ABSTRACT
This study investigated the treatment of wastewater from tomato paste (TP) production using electrocoagulation (EC) and electrooxidation (EO). The effectiveness of water recovery from the pretreated water was then investigated using the membrane process. For this purpose, the effects of independent control variables, including electrode type (aluminum, iron, graphite, and stainless steel), current density (25–75 A/m2), and electrolysis time (15–120 min) on chemical oxygen demand (COD) and color removal were investigated. The results showed that 81.0% of COD and 100% of the color removal were achieved by EC at a current density of 75 A/m2, a pH of 6.84 and a reaction time of 120 min aluminum electrodes. In comparison, EO with graphite electrodes achieved 55.6% of COD and 100% of the color removal under similar conditions. The operating cost was calculated to be in the range of $0.56–30.62/m3. Overall, the results indicate that EO with graphite electrodes is a promising pretreatment process for the removal of various organics. In the membrane process, NP030, NP010, and NF90 membranes were used at a volume of 250 mL and 5 bar. A significant COD removal rate of 94% was achieved with the membrane. The combination of EC and the membrane process demonstrated the feasibility of water recovery from TP wastewater.
HIGHLIGHTS
The treatment of tomato paste wastewater via electrocoagulation (EC) and electrooxidation (EO) processes was examined.
The membrane process was used as the post-treatment.
Aluminum and graphite electrodes were chosen for EC and EO, respectively.
INTRODUCTION
Turkey is one of the five tomato paste (TP)-producing countries in the world. With the expansion of livestock in the extensive production of TP, the treatment of TP wastewater (TPW) has become an important issue. TPW is characterized by high chemical oxygen demand (COD) of 2,800–15,500 mg/L, biochemical oxygen demand (BOD) of 1,750–7,950 mg/L, total Kjeldahl nitrogen of 48–340 mg/L, and ammonium-nitrogen (NH4+-N) of 21–235 mg/L (Gohil & Nakhla 2006). According to Mannapperuma et al. (1993), about 3.5 m3 of wastewater is generated per tonne of TP produced. However, studies on the chemical and microbiological properties and treatment options of TP are very limited. So far, only a few methods have been investigated including UV oxidation (Mahoney et al. 2018), combined biological treatment (Gohil & Nakhla 2006), and nanofiltration (Alghooneh et al. 2016) as they are highly efficient, cost-effective, and environmentally friendly. Therefore, there is a need to develop an alternative treatment process that offers high efficiency with less duration and chemical consumption as well as being easy to install.
Electrocoagulation (EC), which combines the advantages of coagulation, flotation, and electrochemistry, has been extensively studied in recent years due to its ecological versatility and has been adapted for the treatment of food processing wastewater (Sardari et al. 2018; Akarsu et al. 2021, Arslan et al. 2023). During the EC process, metal hydroxides, whose density is higher than that of water, settle, while floc floats by bubbles due to the neutralization of the surface charge of the pollutants (Qasem et al. 2021). The superior electrode is one of the control parameters that not only affects the generation of in situ coagulants but is also associated with the cost (Delil & Gonen 2019). In general, aluminum (Al), iron (Fe), magnesium, steel, and zinc have been used as electrode materials (Safwat 2020; Ebba et al. 2021; Figueiredo et al. 2022). Among them, Al and Fe are the most widely used due to their various properties such as low cost, non-toxicity, and high valence, which lead to the efficient removal of pollutants (Hakizimana et al. 2017). In electrooxidation (EO), in contrast to EC, the main function of the anode is the oxidative degradation of organic substances by the hydroxyl radicals generated at high reaction rates. Consequently, non-consumable electrodes are used in EO to attenuate pollutants directly or indirectly through the generation of oxidizing agents in solution (Lynn et al. 2019).
Membrane treatment systems are one of the best systems for removing pollutants and recycling water. The membrane system is widely used because it does not require chemicals for purification, is easy to operate, and has good separation properties due to its pore diameters (Majidi et al. 2022). The performance of the membrane system varies depending on the characterization of wastewater, pressure, and membrane type. While the pollutants that cannot pass due to the pore diameter on the membrane surface remain concentrated in the upper part, the part that passes through the membrane is taken as a filtrate (Dadari et al. 2022). While suspended solids and dissolved substances with a high molecular weight are retained in the membrane, water and substances with a low molecular weight pass through the membrane (Esfahani et al. 2019). One of the major disadvantages of the membrane system is the contamination problem. Clogging problems due to the contamination of the membrane are caused by both biological and chemical substances. If this clogging problem increases, the lifespan of the membrane is shortened (Tummons et al. 2020).
In recent years, EC and EO applications have been extensively studied for the treatment of industrial sewage such as textile industry (Can et al. 2006), olive mill (Inan et al. 2004), brewery (Wysocka & Masalski 2018), dairy (Reilly et al. 2019), acid mine (Alam et al. 2022), furniture industry (Vicente et al. 2023), and slaughterhouse (Adou et al. 2022). To our knowledge, however, despite these advantages of electrochemical methods, there is little literature research on the treatment of TP. The aim of this study is therefore to treat TPW using EC and EO processes. Subsequently, it is aimed to recover water by the membrane process after pretreatment under optimum conditions. In this context, control variables such as electrode combination (Al–Al, Fe–Fe, G–G, and SS–SS), current density (25, 50, and 75 A/m2), and electrolysis time (15, 30, 45, 60, and 120 min) were analyzed to determine the optimum operating conditions for COD and color removal performance as well as on the total operating cost. For the membrane process, NP030, NP010, and NF90 membranes were used with a volume of 250 mL, a pressure of 5 bar, and an operation time of 2 h. The flux was collected every minute, and COD analysis was performed on the filtrate sample.
MATERIALS AND METHODS
TPW characterization and analysis
TPW used in this study was taken from a company in Mersin, Turkey. It was collected from the sewerage system in September 2022 and cooled down to 4 °C for further procedure. The specific characteristics of TPW are given in Table 1.
TPW characteristics
Parameter . | Unit . | Value . |
---|---|---|
pH | – | 6.84 ± 0.12 |
COD | mg/L | 1,260 ± 82 |
Color | Pt–Co | 150 ± 14 |
Conductivity | μS/cm | 1,490 ± 10 |
Parameter . | Unit . | Value . |
---|---|---|
pH | – | 6.84 ± 0.12 |
COD | mg/L | 1,260 ± 82 |
Color | Pt–Co | 150 ± 14 |
Conductivity | μS/cm | 1,490 ± 10 |
The changes of COD and the color decrease depending on the electrode (Al–Al, Fe–Fe, G–G, and SS–SS), current density (25, 50, and 75 A/m2), and electrolysis time (15, 30, 45, 60, 90, and 120 min) were observed. pH and conductivity were measured by a multi-parameter instrument (Hach-Lange HQ40d). COD of the samples was determined by the dichromatic closed reflux method of the standard methods. The color was determined at 455 nm with a spectrophotometer (Hach-DR 6000) and calibrated using the platinum–cobalt method (Pt/Co). All analyses were performed in duplicate and mean values were reported.
Experimental setup
Electrochemical experiment
Membrane experiment
General properties of the membranes used in the study
Membrane . | Material . | Maximum temperature (°C) . | pH . | Pore size (Da) . | Flux . | Contact angle (°) . | Firm . |
---|---|---|---|---|---|---|---|
NP010 | Polyethersulfone (PES) | 95 | 0–14 | ∼1,000 | 24–30 LMH/5 bar | 47.3 | Microdyn Nadir™ |
NP030 | Polyethersulfone (PES) | 95 | 0–14 | ∼400–600 | 9–12 LMH/5 bar | 88.5 | Microdyn Nadir™ |
NF90 | Polyamide (PA) | 45 | 2–11 | ∼90–180 | 12–15 LMH/5 bar | 72 | DOW Filmtec |
Membrane . | Material . | Maximum temperature (°C) . | pH . | Pore size (Da) . | Flux . | Contact angle (°) . | Firm . |
---|---|---|---|---|---|---|---|
NP010 | Polyethersulfone (PES) | 95 | 0–14 | ∼1,000 | 24–30 LMH/5 bar | 47.3 | Microdyn Nadir™ |
NP030 | Polyethersulfone (PES) | 95 | 0–14 | ∼400–600 | 9–12 LMH/5 bar | 88.5 | Microdyn Nadir™ |
NF90 | Polyamide (PA) | 45 | 2–11 | ∼90–180 | 12–15 LMH/5 bar | 72 | DOW Filmtec |
Cost calculation




RESULTS AND DISCUSSION
Electrochemical experiment
COD and color removal
Effect of operating variables on COD removal by using (a) Al–Al, (b) Fe–Fe, (c) G–G, and (d) SS–SS.
Effect of operating variables on COD removal by using (a) Al–Al, (b) Fe–Fe, (c) G–G, and (d) SS–SS.
Operating time is another key variable in electrochemical treatment processes, as the coagulant concentration generated by electrolysis is time-bound and directly dependent on the electrical charge supplied per unit volume of EC, while the desired oxidation of substances with the gases O2 and H2 was achieved by the cathodic activity of the anode of EO (Khan et al. 2023). Metallic polymer types in the form of hydroxides produce more hydroxyl radicals over a longer period of time, resulting in higher removal efficiency (Asaithambi et al. 2016). However, the pollutant removal efficiency increases with the duration of electrolysis until it reaches saturation under optimal conditions (Can et al. 2014). The duration of operation is closely related to the current density and therefore needs to be optimized, as the desired removal efficiency depends on the availability of the coagulation floc for the desired pollutant removal. Therefore, the effect of electrolysis time on the process has been extensively studied in the literature. For example, Deveci et al. (2019) used a laboratory-scale EC to remove chromium and organics from tannery wastewater under different operating parameters and found that increasing the treatment time significantly improved removal. Similar results were found for the removal of other pollutants, e.g. total phenol (Arslan et al. 2023), surfactants (Akarsu et al. 2022), inorganics (Akbay et al. 2019), and COD (Delil & Gönen 2019).
After 15 min of electrochemical treatment, the color removal for the electrode pairs Al, Fe, and SS was 100%, while the maximum color removal for the G pairs was achieved in 45 min. Similarly, several studies also report that different Fe and Al electrode arrangements were effectively used in color removal (Majlesi et al. 2016; Dizge et al. 2018). As expected, in the case of Al, Fe, and SS electrode pairs, not only phase transfer but also redox reactions can contribute to color removal (Kabdasli et al. 2009).
Effect of electrolysis time on pH changes
Variation of pH in the reactor by using Al–Al, Fe–Fe, G–G, and SS–SS.
The ionic distribution of aluminum and iron when dissolved in water gives us much to understand, as the coagulation process is pH dependent. Al(OH)2 is the first metal hydroxide to be formed at acidic pH. When the OH− ion concentration increases, Al(OH)3 tends to form equilibria. Moreover, the isoelectric points of the metal hydroxides for Al and Fe electrodes are close to each other at neutral pH or below (Pourabdollah 2021). Cationic iron species such as Fe3+, Fe(OH)2, and Fe(OH)3 may be present in the solution if the hydroxides do not reach a sufficient solubility constant at acidic pH. The iron hydrolysis balance can cause several equilibria compared to the aluminum hydrolysis balance due to the divalent iron. Most importantly, the reduction potential of the Fe2+ ions can take place if sufficient oxygen is present in the water.
The increase in the concentration of OH− ions during electrolysis leads to an increase in the pH value of the wastewater for Al–Al and Fe–Fe electrodes. However, for EO, the increase is less pronounced or in some cases a decrease in pH is observed, as in this study. In addition, the release of CO2 due to hydrogen bubbles that can form on the cathode surface is another reason for the increase in the pH value (Lakshmi & Sivashanmugam 2013). The decrease in pH during electrooxidation can be explained by a possible higher H+ production compared to OH−, leading a shift in pH toward the acidic region.
OC evaluation
OCs are an important parameter in wastewater treatment technologies when considering their application on an industrial scale. In this study, the calculation of process cost was therefore estimated only for energy and anode consumption, excluding sludge treatment and maintenance costs. The calculations were based on previous studies (Villasenor-Basulto et al. 2022).
Treatment cost of TPW and the efficiencies of EC and EO processes at different current densities (initial pH of 6.84 and electrolysis time of 120 min)
j (A/m2) . | ![]() | Cost ($/m3) . | Cost ($/kg COD) . |
---|---|---|---|
Al–Al | |||
25 | 40.5 | 0.56 | 1.10 |
50 | 63.5 | 15.69 | 19.60 |
75 | 81.0 | 30.62 | 30.02 |
Fe–Fe | |||
25 | 41.3 | 1.88 | 3.64 |
50 | 46.7 | 5.55 | 9.44 |
75 | 51.1 | 33.39 | 51.85 |
G–G | |||
25 | 28.9 | 0.07 | 0.20 |
50 | 35.9 | 0.69 | 1.54 |
75 | 55.5 | 2.86 | 4.08 |
SS–SS | |||
25 | 35.4 | 0.11 | 0.24 |
50 | 43.8 | 1.56 | 2.83 |
75 | 50.5 | 8.22 | 12.92 |
j (A/m2) . | ![]() | Cost ($/m3) . | Cost ($/kg COD) . |
---|---|---|---|
Al–Al | |||
25 | 40.5 | 0.56 | 1.10 |
50 | 63.5 | 15.69 | 19.60 |
75 | 81.0 | 30.62 | 30.02 |
Fe–Fe | |||
25 | 41.3 | 1.88 | 3.64 |
50 | 46.7 | 5.55 | 9.44 |
75 | 51.1 | 33.39 | 51.85 |
G–G | |||
25 | 28.9 | 0.07 | 0.20 |
50 | 35.9 | 0.69 | 1.54 |
75 | 55.5 | 2.86 | 4.08 |
SS–SS | |||
25 | 35.4 | 0.11 | 0.24 |
50 | 43.8 | 1.56 | 2.83 |
75 | 50.5 | 8.22 | 12.92 |
Currently, there is no study in the literature on the electrochemical treatment of wastewater from TP. Therefore, the costs were compared with current literature studies. For example, Isik et al. (2020) reported that the cost of treating pistachio wastewater with the EO process via boron-doped diamond (BDD) electrodes is about $4.38/m3. In another study by Vidal et al. (2019), the EO could be applied for $1.4/m3.
Membrane experiment
Flux chart of membrane treatment after pretreated water (experimental conditions: ΔP: 5 bar, pH: 7.0, volume: 250 mL, time: 2 h).
Flux chart of membrane treatment after pretreated water (experimental conditions: ΔP: 5 bar, pH: 7.0, volume: 250 mL, time: 2 h).
The initial and steady-state fluxes of pretreated wastewater for the NP030 membrane were 8 and 6.7 L/m2·h, respectively. For the NP010 and NF90 membranes, the initial fluxes are 20 and 8 L/m2·h, while the stable fluxes are 15 and 5 L/m2·h, respectively. Remarkably, no significant decrease in flux was observed for the membranes, which can be attributed to the lower contaminant load of the pretreated water. While the COD value of pretreated wastewater was 240 mg/L, after passing through NP030, NP010, and NP90 membranes, the COD values were 95, 110, and 70 mg/L, respectively. Comparative experimental results of studies on the removal of organic substances and color, as found in the literature, are summarized in Table 4. Additionally, the results of the current study are included.
Comparison of combined electrochemical and membrane filtration processes for various pollutant types
Wastewater Type . | Process . | Operating conditions . | Removal efficiency . | OC (US$/m3) . | Reference . | ||||
---|---|---|---|---|---|---|---|---|---|
pH . | Current density (mA/cm2) . | Retention time (min) . | Electrode material . | Other . | |||||
Wastewater from industrial Cu production | EC + MF | – | 48 | 20 | – | – | Se: 98.7%, Cu: 98%, Pb: 98%, As: 99.9%, Zn: 99.9% and Cd: 99.9% | – | Mavrov et al. (2006) |
Fabric dyeing wastewater | EO + NF | 10.6 | 200 | 60 | Graphite | Flow: 200 m3/d | SS: 100% and color: 99.4% | 2.20 | Yildirim et al. (2023) |
Turnip juice wastewater | EC | 5.4 | 100 | 45 | Al | – | COD: 100% and TPh: 100% | 1.58 | Arslan et al. (2023) |
Turnip juice wastewater | EO | 5.4 | 100 | 45 | BDD-Pt | – | COD: 100% and TPh: 100% | 0.61 | Arslan et al. (2023) |
Steel wastewater | EC + Fenton | 4.0 | 1.5 | 25 | – | Fe2+/H2O2 = 1.5 | COD: 98% and TPh: 100% | – | Malakootian & Heidari (2018) |
Personal care product wastewater | EC + NF/UF/RO | 4.0 | 500 | 180 | SS | – | COD: 89.6%, surfactant: 99.4% and oil-grease: 99.3% | – | Akarsu et al. (2022) |
Wet scrubber wastewater | EO + RO | 8.0 | 150 | 180 | Graphite | – | COD: 95% and TPh: 98% | Belibagli et al. (2022) | |
Textile industry wastewater | EC + NF | 7.0 | 2 V | 24 | Fe | 10 ppm celestine blue dye solution | Dye: 79.4% | – | Saad et al. (2020) |
Poultry processing wastewater | EC + UF | 6.7 | 3 | – | Al | Electrodes spacing: 0.9 cm; flow speed: 1.2 L/min, applied pressure: 70 kPa | BOD: 98%, COD: 92% and TSS: 100% | – | Sardari et al. (2018) |
Hypersaline oil field produced water | EC + MBR | 7.2 | 2.78 | 120 | Al | Electrodes spacing: 5.8 cm | COD: 97% and O&G: 95% | – | Al-Malack & Al-Nowaiser (2018) |
Biologically treated textile effluent | EC + NF + RO | 8.5 | 11 | 90 | Fe | Q: 2 L/min, ΔP: 2 bar | COD: 93% and TDS: 91% | – | Güneş & Gönder (2021) |
Hospital wastewater | EC + UF | 7–8 | 88.5 | 60 | Al | ΔP: 1 bar | TSS: 95.12% TDS: 97.53% BOD: 95.18% COD: 97.88% | 0.89–3.92 | Djajasasmita et al. (2022) |
Hospital wastewater | EC + RO | 7–8 | 110.6 A | 60 | Al | ΔP: 4 bar | TSS: 97.64% TDS: 99.85% BOD: 97.88% COD: 98.38% | 0.93–4.02 | Djajasasmita et al. (2022) |
TP wastewater | EC + NF | 6.84 | 75 | 120 | Al | ΔP: 5 bar | COD: 94% and color: 100% | 0.56–30.62 | This study |
Wastewater Type . | Process . | Operating conditions . | Removal efficiency . | OC (US$/m3) . | Reference . | ||||
---|---|---|---|---|---|---|---|---|---|
pH . | Current density (mA/cm2) . | Retention time (min) . | Electrode material . | Other . | |||||
Wastewater from industrial Cu production | EC + MF | – | 48 | 20 | – | – | Se: 98.7%, Cu: 98%, Pb: 98%, As: 99.9%, Zn: 99.9% and Cd: 99.9% | – | Mavrov et al. (2006) |
Fabric dyeing wastewater | EO + NF | 10.6 | 200 | 60 | Graphite | Flow: 200 m3/d | SS: 100% and color: 99.4% | 2.20 | Yildirim et al. (2023) |
Turnip juice wastewater | EC | 5.4 | 100 | 45 | Al | – | COD: 100% and TPh: 100% | 1.58 | Arslan et al. (2023) |
Turnip juice wastewater | EO | 5.4 | 100 | 45 | BDD-Pt | – | COD: 100% and TPh: 100% | 0.61 | Arslan et al. (2023) |
Steel wastewater | EC + Fenton | 4.0 | 1.5 | 25 | – | Fe2+/H2O2 = 1.5 | COD: 98% and TPh: 100% | – | Malakootian & Heidari (2018) |
Personal care product wastewater | EC + NF/UF/RO | 4.0 | 500 | 180 | SS | – | COD: 89.6%, surfactant: 99.4% and oil-grease: 99.3% | – | Akarsu et al. (2022) |
Wet scrubber wastewater | EO + RO | 8.0 | 150 | 180 | Graphite | – | COD: 95% and TPh: 98% | Belibagli et al. (2022) | |
Textile industry wastewater | EC + NF | 7.0 | 2 V | 24 | Fe | 10 ppm celestine blue dye solution | Dye: 79.4% | – | Saad et al. (2020) |
Poultry processing wastewater | EC + UF | 6.7 | 3 | – | Al | Electrodes spacing: 0.9 cm; flow speed: 1.2 L/min, applied pressure: 70 kPa | BOD: 98%, COD: 92% and TSS: 100% | – | Sardari et al. (2018) |
Hypersaline oil field produced water | EC + MBR | 7.2 | 2.78 | 120 | Al | Electrodes spacing: 5.8 cm | COD: 97% and O&G: 95% | – | Al-Malack & Al-Nowaiser (2018) |
Biologically treated textile effluent | EC + NF + RO | 8.5 | 11 | 90 | Fe | Q: 2 L/min, ΔP: 2 bar | COD: 93% and TDS: 91% | – | Güneş & Gönder (2021) |
Hospital wastewater | EC + UF | 7–8 | 88.5 | 60 | Al | ΔP: 1 bar | TSS: 95.12% TDS: 97.53% BOD: 95.18% COD: 97.88% | 0.89–3.92 | Djajasasmita et al. (2022) |
Hospital wastewater | EC + RO | 7–8 | 110.6 A | 60 | Al | ΔP: 4 bar | TSS: 97.64% TDS: 99.85% BOD: 97.88% COD: 98.38% | 0.93–4.02 | Djajasasmita et al. (2022) |
TP wastewater | EC + NF | 6.84 | 75 | 120 | Al | ΔP: 5 bar | COD: 94% and color: 100% | 0.56–30.62 | This study |
As: arsenic, Cd: cadmium, Cu: copper, MBR: membrane bioreactor, MF: microfiltration, NF: nano-filtration, O&G: oil and grease, Pb: lead, Pt: platinum, RO: reverse osmosis, Se: Selenium, TDS: total dissolved solids, TPh: total phenol, TSS: total suspended solids, UF: ultra-filtration, Zn: zinc.
CONCLUSION
In this work, the electrochemical degradation of COD and color from TP wastewater was investigated on a laboratory scale using commonly available and cost-effective electrode materials such as aluminum, iron, graphite, and stainless steel. The aim was to improve the performance of water recovery by treating the pretreated water with the membrane process. Operating parameters such as current density (25–75 A/m2) and electrolysis time (15–120 min) were also systematically tested. The EC process showed remarkable efficiency, achieving 81% COD and 100% color removal under optimum conditions of 75 A/m2, pH 6.84, and 120 min using Al–Al electrodes. In contrast, the EO process provided 55.6% COD and 100% color removal efficiency under the conditions of 75 A/m2 current density, pH 6.84, and 120 min reaction time with G–G electrode pairs. These results emphasize the feasibility of EC and electrooxidation processes for the degradation of organics from TP wastewater, depending on the selection of suitable electrode pairs and operating parameters. The OCs for Al–Al were calculated in the range of 0.56–30.62$, while for BDD-Pt they were in the range of 0.07–2.87 US$/m³. The study conducted with BDD showed a very low dissolution of the electrode as expected, which explains the significant difference in the cost calculation results. In addition, it is worth noting that increasing the current density has a significant impact on the total energy consumption results and therefore influences the first stages of the cost calculation.
Furthermore, the results presented here provide a basis for future research on a combined process as well as for pilot- and real-scale studies. Possible future studies could test mathematical and statistical techniques to further improve and optimize the processes. Three different membranes (NP030, NP010, and NF90) were used to ensure water recovery in the dead-end filtration system operating at a pressure of 5 bar. The initial treatment of the TP wastewater proved to be effective in preventing membrane fouling and a significant drop in flux. The COD removal rates for the NP030, NP010, and NF90 membranes were determined as 92, 91, and 94%, respectively. The results show that water recovery from TP wastewater is achieved with the EC and membrane hybrid system.
FUNDING
This research was supported by the Mersin University Scientific Research Project (BAP), Project no: 2021-2-TP2-4462.
AUTHORS’ CONTRIBUTIONS
C.A., H.A., and N.D. contributed to the study conception and design. Material preparation and analysis were performed by A.Ş., Z.B., and C.A. C.A. wrote the manuscript with support from N.D., and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
DATA AVAILABILITY STATEMENT
All relevant data are included in the paper or its Supplementary Information.
CONFLICT OF INTEREST
The authors declare there is no conflict.