by EC-EF with an applied current density of 2 mA/cm² followed by 15 min of mixing at 100 rpm and 24 h of sedimentation. With competitive treatment efficiency and advantages in preserving rendering fats with minimal changes in quality, the implementation of EC-EF in RW primary treatment offers a potentially valuable opportunity for the rendering industry and perhaps other industries where high-strength wastewater containing fats, oils, and/or grease is generated and requires primary separation.

5. Acknowledgments

The authors would like to acknowledge the Fats and Proteins Research Foundation for providing the funding to conduct this study through the Animal Co-products Research and Education Center at Clemson University.

6. References

[1] B. Kiepper, Characterisation of Poultry Processing Operations, Wastewater Generation and Wastewater Treatment (2003) 1–141.
[2] US EPA, Technical development document for the final effluent limitations guidelines and standards for the meat and poultry products point source category (40 CFR 432), (2005). https://www.epa.gov/sites/production/files/2015-11/documents/meat-poultry-products_tdd_2004_0.pdf.
[3] G.L. Sindt, Environmental issues in the rendering industry, (2006). http://assets.nationalrenderers.org/essential_rendering_environmental_impact.pdf.
[4] L. Heninger, Wastewater Treatment Manual Poultry Processing Facilities U.S. Poultry & Egg Association, Manag. As Mission Nonprofit Manag. Sustain. Chang.(2017), https://doi.org/10.1201/9781315180656-4.
[5] R.K.L. Yap, M. Whittaker, M. Diao, R.M. Stuetz, B. Jefferson, V. Bulmus, W. L. Peirson, A.V. Nguyen, R.K. Henderson, Hydrophobically-associating cationic polymers as micro-bubble surface modifiers in dissolved air flotation for cyanobacteria cell separation, Water Res. 61 (2014) 253–262, https://doi.org/10.1016/j.watres.2014.05.032.
[6] K.O. Iwuozor, Prospects and Challenges of Using Coagulation-Flocculation method in the treatment of Effluents, Adv. J. Chem. A. 2 (2019) 105–127. https://doi.org/10.29088/sami/ajca.2019.2.105127.
[7] J.T. Nwabanne, C.C. Obi, Abattoir Wastewater Treatment by Electrocoagulation Using Iron Electrodes, Der Chem. Sin. 8 (2017) 254–260.
[8] D.T. Moussa, M.H. El-Naas, M. Nasser, M.J. Al-Marri, A comprehensive review of electrocoagulation for water treatment: Potentials and challenges, J. Environ. Manage. 186 (2017) 24–41, https://doi.org/10.1016/j.jenvman.2016.10.032.
[9] M. Kobya, E. Senturk, M. Bayramoglu, Treatment of poultry slaughterhouse wastewaters by electrocoagulation, J. Hazard. Mater. 133 (1-3) (2006) 172–176, https://doi.org/10.1016/j.jhazmat.2005.10.007.
[10] I. Papadopoulos, Effect of sulphate waters on soil salinity, growth and yield of tomatoes, Plant Soil. 81 (1984) 353–361. https://www.jstor.org/stable/42934681.
[11] E. Bazrafshan, L. Mohammadi, A. Ansari-Moghaddam, A.H. Mahvi, Heavy metals removal from aqueous environments by electrocoagulation process – A systematic review, J. Environ. Heal. Sci. Eng. 13 (1) (2015), https://doi.org/10.1186/s40201-015-0233-8.
[12] ˙I.A. S¸ engil, M. ozacar, ¨ Treatment of dairy wastewaters by electrocoagulation using
mild steel electrodes, J. Hazard. Mater. 137 (2) (2006) 1197–1205, https://doi.org/10.1016/j.jhazmat.2006.04.009.
[13] M. Asselin, P. Drogui, H. Benmoussa, J.-F. Blais, Effectiveness of electrocoagulation process in removing organic compounds from slaughterhouse wastewater using monopolar and bipolar electrolytic cells, Chemosphere. 72 (11) (2008) 1727–1733, https://doi.org/10.1016/j.chemosphere.2008.04.067.
[14] R.M. Ziels, D.A.C. Beck, M. Martí, H.L. Gough, H.D. Stensel, B.H. Svensson, Monitoring the dynamics of syntrophic β-oxidizing bacteria during anaerobic degradation of oleic acidby quantitative PCR, FEMS Microbiol. Ecol. 91 (2015) 1–13, https://doi.org/10.1093/femsec/fiv028.
[15] A.M. Burja, H. Radianingtyas, A. Windust, C.J. Barrow, Isolation and characterization of polyunsaturated fatty acid producing Thraustochytrium species: Screening of strains and optimization of omega-3 production, Appl.
Microbiol. Biotechnol. 72 (6) (2006) 1161–1169, https://doi.org/10.1007/s00253-006-0419-1.
[16] C.Y. Teh, P.M. Budiman, K.P.Y. Shak, T.Y. Wu, Recent Advancement of Coagulation-Flocculation and Its Application in Wastewater Treatment, Ind. Eng. Chem. Res. 55 (16) (2016) 4363–4389, https://doi.org/10.1021/acs.iecr.5b04703.
[17] Y. Liu, X. Zhang, WenMing Jiang, MingRui Wu, ZhiHui Li, Comprehensive review of floc growth and structure using electrocoagulation: Characterization, measurement, and influencing factors, Chem. Eng. J. 417 (2021) 129310, https://doi.org/10.1016/j.cej.2021.129310.
[18] Z. Zhang, D. Liu, D. Hu, D. Li, X. Ren, Y. Cheng, Z. Luan, Effects of slow-mixing on the coagulation performance of polyaluminum chloride (PACI), Chinese, J Chem.Eng. 21 (3) (2013) 318–323, https://doi.org/10.1016/S10049541(13)60463-2.
[19] H.-H. Tsai, J.-B. Lee, H.-S. Li, T.-Y. Hou, W.-Y. Chu, P.-C. Shen, Y.-Y. Chen, C.- J. Tan, J.-C. Hu, C.-C. Chiu, Geometrical effects of phospholipid olefinic bonds on the structure and dynamics of membranes: A molecular dynamics study, Biochim. Biophys. Acta – Biomembr. 1848 (5) (2015) 1234–1247, https://doi.org/10.1016/j.bbamem.2015.02.016.
[20] C. Scheidler, J. Sobotta, W. Eisenreich, G. W¨achtersh¨auser, C. Huber, Acids by Aqueous, One-Pot Carbon Fixation : Possible Relevance for the Origin of Life, Nat. Publ. Gr. (2016) 1–7, https://doi.org/10.1038/srep27595.