**Performance of Newly Developed Intermittent Aerator for Flat-Sheet Ceramic Membrane in Industrial MBR System**

**Hiroshi Noguchi 1,\* , Qiang Yin <sup>1</sup> , Su Chin Lee <sup>1</sup> , Tao Xia <sup>1</sup> , Terutake Niwa <sup>1</sup> , Winson Lay <sup>2</sup> , Seng Chye Chua <sup>2</sup> , Lei Yu <sup>2</sup> , Yuke Jen Tay <sup>2</sup> , Mohd Jamal Nassir <sup>2</sup> , Guihe Tao <sup>2</sup> , Shu Ting Ooi <sup>3</sup> , Adil Dhalla <sup>3</sup> and Chakravarthy Gudipati 3,\***


**Abstract:** An intermittent aerator was newly developed to reduce energy costs in a flat-sheet ceramic membrane bioreactor (MBR) for industrial wastewater treatment. Large air bubbles were supplied over a short time interval by the improved aerator technology at the bottom of the flat-sheet membrane. Performance tests for the intermittent aerator were carried out in a pilot system with two cassettes immersed in a membrane tank of the 1-MGD demonstration plant at Jurong Water Reclamation Plant (JWRP) in Singapore. Stable operation was achieved at an average flow of 19–22 LMH with every-2 days MC and peak flow of 27 to 33 LMH with daily MC with reduced air flow for membrane aeration. This indicates that energy costs for membrane aeration can be reduced by using the intermittent aerator. Stable MBR operation with a projected 43% reduction in the overall operating costs could be achieved with an improved aerator together with improved MC regime and membrane cassette.

**Keywords:** ceramic membranes; industrial wastewater; intermittent aerator; water reclamation; MBR

#### **1. Introduction**

The membrane bioreactor (MBR) has been widely used for treatment of domestic sewage and industrial wastewater [1,2]. The MBR system has the advantages of smaller footprint, higher quality of product water, stability of bioprocess and shorter retention times compared to conventionally activated sludge processes [3–5]. Membrane aeration is a key operational parameter and has a significant impact on the energy costs of the MBR system [6–8]. Membrane aeration is applied to prevent fouling on the surface of the membrane, and not only provides oxygen to the biomass in the membrane tank, but also scours the membrane surface and maintains solids in suspension to control the fouling layer on the membrane surface. Many researchers have carried out trials to reduce energy for membrane aeration through novel methods such as controlling aeration for bioprocess [5,9–13] and reducing membrane scouring air [14–17]. On–off control of scouring air has been developed to reduce scouring air [14,15] while pulsed air for membrane scouring has been utilized to reduce the required amount of air [16,17]. An intermittent aerator has been developed by MEIDEN and Separation Technologies Applied Research and Translation (START) Centre to reduce the required air amount for membrane scouring, which results in a reduction of energy consumption in the membrane filtration system. PUB, Singapore's National Water Agency, has developed a one million gallons per day (1 MGD) demonstration plant (DEMO) at Jurong Water Reclamation Plant (JWRP), Singapore, to

**Citation:** Noguchi, H.; Yin, Q.; Lee, S.C.; Xia, T.; Niwa, T.; Lay, W.; Chua, S.C.; Yu, L.; Tay, Y.J.; Nassir, M.J.; et al. Performance of Newly Developed Intermittent Aerator for Flat-Sheet Ceramic Membrane in Industrial MBR System. *Water* **2022**, *14*, 2286. https:// doi.org/10.3390/w14152286

Academic Editors: Martin Wagner and Sonja Bauer

Received: 19 May 2022 Accepted: 20 July 2022 Published: 22 July 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

treat high-strength industrial wastewater from an industrial estate. MEIDEN's Ceramic Membrane Bioreactor (CMBR) process has been installed and operational at the plant since 2014 [14]. It has been demonstrated and reported widely in the literature that the CMBR system can produce high-quality water for reuse application [6,7,9,15–18]. A new intermittent aerator with pulsated, lower air flow was developed by MEIDEN and START Centre, for energy reduction in the CMBR system. The performance evaluation of the intermittent aerator was carried out in the DEMO plant at JWRP to assess the feasibility and extent of energy reduction for the CMBR system. The stability of membrane filtration was investigated with the intermittent aerator at reduced air flow compared with that for the conventional continuous aerator.

#### **2. Materials and Methods**

#### *2.1. Intermittent Aerator Tank Operation Principles*

An intermittent aerator was developed to release large air bubbles at the bottom of the flat-sheet ceramic membrane with a width of 250 mm (Figure 1). The intermittent aerator can release larger bubbles from φ24 mm air release holes as compared to the φ6 mm air release holes from a conventional continuous aerator. Air is continuously supplied from air pipes at the bottom of the intermittent aerator; it is accumulated in the air chamber of the aerator and is released in 0.75 sec with 3.5 sec of air release interval to produce large air bubbles. Air bubbles from the φ24 mm air release holes were split into the gaps of 10 mm to produce mixed air and water flow in the gaps as is shown in Figure 2.

**Figure 1.** Membrane cassette with the intermittent aerator.

**Figure 2.** CFD analysis result for air flow with the intermittent aerator cassette.

Multiphase CFD simulations were performed by the Eulerian model in Ansys® Fluent, Release 19.2 (Ansys, Canonsburg, PA, USA) with water as the primary phase and air as the secondary phase. The boundary conditions assumed for the modelling include: (a) symmetry on the size of the membrane cassette, and (b) air entrainment from bottom to top. The material properties used in the simulation are defined in Table 1 below. The air mass flow rate was converted accordingly to correspond with the air flowrate of 2.2 m3/min and the number of holes that exist throughout the air-scouring pipes. To improve the convergence behavior, an initial solution was computed before solving the complete Eulerian multiphase model. In this work, a mass-flow inlet boundary condition was utilized to initialize the flow conditions. It is also more recommended to set the value of the volume fraction close to the value of the volume fraction at the inlet. At the beginning of the solution, a lower time step i.e., 0.001 s was used and recommended in order to reach convergence. In addition, a sufficient number of iterations (i.e., 30) are required to maintain the convergence level at each time step. The simulation was executed until the end time of 9 s to ensure that there has been a flow stability reaching the top part of the two-level stack membrane unit that would be fully submerged inside water.

**Table 1.** Material properties employed in the CFD simulation.


#### *2.2. Test Systems*

Performance tests of the new aeration system were carried out at the 1-MGD DEMO MBR plant at JWRP. The DEMO plant consisted of an Up-flow Anaerobic Sludge Blanket (UASB) reactor, aeration and membrane separation tanks. Wastewater from the industrial estate was treated in a UASB reactor followed by an MBR system using flat-sheet ceramic membrane [6]. Two commercial membrane cassettes with flat-sheet ceramic membranes manufactured by MEIDENSHA Corporation were used for the performance tests. Each cassette had 400 membrane sheets with an effective area of 200 m<sup>2</sup> . One membrane cassette (Train 1) was equipped with the newly developed intermittent aerator while the other cassette (Train 2) was installed with a conventional continuous aerator, in which air was supplied from 6-mm air release holes on the PVC pipes. The two cassettes were immersed in a membrane tank of the 1-MGD DEMO plant, and they were independently operated by using a control panel. ON–OFF control air supply for the continuous aerator was also carried out in Train 2, for comparison.

#### *2.3. Operating Conditions*

Membrane filtration was carried out under similar conditions for Train 1 and Train 2, except for the membrane aeration condition. The filtration/backwash cycle was fixed at 10 min; backwash duration was 30 s with a flow rate of 1.5 Q (Q = filtration flow). Different flux conditions were set to check membrane stability at average and peak flow conditions (Table 2).


**Table 2.** The operating conditions for the MBR system.

Maintenance cleaning (MC) was carried out with 250 mg/L sodium hypochlorite (NaClO) every two days for the average flow and daily for the peak flow (Table 2). Average flow setting assumed normal operating conditions with full operation of membrane filtration tanks in the membrane systems. Peak flow setting was for higher flux condition when one or two membrane tanks were out of service during maintenance or recovery cleaning. Long-term operation was carried out with average and continuous flux settings to observe filtration stability and impact of peak flow between average flow operation.

Air flow for the intermittent aerator (Train 1) and the continuous aerator (Train 2) was set to 35 m3/h/train and that for the continuous aerator (Train 2) was set to 66 m3/h/train, which corresponded to 5.3 m<sup>3</sup> -air/m<sup>3</sup> -permeate and 10 m<sup>3</sup> -air/m<sup>3</sup> -permeate as specific air demand per permeate (SADp) at 33 L/m2/h (LMH). Air flow setting for the intermittent aerator was 53% of that for conventional continuous aerator.

#### **3. Results and Discussion**

#### *3.1. CFD Analysis for the Intermittent Aerator*

Figure 2 shows CFD analysis results of air flow by the intermittent aerator. The large air bubbles released from the bottom of the membrane cassette were evenly distributed in between the gaps of 10 mm for the membrane sheets. Multiphase CFD simulations were performed with the Eulerian model in Ansys® Fluent, Release 19.2. As an unstructured solver, Ansys® Fluent used internal data structures to assign an order to the cells, faces, and grid points in a mesh and to maintain contact between adjacent cells. Upon generating the grids, the minimum orthogonal quality of the whole CFD model was up to 0.067 which shows a high quality of the mesh, and that quality plays a significant role in the accuracy and stability of the numerical computation. Ansys Fluent allows the user to evaluate the mesh quality through a quantity called orthogonality. Orthogonal quality was computed for cells using cell skewness and the vector from the cell centroid to each of its faces, the corresponding face area vector, and the vector from the cell centroid to the centroids of each of the adjacent cells. If the orthogonality exceeds the limits the numerical error will be higher, which will lead to solution convergence issues. The minimum orthogonal quality for all types of cells should be more than 0.01, with an average value that is significantly higher.

#### *3.2. Performance with Intermittent Aerator*

The trans-membrane pressure (TMP) and the permeate flux data were obtained and collated for five months of continuous operation, to determine the long-term efficiency of the intermittent aerator. Figures 3 and 4 show the profiles of comparison (1) TMP and (2) permeability between the two aerators with permeate flux settings at 22 LMH and 33 LMH, respectively. From Figure 3, base TMP after each MC for the intermittent aerator was lower than a conventional continuous aerator. However, TMP trends became similar for intermittent and continuous aerators after plant shut down on 15th August. Aeration for bioprocess was stopped during the shut down period, which might have caused lower COD or BOD removal in the activated sludge system; this could be attributed to potential membrane fouling when the organics became more dominant after the plant shut down and the reducing impact of the larger bubbles from the intermittent aerator on membrane surface resulted in a similar TMP increase for intermittent and continuous aerators.

**Figure 3.** Comparison of aeration methods at 22 LMH flux: (**a**) TMP trend; (**b**) Permeability trend.

**Figure 4.** Comparison of aeration methods at a flux of 33 LMH: (**a**) TMP trend; (**b**) Permeability trend.

Figure 4 shows TMP and permeability for the intermittent and continuous aeration methods at flux of 33 LMH. TMP increases between MC for the intermittent aerator, ranging from 5 to 23 kPa. This was much higher than that at 22 LMH which has the TMP ranged from 2 to 5 kPa for the intermittent aerator. Both base TMP and TMP after MC have become stable at around 15 to 17 kPa after 24th October, which indicates that stable operation was achieved with the intermittent aerator at 33 LMH.

From Figure 4, it is shown that the performance of TMP and permeability of the intermittent aeration method surpassed the conventional continuous aeration method at a higher flux of 33 LMH. This can be observed from Figure 4a where TMP increase between MC for the intermittent aeration was lower than the conventional continuous aeration method. The Figure 4b also shows a higher consistency in permeability data using the intermittent aeration method than that of the conventional continuous aeration method.

Membrane fouling occurs on the surface of the membrane by formation of a cake layer and inside of the pores by the accumulation of foulants. As membrane air can remove foulants on the surface of the membrane and foulants inside of the membrane pores can be removed by backwashing and chemical cleaning during MC. As air flow on the surface of the membrane mitigates the cake layer formation [18], the difference in increase in TMP between MC for the intermittent and continuous aerator might be attributed to the difference in formation of cake layer on the surface. Shear force was obtained from the CFD analysis. Average shear force in a membrane cassette for the intermittent aerator was 11.22 µST, while that for the continuous aerator was 6.97 µST. Higher shear force resulting from larger air bubbles from the intermittent aerator can strongly reduce foulants on the surface compared with the continuous aerator. This might result in a lower increase in TMP between MC for the intermittent aerator. Reduced formation of cake layer during filtration can affect the efficiency of backwashing, which might result in lower base TMP for the intermittent aerator. Cake layer formation might become faster at higher flux setting of 33 LMH. This might result in more significant suppression of TMP increase with intermittent aerator than the flux condition of 22 LMH.

#### *3.3. Long-Term Operation*

Figure 5a presents the TMP trend graph of the intermittent aerator operation with various flux settings over a period of five months without recovery cleaning (RC). The results indicate that stable operation can be achieved with the intermittent aerator to reduce air consumption while achieving sustainable flux.

SADp was reduced to 5.3 m<sup>3</sup> -air/m<sup>3</sup> -permeate by using the newly developed intermittent aerator, and longer-term stability was shown in the large-scale MBR plant in JWRP. SADp in full-scale plant ranges from 10 to 50 m<sup>3</sup> -air/m<sup>3</sup> -permeate [19]. Some researchers reported achievements of lower SADp ranging from 6 to 9 m<sup>3</sup> -air/m<sup>3</sup> -permeate with polymeric hollow fiber membrane in lab or pilot scale tests for MBR systems [7,10]. Reduction of SADp to 5.3 m<sup>3</sup> -air/m<sup>3</sup> -permeate for the flat-sheet membrane system was similar to the range achieved for the hollow fiber membrane system, suggesting that similar energy for membrane scouring for the flat-sheet ceramic membranes can be almost the same level as that for hollow fiber polymeric membrane. Achievement of lower SADp at 5.3 m<sup>3</sup> -air/m<sup>3</sup> -permeate is a significant development for MBR systems, especially for flat-sheet membranes whose footprint is larger than those for hollow fiber membranes. Flat-sheet ceramic membrane can withstand higher shear force by larger air bubbles, which enables lower SADp in MBR system.

Turbidity is an indicator of performance in water and wastewater applications. Thus, it is monitored regularly to ensure treatment systems are operating effectively. Figure 5b shows that the turbidity of the MBR permeate remained constant at ≤0.1 NTU throughout the testing period, regardless of the wide variations in turbidity for raw wastewater that ranged between 30 NTU and 600 NTU. As shown in Figure 5c, due to the industrial nature of influent, the COD of feed water to the MBR fluctuated with values ranging between 150 and 2500 mg/L. Despite the variable characteristics of the influent, the MBR was able to consistently produce a permeate with COD ≤ 50 mg/L. The results showed the robustness of the aerobic biological process. The consistent permeate quality also indicated stable operation of the ceramic membrane system.

**Figure 5.** TMP trend, turbidity, and COD of MBR: (**a**) TMP trend with intermittent aerator; (**b**) Turbidity; (**c**) COD.

#### *3.4. Analysis of Operating Costs*

Analysis of operation costs was carried out to show the impact of the improved membrane aeration method, and cost analysis was done for large scale MBR with a capacity of 20 MGD as a case study. Energy costs were estimated for the intermittent aerator and compared with the conventional continuous aerator and ON–OFF control of the continuous aerator. The 7 min-ON and 3 min-OFF conditions were used for the analysis of ON–OFF control. Chemical costs included MC with NaClO and recovery cleaning (RC) with NaClO and citric acid. Improved MC regime has been applied in the estimation for developing this improved system with new aeration method. Costs for spare parts were also estimated based on this improved system, which has the improvement in membrane cassette design to eliminate spare parts of tubes and connectors for permeate collection from each membrane sheet. Total life cycle costs were estimated with a forecast of 20 years operation of the MBR system.

Conventional continuous aeration required 66 m3/h/train, which corresponds to 0.258 kWh/m<sup>3</sup> . This results in 5.68 ¢/m<sup>3</sup> in Singapore Dollars converted with an electrical cost of 0.22 ¢/kWh. The energy cost was reduced to 3.98 ¢/m<sup>3</sup> by using 7 min-ON and 3 min-OFF condition, 30% reduction of energy; the energy cost was further reduced to 3.01 ¢/m<sup>3</sup> with intermittent aerator, which required 35 m3/h/train. The chemical cost was 2.14 ¢/m<sup>3</sup> , which was obtained from 44 g/m<sup>3</sup> usage of NaClO (29.5 ¢/kg) and 14 g/m<sup>3</sup> of citric acid (60 ¢/kg) for MC and RC. This was reduced to 1.67 ¢/m<sup>3</sup> with improved regime of MC. The cost for spare parts was 0.47 ¢/m<sup>3</sup> and was reduced to 0.05 ¢/m<sup>3</sup> with the improved membrane cassette. The total cost was 8.29 ¢/m<sup>3</sup> and was reduced to 5.55 ¢/m<sup>3</sup> with ON–OFF control, 4.73 ¢/m<sup>3</sup> with intermittent aerator together with improvement of MC regime and membrane cassette.

Figure 6 shows analysis of the projected operating costs that cover energy, chemical and parts replacement. Energy costs here relate to membrane air scouring. The current operation with continuous air scouring serves as the reference condition at 100%. It can be noted that about 70% of the overall cost is due to energy for membrane air scouring. Through implementation of the newly developed intermittent aeration technology, higher efficiency could be achieved with reduction of the overall cost by approximately 43% as shown in Figure 6. Chemical costs and costs related to parts replacement could also be reduced, as the improved MBR performance would also reduce the frequency of membrane cleanings. This could result in lower life cycle cost (LCC) for MBR system using flat-sheet ceramic membrane.

**Figure 6.** Operating cost comparison between continuous and intermittent aerators.

#### **4. Conclusions**

An intermittent aerator for flat-sheet ceramic membrane was developed to release larger bubbles at the bottom of the membrane cassette. CFD analysis was carried out to observe air distribution in the membrane cassette. Membrane filtration stability with the intermittent aerator was investigated in 1-MGD DEMO plant, in which wastewater from an industrial estate was treated by the combined process of UASB and MBR. TMP increase was suppressed with the intermittent aerator compare with that for the conventional continuous aerator. Five months of operation with the intermittent aerator showed long-term stability of the membrane filtration system with the developed aerator. Reduction in operating costs was estimated to be 43% together with an improved chemical cleaning regime and membrane cassette. A performance test with the developed intermittent aerator in the MBR system for domestic sewage treatment will be carried out to investigate energy reduction for the domestic MBR system.

**Author Contributions:** Conceptualization, H.N. and T.N.; methodology, H.N. and S.T.O.; validation, H.N., Q.Y., T.X. and S.T.O.; formal analysis, H.N.; investigation, H.N., S.C.L., W.L., S.C.C., L.Y., S.C.L., M.J.N. and G.T.; resources, T.N., Y.J.T., A.D. and G.T.; data curation, H.N.; writing—original draft preparation, H.N.; writing—review and editing, C.G. and A.D.; visualization, C.G. and G.T.; supervision, H.N. and T.N.; project administration, H.N. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


### *Article* **Reuse of Textile Dyeing Wastewater Treated by Electrooxidation**

**Cláudia Pinto, Annabel Fernandes , Ana Lopes \* , Maria João Nunes , Ana Baía, Lurdes Ciríaco and Maria José Pacheco**

> Fiber Materials and Environmental Technologies (FibEnTech-UBI), Universidade da Beira Interior, R. Marquês de D'Ávila e Bolama, 6201-001 Covilhã, Portugal; garcia.pinto@ubi.pt (C.P.); annabelf@ubi.pt (A.F.); maria.nunes@ubi.pt (M.J.N.); ana.isabel.mota.baia@ubi.pt (A.B.); lciriaco@ubi.pt (L.C.); mjap@ubi.pt (M.J.P.) **\*** Correspondence: analopes@ubi.pt

> **Abstract:** Wastewater reuse has been addressed to promote the sustainable water utilization in textile industry. However, conventional technologies are unable to deliver treated wastewater with the quality required for reuse, mainly due to the presence of dyes and high salinity. In this work, the feasibility of electrooxidation, using a boron-doped diamond anode, to provide treated textile dyeing wastewater (TDW) with the quality required for reuse, and with complete recovery of salts, was evaluated. The influence of the applied current density on the quality of treated TDW and on the consecutive reuse in new dyeing baths was studied. The ecotoxicological evaluation of the process towards *Daphnia magna* was performed. After 10 h of electrooxidation at 60 and 100 mA cm−<sup>2</sup> , discolorized treated TDW, with chemical oxygen demand below 200 (moderatequality) and 50 mg L−<sup>1</sup> (high-quality), respectively, was obtained. Salt content was unchanged in both treatment conditions, enabling the consecutive reuse without any salt addition. For the two reuse cycles performed, both treated samples led to dyed fabrics in compliance with the most restrictive controls, showing that an effective consecutive reuse can be achieved with a moderate-quality water. Besides the water reuse and complete salts saving, electrooxidation accomplished an ecotoxicity reduction up to 18.6-fold, allowing TDW reuse without severe ecotoxicity accumulation.

> **Keywords:** wool dyeing wastewater; electrooxidation; boron-doped diamond; ecotoxicity; *Daphnia magna*; salts recovery; wastewater reuse

#### **1. Introduction**

The textile industry is one of the most water-intensive industries in the world. In 2015, the worldwide annual consumption of water, in the textile and clothing industry, was estimated to be around 79 billion cubic meters [1]. During the textile manufacturing process, close to 80% of the used water is discharged as wastewater [2]. This process is also responsible for the vast consumption of different chemical products, mainly in the finishing sector (dyeing, bleaching, washing, etc.), and for the generation of large volumes of highly contaminated wastewater, containing residual dyes, dyeing auxiliaries, and high salinity, which must undergo treatment before being discharged into the environment.

The volume and characteristics of the wastewaters generated by the textile industry depend on the type of fabric processed, the industrial processes applied, the type of equipment used and the water consumption [3,4]. Thus, reducing both the water consumption and the discharge of highly contaminated wastewater represents a huge challenge of efficiently managing water with a strategic focus on sustainability. This issue was covered by a large-scale European project AQUAFIT4USE (7th EU Framework Programme), in which one of the objectives was to reduce the use of resources, in particular the use of fresh high-quality water [2]. For the textile finishing sector, the outcomes of this project point to sustainable thinking related to the treatment of textile wastewater for later reuse. One of the proposed approaches was the separation of waste streams, which increases treatability options using individual treatment technologies, and verification of the same

**Citation:** Pinto, C.; Fernandes, A.; Lopes, A.; Nunes, M.J.; Baía, A.; Ciríaco, L.; Pacheco, M.J. Reuse of Textile Dyeing Wastewater Treated by Electrooxidation. *Water* **2022**, *14*, 1084. https://doi.org/10.3390/w14071084

Academic Editors: Martin Wagner and Sonja Bauer

Received: 28 February 2022 Accepted: 28 March 2022 Published: 29 March 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

quality preservation of the textile material when high-quality process water is replaced by lower quality water in the dyeing processes [2].

Considering that the dyeing process consumes more water than other textile processes and, consequently, generates a larger volume of wastewaters, this specific wastewater is a good candidate to be treated for reuse. However, the water used in textile processing, particularly in the dyeing step, must comply with stricter quality requirements than those applied for discharge into the environment [2,4]. Thus, for its reuse to be economically viable, it is necessary to develop efficient treatment processes, which lead to treated wastewaters with the minimum quality necessary for reuse. On the other hand, the study of new technologies for the treatment of these wastewaters should aim not only to remove the color and the organic matter but also to recover salts and other constituents [5].

Several review papers, found in the literature, focus on the different technologies applied for the remediation of textile wastewaters, from biological treatments to physical and chemical processes, with some integrated processes also [6–9]. Among the described technologies, the electrochemical advanced oxidative processes (EAOPs) are shown to be very effective in the treatment of textile dyeing wastewaters (TDW). EAOPs are based on the electrogeneration of highly reactive hydroxyl radicals, which non-selectively react with organic compounds [10,11]. In particular, the electrochemical oxidation (EO) using a high O2-overpotential anode, such as a boron-doped diamond (BDD) electrode, has proved to be very effective in dyes removal, and the mineralization of the organic compounds can be achieved [11,12]. It has also the advantage of not producing sludge or concentrates and, in this case, the high salinity of the textile wastewater may be an advantage for the dyeing wastewater reuse [13–15].

Most studies found in the literature are focused on the development of methodologies that allow the treatment of textile wastewaters to comply with discharge limits. However, in recent years, given the evidence on the potential for reuse within the textile industry, with the prospect of moving towards the goal of zero discharge of industrial wastewaters, the number of studies in this area have increased, namely with a focus on TDW reuse in new dyeing process [4,12–20]. However, to obtain an acceptable quality of dyed fabrics, in some of these studies, the treated wastewater is diluted before its reuse [12,13,17]. On the other hand, considering the high amounts of salts necessary to obtain a good dye fixation to the fiber, another important economic and environmental aspect to be considered is the salt reuse, which can be accomplished by using electrochemical oxidation to treat the TDW. In the literature, only a few studies can be found focusing on the reuse of TDW treated by EO. Riera-Torres et al. [21] reused a simulated TDW, prepared with previously hydrolyzed reactive dyes and sodium sulfate, treated by EO, using titanium covered by platinum oxides as anode. The treated wastewater was diluted before its reuse, and for most of the dyes used, the level of dye degradation in the wastewater was considered non-relevant in the direct reuse, considering that the color removal was enough. Orts et al. [14] studied the reuse of TDW containing a trichrome mixture of reactive dyes and sulfate ions, treated by EO, using a DSA anode, and the decolorized treated wastewater was diluted before its reuse. The electrochemical treatment and reuse of industrial reactive dyeing wastewaters were performed by Sala et al. [12], and the discoloration of dyeing wastewater treatment and reuse was effective, allowing practitioners to save 70% of water and 60% of electrolytes. The water volume lost during the dyeing process, by adsorption into the fiber and by evaporation, was restored with fresh water. This strategy was also used in a different study [22], where the residual dyeing and washing wastewaters were electrochemically treated using a Ti anode covered by Pt oxides. Once again, by the addition of 30% of fresh water, to reconstitute the dyebath, the reduction of organic matter of 49% was obtained. The residual oxidant compounds generated during the EO treatment, namely the chloride active species, had to be removed before the treated wastewater reuse.

For the reuse of TDW treated by EO, some reformulations of the dye concentration and auxiliary chemicals may be necessary, especially in light color dyeing operations [23]. Furthermore, multiple reuses of the treated wastewater would require changes in salt and other auxiliary chemicals to achieve the same fabric color as fresh process water [23].

Considering that ecotoxicity is one of the major problems of textile wastewaters, some studies have evaluated this parameter and verified its reduction through electrochemical oxidation using a BDD anode [24]. Due to the TDW complexity, especially when several organic auxiliaries and salts are used, the evaluation of the ecotoxicity of the treated wastewaters cannot be ignored. Still, very few TDW reuse studies have addressed this issue. Although it may not seem a relevant aspect for TDW reuse, for other reuse applications and for final disposal, it may be crucial. Furthermore, it is important to acknowledge the effect of the consecutive reuse cycles on the wastewater ecotoxicity. Another aspect that has not been much explored in previous studies is the influence of the dyeing bath composition on the treatment–reusing cycle, since, besides the dyes and salts, different organic auxiliaries, e.g., equalizer and humectant agents, are very often utilized, at an industrial level, in the dyeing of woolen fibers [4]. These TDW present a very high content of organic matter of different natures and the discoloration of the wastewater may not be enough for the reuse, as the reduction of the organic matter to increase the quality of the treated wastewater is necessary. In this context, it is important to evaluate the influence of using treated TDW with different qualities, and without using any dilution, on the quality of the color of the dyed fabrics.

To fill some of the gaps presented by the previous studies performed, this work aims to evaluate the feasibility of an EO process using a BDD anode to: (i) treat a TDW from a woolen fabric dyeing process using a trichromatic acid dyes combination in the presence of sulfate salt, equalizer and humectant agents, widely used at the industrial level; (ii) provide a treated TDW with complete salt recovery and a quality level that allows its successful consecutive reuse in new dyeing baths; (iii) significantly reduce the ecotoxicity towards *Daphnia magna* of the wastewaters, even after consecutive reuses. The effect of EO operational parameters, namely applied current density and treatment duration, on the quality of the treated TDW, and the influence of this quality level in terms of organic load content on the consecutive reuse in new dyeing baths are also addressed. Furthermore, the ecotoxicity towards *Daphnia magna* of the different dyes utilized and of the dyeing auxiliaries was evaluated, aiming to establish which of the compounds contributed most to the TDW ecotoxicity.

#### **2. Materials and Methods**

#### *2.1. Textile Dyeing Wastewater*

The dyeing process is the most intensive water and chemical consumer in the textile industry, generating the largest effluent stream with minimally known composition [25]. For these reasons, it has been highlighted as a good candidate for reusing the wastewater after treatment and, considering the high stability under sunlight and resistance to microbial attack and temperature of the synthetic organic dyes, the textile dyeing wastewater is also an optimum matrix to evaluate the performance of the electrochemical oxidation process [3,11,25].

The TDW utilized in this study was obtained from the dyeing process of a 100% wool twill fabric (weight = 351.6 g m−<sup>2</sup> ; finesse of yarn: warp and weft = 100 Tex) with a trichromatic combination of Nylosan acid dyes (Nylosan® Red N-2RBL (C.I. AR 336), Nylosan® Yellow N-3RL (C.I. AO 67), and Nylosan® Blue N-GL (C.I. AB 230)) (Figure 1).

The dyeing process was carried out in a Mathis Labomat type BFA-12 equipment, purchased from Maquicontrolo (Oporto, Portugal), under a fabric:dyeing bath ratio of 1:50 (5 g of wool fabric per 250 mL of dyeing bath). The dyeing bath consisted of an aqueous solution with a composition as described in Table 1 and a pH of 4.5. The Nylosan acid dyes, the Sarabid PAW (equalizer agent), and the SERA WET C-NR (humectant agent) were purchased, respectively, from Clariant (Leça do Balio, Portugal), CHT (Tübingen, Germany) and Dystar (Oporto, Portugal). Sodium sulfate, sodium acetate, and acetic acid were purchased from Sigma-Aldrich (Lisbon, Portugal).

**Figure 1.** Structure of three Nylosan N acid dyes: (**a**) Red Nylosan N-2RBL; (**b**) Yellow Nylosan N-3RL and (**c**) Acid Blue 230, one of the Blue Nylosan N-GL constituents. **Figure 1.** Structure of three Nylosan N acid dyes: (**a**) Red Nylosan N-2RBL; (**b**) Yellow Nylosan N-3RL and (**c**) Acid Blue 230, one of the Blue Nylosan N-GL constituents.


The dyeing process was carried out in a Mathis Labomat type BFA-12 equipment, **Table 1.** Composition of the dyeing bath utilized in the dyeing process.

**Constituents Concentration**  Nylosan® Red N-2RBL 80 mg L−1 (0.4% of dye/fabric (*w*/*w*)) Nylosan® Yellow N-3RL 60 mg L−1 (0.3% of dye/fabric (*w*/*w*)) Nylosan® Blue N-GL 60 mg L−1 (0.3% of dye/fabric (*w*/*w*)) Equalizer agent (Sarabid PAW) 200 mg L−1 (1% of equalizer/fabric (*w*/*w*)) Humectant agent (SERA WET C-NR) 1 g L−<sup>1</sup> Sodium sulfate anhydrous, ≥99.0%, 1 g L−1 (5% of Na2SO4/fabric (*w*/*w*)) The dyeing program involved a heating step, from room temperature up to 98 ◦C, at a heating rate of 2 ◦C/min, followed by a 30 min period at this final temperature. After this period, the mini-reactors were maintained in the apparatus for cooling to room temperature. After the dyeing process, the wool dyed fabrics were washed at room temperature, to remove the unfixed dye. The TDW obtained from each mini-reactor was collected and combined in a single TDW sample, which was characterized and then utilized in the EO experiments.

#### Sodium acetate anhydrous, ≥99.0% Acetate buffer (2 g L−1) pH 4.5 Acetic acid glacial, ≥99.7% *2.2. Electrochemical Treatment*

The dyeing program involved a heating step, from room temperature up to 98 °C, at a heating rate of 2 °C/min, followed by a 30 min period at this final temperature. After this period, the mini-reactors were maintained in the apparatus for cooling to room temperature. After the dyeing process, the wool dyed fabrics were washed at room temperature, to remove the unfixed dye. The TDW obtained from each mini-reactor was collected and combined in a single TDW sample, which was characterized and then utilized in the EO The TDW electrochemical treatment was conducted in batch mode, with stirring (300 rpm), using an undivided cylindrical cell, containing 250 mL of TDW. A commercial Si/BDD anode, purchased from Neocoat (La Chaux-de-Fonds, Switzerland), and a stainlesssteel cathode, each one with an immersed area of 10 cm<sup>2</sup> , were utilized as electrodes. They were placed in parallel, with an inter-electrode gap of 1 cm, and were centered in the electrochemical cell. A GW, Lab DC, model GPS-3030D (0–30 V, 0–3 A), purchased from ILC (Lisbon, Portugal), was used as the power supply.

experiments. Before the electrochemical treatment, the TDW was filtered to remove any fiber residues.

*2.2. Electrochemical Treatment*  The TDW electrochemical treatment was conducted in batch mode, with stirring (300 rpm), using an undivided cylindrical cell, containing 250 mL of TDW. A commercial In a preliminary set of EO experiments, the influence of the applied current density (*j*) on the electrochemical treatment performance was evaluated. Assays were run at 30, 60, and 100 mA cm−<sup>2</sup> , for 10 h.

Si/BDD anode, purchased from Neocoat (La Chaux-de-Fonds, Switzerland), and a stainless-steel cathode, each one with an immersed area of 10 cm2, were utilized as electrodes. They were placed in parallel, with an inter-electrode gap of 1 cm, and were centered in the electrochemical cell. A GW, Lab DC, model GPS-3030D (0–30 V, 0–3 A), purchased The EO treatment, for TDW reuse purpose, was performed for 10 h, at 60 and 100 mA cm−<sup>2</sup> . A set of five EO assays were run at each *j* studied. The treated TDW obtained from the five assays performed at the same experimental conditions was combined in a single sample, which was characterized and then utilized in a new dyeing process.

#### from ILC (Lisbon, Portugal), was used as the power supply. *2.3. Reuse Experiments*

Before the electrochemical treatment, the TDW was filtered to remove any fiber residues. The strategy adopted in this study is summarized in Figure 2. The reuse experiments comprised the dyeing process as described in Section 2.1, but utilizing, in the dyeing bath, the TDW samples treated by EO instead of fresh water. Furthermore, since it was found that the sulfate ion concentration was maintained during the EO treatment, the dyeing bath, in the reuse experiments, was prepared as described in Table 1, but without the addition of sodium sulfate. the TDW samples treated by EO instead of fresh water. Furthermore, since it was found that the sulfate ion concentration was maintained during the EO treatment, the dyeing bath, in the reuse experiments, was prepared as described in Table 1, but without the addition of sodium sulfate.

sample, which was characterized and then utilized in a new dyeing process.

In a preliminary set of EO experiments, the influence of the applied current density (*j*) on the electrochemical treatment performance was evaluated. Assays were run at 30,

The EO treatment, for TDW reuse purpose, was performed for 10 h, at 60 and 100 mA cm−2. A set of five EO assays were run at each *j* studied. The treated TDW obtained from the five assays performed at the same experimental conditions was combined in a single

The strategy adopted in this study is summarized in Figure 2. The reuse experiments comprised the dyeing process as described in Section 2.1, but utilizing, in the dyeing bath,

*Water* **2022**, *14*, x FOR PEER REVIEW 5 of 15

60, and 100 mA cm−2, for 10 h.

*2.3. Reuse Experiments* 

**Figure 2.** A schematic diagram of the processes. **Figure 2.** A schematic diagram of the processes.

The TDW generated during the first reuse cycle was collected and submitted to a second EO treatment and dyeing process. The second electrochemical treatment was performed under similar experimental conditions of the first treatment, and the second reuse cycle followed the procedure described above. The TDW generated during the first reuse cycle was collected and submitted to a second EO treatment and dyeing process. The second electrochemical treatment was performed under similar experimental conditions of the first treatment, and the second reuse cycle followed the procedure described above.

#### *2.4. Analytical Methods*

*2.4. Analytical Methods*  The TDW samples, before and after the EO treatments, were characterized in terms of chemical oxygen demand (COD), dissolved organic carbon (DOC), ecotoxicity towards the model organism *Daphnia magna*, sulfate ion concentration, pH, and electrical conductivity. COD determinations followed the closed reflux and titrimetric method, according The TDW samples, before and after the EO treatments, were characterized in terms of chemical oxygen demand (COD), dissolved organic carbon (DOC), ecotoxicity towards the model organism *Daphnia magna*, sulfate ion concentration, pH, and electrical conductivity. COD determinations followed the closed reflux and titrimetric method, according to the standard procedures [26].

to the standard procedures [26]. DOC was measured in a Shimadzu TOC-VCPH analyzer, purchased from Izasa Sci-DOC was measured in a Shimadzu TOC-VCPH analyzer, purchased from Izasa Scientific (Carnaxide, Portugal), with samples previously filtered through 0.45 µm membrane filters.

entific (Carnaxide, Portugal), with samples previously filtered through 0.45 μm membrane filters. The ecotoxicity towards *Daphnia magna* was evaluated using a commercial Daphtoxkit F microbiotests, DM230921, purchased from Ambifirst (Moita, Portugal), following the OECD/OCDE Guideline 202 [27], by measuring the number of immobilized *Daphnia magna* neonates exposed to different dilutions of the TDW samples. To clarify which con-The ecotoxicity towards *Daphnia magna* was evaluated using a commercial Daphtoxkit F microbiotests, DM230921, purchased from Ambifirst (Moita, Portugal), following the OECD/OCDE Guideline 202 [27], by measuring the number of immobilized *Daphnia magna* neonates exposed to different dilutions of the TDW samples. To clarify which constituents most contributed to the ecotoxicity of the TDW, the ecotoxicity of the dyes, the equalizer, and the humectant agents used was also assessed.

stituents most contributed to the ecotoxicity of the TDW, the ecotoxicity of the dyes, the equalizer, and the humectant agents used was also assessed. Sulfate ion concentration was determined by ion chromatography, using a Shimadzu Prominance LC-20A system with a Shimadzu CDD 10Avp conductivity detector, purchased from Izasa Scientific (Carnaxide, Portugal). An IC I-524A Shodex (4.6 mm ID × 100 Sulfate ion concentration was determined by ion chromatography, using a Shimadzu Prominance LC-20A system with a Shimadzu CDD 10Avp conductivity detector, purchased from Izasa Scientific (Carnaxide, Portugal). An IC I-524A Shodex (4.6 mm ID × 100 mm) anion column was used at 40 ◦C. The mobile phase was an aqueous solution of 2.5 mM of phthalic acid and 2.3 mM of tris(hydroxymethyl)aminomethane at pH 4, with a flow rate of 1.5 mL min−<sup>1</sup> .

pH was measured with a HANNA pH meter (HI 931400) and the electrical conductivity (EC) with a Mettler Toledo conductivity meter (SevenEasy S30K), both purchased from MT Brandão (Oporto, Portugal).

The presence of the dyes Nylosan® Yellow N-3RL, Nylosan® Red N-2RBL, and Nylosan® Blue N-G in the TDW samples was also monitored, through UV-vis spectrophotometric measurements at 436, 525, and 620 nm, respectively, utilizing a Shimatzu UV-1800 spectrophotometer, purchased from Izasa Scientific (Carnaxide, Portugal).

The performance of the dyeing process was evaluated in terms of total color difference (∆*E*\*) and color fastness to washing of the wool fabric samples dyed with the primary dyeing bath (control fabric), prepared with fresh water, and with the reused dyeing baths. ∆*E*\* was determined through the CIELab color system, following the procedure described

in ISO 105-J03:2009 [28] and utilizing a Spectraflash SF 300X reflectance spectrophotometer, purchased from Datacolor International (Trenton, NJ, USA).

The color fastness to washing was evaluated following the procedure described in ISO 105-C06 A2S:2010 [29]. These washing fastness tests were carried out with a multifiber fabric, which is divided equally by the acetate-cotton-polyamide-polyester-acrylic-wool fibers to which the dyed fabric to be tested was attached.

#### **3. Results and Discussion**

#### *3.1. First EO Treatment and Reuse Cycle*

The characterization of the TDW obtained from the primary dyeing process (utilizing fresh water) is presented in Table 2. This wastewater presents a high content of organic matter, as can be seen by the values of the COD and DOC concentrations, which can be ascribed to the presence of dyes that were not adsorbed by the fibers and to the dyeing auxiliaries utilized. In fact, the brownish color presented by the TDW and the absorbances at 436, 525, and 620 nm, which are the wavelengths of maximum absorbance of the dyes Nylosan® Yellow N-3RL, Nylosan® Red N-2RBL, and Nylosan® Blue N-GL, respectively, indicate the presence of residual dye content.

**Table 2.** Characterization of the TDW obtained from the primary dyeing process, before and after treatment by EO at 60 and 100 mA cm−<sup>2</sup> .


The ecotoxicity towards *Daphnia magna* was assessed and the obtained results confirmed the high toxicity presented by the textile dyeing wastewaters. According to the toxicity classification based on toxic units (TU) reported by Pablos et al. [30], where the toxicity expressed in terms of TU is calculated according to Equation (1), the TDW obtained from the primary dyeing process is classified as very toxic.

$$\text{TU} = \frac{100}{\text{EC}\_{50}(\%)} \tag{1}$$

To clarify which constituents most contributed to the ecotoxicity of the TDW, the ecotoxicity towards *Daphnia magna* of the dyes, and of the equalizer and humectant agents used, was assessed. The obtained results, presented in Table 3, show that the estimated EC50–48 h values for all three dyes and equalizer agent are above the values of the initial dyeing bath conditions (Table 1), indicating that these constituents do not contribute to the observed ecotoxicity. However, the humectant agent, with a calculated EC50–48 h of 75.18 mg L−<sup>1</sup> , is used in a concentration ten times higher than the estimated value.


**Table 3.** Ecotoxicity towards *Daphnia magna* of Nylosan® Yellow N-3RL, Nylosan® Red N-2RBL, Nylosan® Blue N-GL, Sarabid PAW and SERA WET C-NR. **Table 3.** Ecotoxicity towards *Daphnia magna* of Nylosan® Yellow N-3RL, Nylosan® Red N-2RBL, Nylosan® Blue N-GL, Sarabid PAW and SERA WET C-NR.

TU =

75.18 mg L−1, is used in a concentration ten times higher than the estimated value.

100

To clarify which constituents most contributed to the ecotoxicity of the TDW, the ecotoxicity towards *Daphnia magna* of the dyes, and of the equalizer and humectant agents used, was assessed. The obtained results, presented in Table 3, show that the estimated EC50–48 h values for all three dyes and equalizer agent are above the values of the initial dyeing bath conditions (Table 1), indicating that these constituents do not contribute to the observed ecotoxicity. However, the humectant agent, with a calculated EC50–48 h of

ECହ(%) (1)

*Water* **2022**, *14*, x FOR PEER REVIEW 7 of 15

To establish the best EO operational conditions for TDW reuse purpose, a preliminary set of EO experiments was performed, where the influence of applied current density and treatment duration on the color and organic load removal rates was assessed. Both color and organic load removals increased with *j* and treatment time, since, at higher currents and treatment times, the production of oxidative species is higher, promoting the enhanced degradation of the organic compounds [11]. At 100 mA cm−<sup>2</sup> , complete color removal (visual) was accomplished after 8 h assay, but at 60 mA cm−<sup>2</sup> it took 10 h to attain complete discoloration. At the lowest applied *j* (30 mA cm−<sup>2</sup> ), after 10 h assay, complete discoloration was not achieved. These observations were in accordance with the removal of the dyes, which was evaluated through absorbance measurements at 436, 525, and 620 nm. As it can be seen in Figure 3, the dye removal rate was higher in the first hours of assay, since more dye molecules were available to be oxidized. As the dye concentration became lower, its oxidation became diffusion controlled and the byproducts formed were preferably oxidized, decreasing the dye removal rate. To establish the best EO operational conditions for TDW reuse purpose, a preliminary set of EO experiments was performed, where the influence of applied current density and treatment duration on the color and organic load removal rates was assessed. Both color and organic load removals increased with *j* and treatment time, since, at higher currents and treatment times, the production of oxidative species is higher, promoting the enhanced degradation of the organic compounds [11]. At 100 mA cm−2, complete color removal (visual) was accomplished after 8 h assay, but at 60 mA cm−2 it took 10 h to attain complete discoloration. At the lowest applied *j* (30 mA cm−2), after 10 h assay, complete discoloration was not achieved. These observations were in accordance with the removal of the dyes, which was evaluated through absorbance measurements at 436, 525, and 620 nm. As it can be seen in Figure 3, the dye removal rate was higher in the first hours of assay, since more dye molecules were available to be oxidized. As the dye concentration became lower, its oxidation became diffusion controlled and the byproducts formed were preferably oxidized, decreasing the dye removal rate.

**Figure 3.** Decay in time of absorbance at (**a**) 436 nm; (**b**) 525 nm; and (**c**) 620 nm, during the preliminary EO experiments performed at different *j*. **Figure 3.** Decay in time of absorbance at (**a**) 436 nm; (**b**) 525 nm; and (**c**) 620 nm, during the preliminary EO experiments performed at different *j*.

Regarding COD and DOC (Figure 4), they presented similar decays, indicating a high degree of mineralization of the organic compounds for all the *j* studied. Nevertheless, Regarding COD and DOC (Figure 4), they presented similar decays, indicating a high degree of mineralization of the organic compounds for all the *j* studied. Nevertheless, when comparing COD and DOC removal rates with that of the absorbances at 436, 525, and 620 nm, during the first hours of the assay, COD and DOC removal rates are lower, indicating that the dye degradation occurred more rapidly than the overall organic load removal. This can be explained considering that, for the dye fragmentation (chromophore group breaking), only a primary oxidation stage is required, but, for COD and DOC reductions, more complex multistage oxidative reactions are involved, since the dye oxidation usually results in the formation of different byproducts [31,32].

tion usually results in the formation of different byproducts [31,32].

**Figure 4.** Variation in time of (**a**) COD; (**b**) DOC, and (**c**) pH, during the preliminary EO experiments performed at different *j*. **Figure 4.** Variation in time of (**a**) COD; (**b**) DOC, and (**c**) pH, during the preliminary EO experiments performed at different *j*.

when comparing COD and DOC removal rates with that of the absorbances at 436, 525, and 620 nm, during the first hours of the assay, COD and DOC removal rates are lower, indicating that the dye degradation occurred more rapidly than the overall organic load removal. This can be explained considering that, for the dye fragmentation (chromophore group breaking), only a primary oxidation stage is required, but, for COD and DOC reductions, more complex multistage oxidative reactions are involved, since the dye oxida-

Similar results were observed by Solano et al. [32] when studying the electrochemical treatment of a TDW sample, using a BDD anode, and in the presence of sulfate ions. According to these authors, electrolysis under BDD anode, in aqueous media containing sulfate ions, generates peroxodisulfate (Equation (2)), a powerful oxidizing agent that can oxidize organic matter near to the anode surface and in the bulk solution. Even for small sulfate concentrations, as in the case of the TDW under study (672 mg L−1), we found an enhancement in the EO performance, caused, according to the authors, by the production of peroxodisulfate that, together with the reactive oxygen species (ROS) such as hydroxyl radicals, oxidize the organic matter from TDW. In fact, more recent studies on electropersulfate processes show that, when utilizing anode materials with high oxygen evolution potential, such as BDD, not only can peroxodisulfate be produced from the oxidation of SO42<sup>−</sup> (Equation (2)), but also sulfate radicals can be electrogenerated, either by sulfate ions direct oxidation (Equation (3)) or by oxidation through a hydroxyl radical (Equation (4)) [33]. Moreover, peroxodisulfate can be also obtained through sulfate ions oxidation by hydroxyl radicals (Equation (5)) and sulfate radicals can be generated from the electrochemical activation of peroxodisulfate, according to Equation (6) [33]. According to the literature, the sulfate radical is a strong and highly oxidizing species that presents higher redox potential than that of hydroxyl radicals or peroxodisulfate. It readily reacts at a wide range of pH values, promoting the nonselective oxidation and efficient removal of a wide range of organic compounds [33]. Similar results were observed by Solano et al. [32] when studying the electrochemical treatment of a TDW sample, using a BDD anode, and in the presence of sulfate ions. According to these authors, electrolysis under BDD anode, in aqueous media containing sulfate ions, generates peroxodisulfate (Equation (2)), a powerful oxidizing agent that can oxidize organic matter near to the anode surface and in the bulk solution. Even for small sulfate concentrations, as in the case of the TDW under study (672 mg L−<sup>1</sup> ), we found an enhancement in the EO performance, caused, according to the authors, by the production of peroxodisulfate that, together with the reactive oxygen species (ROS) such as hydroxyl radicals, oxidize the organic matter from TDW. In fact, more recent studies on electro-persulfate processes show that, when utilizing anode materials with high oxygen evolution potential, such as BDD, not only can peroxodisulfate be produced from the oxidation of SO<sup>4</sup> <sup>2</sup><sup>−</sup> (Equation (2)), but also sulfate radicals can be electrogenerated, either by sulfate ions direct oxidation (Equation (3)) or by oxidation through a hydroxyl radical (Equation (4)) [33]. Moreover, peroxodisulfate can be also obtained through sulfate ions oxidation by hydroxyl radicals (Equation (5)) and sulfate radicals can be generated from the electrochemical activation of peroxodisulfate, according to Equation (6) [33]. According to the literature, the sulfate radical is a strong and highly oxidizing species that presents higher redox potential than that of hydroxyl radicals or peroxodisulfate. It readily reacts at a wide range of pH values, promoting the nonselective oxidation and efficient removal of a wide range of organic compounds [33].

$$\rm 2SO\_4^{2-} \rightarrow \rm S\_2O\_8^{2-} + 2e^- \tag{2}$$

$$\rm{SO}\_4^{2-} \rightarrow \rm{SO}\_4^{\bullet-} + e^- \tag{3}$$

$$\rm{SO}\_4^{2-} + \rm{OH}^{\bullet} \rightarrow \rm{SO}\_4^{\bullet-} + \rm{OH}^- \tag{4}$$

$$2\text{SO}\_4^{2-} + 2\text{OH}^\bullet \rightarrow \text{S}\_2\text{O}\_8^{2-} + 2\text{OH}^-\tag{5}$$

$$\rm{^1S\_2O\_8^{2-}} + e^- \rightarrow \rm{SO\_4^{\bullet-}} + \rm{SO\_4^{2-}} \tag{6}$$

Considering the existence of sulfate ions in the TDW sample under study and that a BDD anode is employed, it can be assumed that the oxidation of the organic compounds occurred in parallel by ROS and reactive sulfate species. It should be noted that, according to Solano et al. [32], the degradation process in the presence of sulfate occurs with significant formation of intermediates, but without formation of organochlorinated or other carcinogenic compounds, as in the case when oxidation is mediated by chloride species.

After 10 h of assay, the COD of the TDW samples was 468, 159, and 46 mg L−<sup>1</sup> , for the EO experiments run at 30, 60, and 100 mA cm−<sup>2</sup> , respectively. According to the literature, a

high-quality water for reuse purposes in the textile industry should present a maximum COD of 50 mg L−<sup>1</sup> , with the maximum COD of 200 mg L−<sup>1</sup> being a moderate-quality water [4]. Considering this criterion and the results obtained for color and organic load removal at the different *j* studied, the TDW treatment, for reuse purpose in new dyeing processes, was conducted at 60 and 100 mA cm−<sup>2</sup> , for 10 h, aiming to evaluate the influence of the treated TDW quality (moderate or high) on the performance of the dyeing process.

The characterization of the TDW samples treated by EO at 60 and 100 mA cm−<sup>2</sup> , for 10 h, which were utilized in the first reuse cycle, is presented in Table 2. Both treated samples were color and dye free. Although the COD of the TDW samples treated at 60 and 100 mA cm−<sup>2</sup> complied with the moderate-quality and high-quality requirements, respectively, the pH and the electrical conductivity values were higher than those required for wastewater reuse purposes in the textile industry [4]. The pH variation in time for the different *j* studied, presented in Figure 4c, shows an increase in pH during the EO treatment, which is more pronounced at higher *j*. This increase in pH during the EO process is well described in the literature and can be attributed to: (i) side reactions that occur at the cathode, such as the reaction described in Equation (7); (ii) sulfate oxidation through hydroxyl radicals (Equations (4) and (5)); and (iii) formation of carbonates and bicarbonates, from the reaction between the hydroxide ions and the CO<sup>2</sup> generated during the oxidation of the organic pollutants [33,34]. Such reactions are enhanced by the increase in *j*, explaining the higher pH values attained by the samples treated at higher *j*.

$$2\text{H}\_2\text{O} + 2\text{e}^- \rightarrow \text{H}\_2 + 2\text{OH}^- \tag{7}$$

According to the requirements for reuse purposes in textile industry [4], pH should not be higher than 8.0. Nevertheless, in the case of the wool fabric dyeing with Nylosan acid dyes, the dyeing bath is required to be at pH 4.5, involving an acidification process through the addition of an acetate buffer. Thus, although the pH of the treated TDW samples, at 60 and 100 mA cm−<sup>2</sup> , is higher than 8.0, its mandatory correction during the dyeing bath preparation eliminates the possible constraint associated with this high pH. It should be noted that, for most dyes used in the textile industry, pH correction is required.

Regarding the high electrical conductivity presented by the treated TDW samples (>4 mS cm−<sup>1</sup> ), this is mainly due to the sodium and sulfate ions, whose concentration was practically unchanged during the dyeing process and the subsequent EO treatment. In fact, although we assumed there was the generation of reactive sulfate species, from the oxidation of sulfate ions present in the TDW, sulfate ion concentration at the end of the EO treatments was practically unchanged. This is explained by the reaction of reactive sulfate species with dissolved organic matter (DOM) (Equation (8)) [35], and by reactions, such as that described in Equation (9), that occur when there is no DOM available for oxidation [33], which restore the sulfate ions in TDW. Thus, despite not complying with the general requirements for reuse purposes in the textile industry (<1.5 mS cm−<sup>1</sup> ) [4], EO treatment has the advantage of enabling the reuse of the treated TDW without the addition of more sodium sulfate salt.

$$\text{DOM} + \text{SO}\_4^{\bullet-} \rightarrow \text{DOM}\_{\text{ox}}^{+} + \text{SO}\_4^{2-} \tag{8}$$

$$\rm SO\_4^{\bullet-} + SO\_4^{\bullet-} \to 2SO\_4^{2-} \tag{9}$$

Despite not being a requirement for wastewater reuse purposes in textile industry, ecotoxicity towards *Daphnia magna* was evaluated. It was found that the ecotoxicity was drastically reduced during the EO treatment, indicating that the oxidation products are less toxic than the parent compounds. This fact is important for TDW reuse, since it shows that there is no significant increase in the dyeing bath ecotoxicity when prepared with treated TDW. Furthermore, we found that the reduction in ecotoxicity was slightly more pronounced in the TDW treated at 100 mA cm−<sup>2</sup> , which is in accordance with the higher oxidation degree of the degradation products at higher *j*.

The first reuse cycle was performed following the procedure described for the initial dyeing process, but utilizing the treated TDW instead of fresh water, and without adding sodium sulfate. The dyeing process was evaluated in terms of total color difference and color fastness to washing of the wool fabric samples dyed with the primary dyeing bath, prepared with fresh water, and with the reused dyeing baths, with the obtained results presented in Tables 4 and 5.

**Table 4.** Total color difference (∆*E*\*) and differences between *L*\*, *a*\*, and *b*\* obtained from the first reuse cycle.


**Table 5.** Color fastness to washing of the wool fabrics dyed using fresh water (primary dyeing) and TDW treated by EO (scale: 1 (poor) to 5 (excellent)).


∆*E*\* was calculated through Equation (10), which considers the differences between *L*\*, *a*\*, and *b*\* values of the fabrics dyed with treated TDW and that of the fabrics dyed with fresh water. The lightness/darkness of the color is given by parameter *L*\*: ∆*L*\* negative values indicate darker color and ∆*L*\* positive values indicate lighter color. Parameter *a*\* gives the color position between red and green: ∆*a*\* negative values indicate greener color and ∆*a*\* positive values indicate redder color. Parameter *b*\* represents the color position between yellow and blue: ∆*b*\* negative values indicate bluer color and ∆*b*\* positive values indicate yellower color [4]. According to the norm DIN ISO 11664–4:2012–06 [36], only for a ∆*E*\* above 1.5 are differences in the color between the sample and control fabrics visible. Nevertheless, there are restrictive controls that require a maximum ∆*E*\* of 1.0 (the smallest value the human eye can detect) [31].

$$
\Delta E^\* = \sqrt{\left(\Delta L^\*\right)^2 + \left(\Delta a^\*\right)^2 + \left(\Delta b^\*\right)^2} \tag{10}
$$

According to the results presented in Table 4, the fabrics obtained from the dyeing process that utilized the treated TDW were slightly lighter, greener, and yellower than the fabrics dyed with fresh water, with the lighter and yellower color more pronounced in the fabrics resultant from the dyeing with the treated TDW at 60 mA cm−<sup>2</sup> and the greener color more pronounced in the fabrics that utilized the treated TDW at 100 mA cm−<sup>2</sup> . These variations in the color parameters presented by the fabrics dyed with the treated TDW are, according to the literature, mainly due to the accumulation of byproducts from the degradation of the dyes and auxiliary products, which influence the dyeing process, in either exhaustion or fixation stages [22,31]. In a study performed by López-Grimau et al. [22], positive ∆*L*\* values were also obtained when reusing electrochemically treated TDW. According to the authors, this is due to a lower dye exhaustion, probably caused by an affinity between the residual hydrolyzed reactive dyes and the fibers or by its reaction with the new dyestuff, decreasing the final depth of shade. The differences found between the fabrics dyed with TDW samples treated at 60 and 100 mA cm−<sup>2</sup> are probably related to the different byproducts that resulted from the treatments, which are expected to have a higher oxidation degree in the EO treatment performed at 100 mA cm−<sup>2</sup> . Nevertheless, both dyeing processes, utilizing treated TDW at 60 and 100 mA cm−<sup>2</sup> , attained ∆*E*\* values of 1.0, complying with the most restrictive controls that require a maximum ∆*E*\* of 1.0.

The color fastness to washing was evaluated to verify differences between the primary dyed fabric and the fabrics dyed with the different treated TDW. The color transference was evaluated on a scale between 1 and 5, where 5 corresponds to no color transference. Color fastness to washing results, presented in Table 4, show that the wool fabrics dyed with the treated TDW presented similar behavior to that observed by the fabrics of the primary dyeing, validating the utilization of the treated wastewaters in new dyeing processes, regarding this parameter.

Both ∆*E*\* and color fastness to washing results showed the suitability of the treated TDW for reuse in new dyeing processes, with no significant differences found between the high-quality (treated by EO at 100 mA cm−<sup>2</sup> ) and moderate-quality (treated by EO at 60 mA cm−<sup>2</sup> ) samples. A similar conclusion was attained in a study performed by Silva et al. [4], where it was found that it is not strictly necessary to meet the reuse requirements to effectively reuse textile wastewater in the textile industry.

#### *3.2. Second EO Treatment and Reuse Cycle*

To evaluate the potential consecutive reuse of the treated TDW, the TDW samples obtained from the first reuse cycle were submitted to a second EO treatment, at the same experimental conditions utilized in the first EO treatment, and were then utilized in a second reuse cycle. Table 6 presents the characterization of the TDW samples obtained from the first reuse cycle, before and after the second EO treatment.

**Table 6.** Characterization of the TDW samples obtained from the first reuse cycle, before and after the second EO treatment.


Similar to that described in previous studies [14,18,25], TDW samples obtained from the first reuse cycle presented higher dye contents (given by the absorbances at 436, 525, and 620 nm) in comparison to those of the primary dyeing wastewater. Furthermore, COD and DOC values and the ecotoxicity towards *Daphnia magna* were higher for the TDW samples obtained from the first reuse cycle. According to the literature, this is ascribed to the accumulation of residual organic load and ecotoxicity [18]. Nevertheless, the second EO treatment was effective in the reduction of the organic load and ecotoxicity, with even better results than those obtained with the first EO treatment.

During the second EO treatment, color and dyes were completely removed. COD and DOC absolute removal and ecotoxicity reduction were higher than that attained for the first EO treatment, which can be ascribed to the higher organic load content of the TDW obtained from the first reuse cycle, since, for higher organics concentration, the mass transfer limitations in the electrochemical cell are reduced and thus the EO process is more effective [37]. As observed during the first EO treatment, sulfate ion concentration was practically unchanged during the second EO treatment, enabling the use of the treated

TDW in the second reuse cycle without the addition of sodium sulfate salt being necessary. This unchanged sulfate ion concentration after EO treatment is in agreement with the inability of advanced oxidation processes to remove salinity, which is usually highlighted as a constraint of these processes. Nevertheless, when considering the use of treated TDW in new dyeing baths, it becomes an advantage. According to Bezerra et al. [18], the TDW reuse without requiring salts addition can result in a saving of 10 tons of salt per year, besides reducing the environmental impact, since the disposal and treatment of saline textile wastewaters is one of the major environmental concerns of the textile industry. In the study performed by these authors, where a TDW was decolorized with H2O<sup>2</sup> catalyzed by UV light, the reuse of the treated TDW without requiring salts addition was not accomplished, with a salt adjustment being necessary.

The second reuse cycle and its performance evaluation followed the procedure applied to the first reuse cycle, utilizing the two TDW samples obtained from the second EO treatment. Similar performance was observed for the first and second reuse cycle. Regarding the total color differences (Table 7), the dyed fabrics obtained from the second reuse cycle presented lower ∆*E*\* values than those of the dyed fabrics from the first reuse cycle.

**Table 7.** Total color difference (∆*E*\*) and differences between *L*\*, *a*\*, and *b*\* obtained from the second reuse cycle.


The color fastness to washing assays (Table 8) showed no significant differences between the fabrics of the primary dyeing and the fabrics dyed with the treated wastewaters from the second EO treatment.

**Table 8.** Color fastness to washing of the wool fabrics dyed using fresh water (primary dyeing) and TDW after the second EO treatment.


#### **4. Practical Implications of the Study**

The United Nations 2030 Sustainable Development Agenda is "a universal call to action to end poverty, protect the planet and improve the lives and prospects of everyone, everywhere" [38]. The work developed in this study is fully aligned with this agenda, namely with Goal 6—Ensure availability and sustainable management of water and sanitation for all, and Goal 12—Ensure sustainable consumption and production patterns, since it presents a viable treatment solution for textile dyeing recalcitrant wastewater and allows its reuse in new dyeing processes. Many other studies have also presented viable treatment solutions to obtain wastewaters with good quality to be reutilized. However, this study went further, since it proved that salts can be completely recovered and reutilized in new textile dyeing processes, without losing dye uptake by the wool fibers. In addition, a good quality wool dyeing was attained, regardless of the use of a high- or moderate-quality water for preparing the dyeing baths. In fact, this was one of the main objectives in this study, together with the ecotoxicological evaluation towards *Daphnia magna* and the potential consecutive reuse of the treated wastewater. Of course, this is "a drop in the ocean", and many more studies should be performed to validate the reuse of treated wastewaters with

different quality levels and without salts addition/readjustment to the dyeing bath, namely from electrochemical oxidation with other anode materials. This is mainly a challenge for the dyeing process of other type of fibers, such as cotton or synthetics. Furthermore, in textile factories where the dyeing wastewater is not separated from the other textile processes, the reuse of the wastewaters with salts recovery, without membrane processes, is an even bigger challenge. If even today the engine that moves the world is based on purely economic reasons, it is necessary to establish a new paradigm that, although always considering economic constraints, promotes the sustainable reuse of all natural resources.

#### **5. Conclusions**

Electrochemical oxidation treatment of textile dyeing wastewaters, utilizing a BDD anode, is an effective strategy to obtain moderate-quality (COD <sup>≤</sup> 200 mg L−<sup>1</sup> ) or a highquality (COD <sup>≤</sup> 50 mg L−<sup>1</sup> ) water for reuse purposes in the textile industry. This strategy not only provides a reduction in water consumption, but also saves salts, since it allows the complete recovery of the salts utilized in the dyeing process, eliminating the need of salts addition in the subsequential dyeing baths. Furthermore, it promotes a drastic reduction of the wastewater ecotoxicity towards *Daphnia magna*, allowing wastewater reuse without severe ecotoxicity accumulation.

By varying the applied current density and treatment time, it is possible to obtain treated wastewaters with different qualities. High applied current densities result in higher color and organic load removal rates and in more oxidized byproducts. However, when comparing the performance of the dyeing processes that utilized treated wastewaters at 60 and 100 mA cm−<sup>2</sup> , which complied with the moderate- and high-quality requirements, respectively, no significant differences were found. Both treated samples lead to dyed fabrics with ∆*E\** values of 1.0 and with color fastness to washing results similar to that obtained by the fabrics of the primary dyeing (utilizing fresh water), complying with the most restrictive controls of the textile industry. Thus, it is concluded that, to efficiently reuse textile dyeing wastewater, a high-quality water, with a COD lower than 50 mg L−<sup>1</sup> , is not required.

Electrochemical oxidation process showed also to be feasible for the consecutive reuse of the treated wastewater. The increase in the organic load of the dyeing wastewater, caused by the accumulation of residual organic load from the previous treatment, is not a constraint for the subsequential treatment and reuse. In fact, higher organic load removals were attained during the second treatment and lower ∆*E*\* values were observed in the fabrics from the second reuse cycle.

**Author Contributions:** Conceptualization, A.L. and M.J.P.; Data curation, A.F. and M.J.P.; Formal analysis, A.F.; Investigation, C.P., M.J.N. and A.B.; Methodology, A.F. and M.J.P.; Project administration, A.L.; Resources, M.J.P.; Supervision, A.F. and M.J.P.; Validation, A.L. and M.J.P.; Visualization, A.L. and L.C.; Writing—original draft, C.P. and M.J.N.; Writing—review & editing, A.F., A.L., L.C. and M.J.P. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by FUNDAÇÃO PARA A CIÊNCIA E A TECNOLOGIA, FCT, project UIDB/00195/2020, PhD grants SFRH/BD/132436/2017 and COVID/BD/151965/2021 awarded to M.J. Nunes, and contract awarded to A. Fernandes, and by INSTITUTO NACIONAL DE GESTÃO DE BOLSAS DE ESTUDO, INAGBE, Ph.D. grant awarded to C. Pinto.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Data sharing is not applicable to this article.

**Acknowledgments:** The authors are very grateful for the support given by research unit Fiber Materials and Environmental Technologies (FibEnTech-UBI), on the extent of the project reference UIDB/00195/2020, funded by the Fundação para a Ciência e a Tecnologia, IP/MCTES through national funds (PIDDAC).

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

#### **References**

