Malachite green removal using ionic flocculation

The textile, paper and cellulose, cosmetic and food industries use more than 10,000 types of dyes, resulting in large amounts and variability of colored effluents (Dahri et al. 2014).

Malachite green (MG) is a toxic cationic dye, widely used to dye silk, cotton, wool, wood and leather. It is also used as an antibacterial agent and fungicide in fish farming, aquaculture and animal breeding due to its efficacy and low cost (Oyelude et al. 2018; Xie et al. 2020). However, many hostile effects such as carcinogenicity, mutagenesis and malformation have been related to different organisms, especially mammals (Srivastava et al. 2004). The use of MG in aquaculture companies is prohibited in several countries and its tolerance limit in water bodies was established at a concentration of 0.5–100 μg L⁻¹ (Zhang et al. 2017). Thus, specific monitoring of MG consumption and its removal from industrial wastewater should be considered, due to its extensive destructive effects on the environment and health of fish and mammals (Salamat et al. 2019).

The techniques used in wastewater treatment include: coagulation-flocculation (Albuquerque et al. 2013), advanced oxidation processes (Karthikeyan et al. 2011), nanofiltration (Gozálvez-Zafrilla et al. 2008), photocatalysis (Das & Dhar 2020) and adsorption (Mahanta et al. 2009). Adsorption has been used widely, relative to other techniques, for its simplicity, facility and high efficiency (Wong et al. 2003).

Several types of adsorbents have been tested for their capacity to remove dyes from effluents – e.g., silica gel (Andrzejewska et al. 2007), alumina (Adak et al. 2005), zeolites (Alver & Metin 2012) and modified chitosan (Sadiq et al. 2020). With respect to MG removal, the adsorbents tested include magnetic microspheres (Pan et al. 2019), microplastics (Lin et al. 2020), activated carbon (Qu et al. 2019) and carbon composites (Adebayo et al. 2020).

Surfactants were tested as separation agents in a number of processes, such as cloud point extraction (Melo et al. 2014), microemulsion (Beltrame et al. 2005) and micellar ultrafiltration (Ahmad et al. 2006). Another promising process is ionic flocculation, in which low-cost surfactants are used. Amphiphilic surfactants are easily obtained by the saponification reaction between plant and animals fats, and a strong base. In aqueous media, these surfactants react with calcium ions, forming surfactant flocs. Their polar-nonpolar characteristics make them useful for ionic flocculation because they can adsorb organic compounds (Melo et al. 2017). The use of ionic flocculation is proposed in this study to remove MG from a synthetic textile effluent.

The surfactant used was synthesized in the laboratory from industrialized sunflower oil and sodium hydroxide p.a. by saponification, with mass proportions of 55.55 and 44.45%, respectively. The sunflower oil (10 g) and NaOH (8 g) were diluted in 90 mL of ethyl alcohol and 40 mL of distilled water, respectively. The resulting solutions were mixed and then heated. An Allihn condenser was used for alcohol reflux during the 90-minute saponification process. After saponification, the alcohol and water were evaporated to collect the surfactant formed. The surfactant was dried at 90 °C for 120 minutes.

The MG-contaminated effluent was prepared at 120 mg L⁻¹. Next, the surfactant and CaCl₂ were weighed to obtain the desired surfactant concentration in 50 mL of the effluent. The CaCl₂ was weighed so that the calcium ion concentration in solution was half the surfactant concentration. This proportionality has little effect on process efficiency (Cavalcante et al. 2018).

The surfactant was dissolved totally in the effluent solution, which was agitated at 20 rpm, and the pH adjusted to 9. The CaCl₂ was always added after complete dissolution of the surfactant and pH adjustment. The CaCl₂ addition causes surfactant precipitation and floc formation. The resulting combined system was agitated for 5 minutes at 20 rpm to help the interaction between surfactant flocs and MG, promoting pollutant adsorption on the floc surface.

Table 1 summarizes the tests performed (adsorbent dosage, dye concentration, contact time, influence of pH and influence of electrolytes).

A concentration of 10 mg-MG L⁻¹ was selected as the standard dose for the adsorbate because it exceeds the MG tolerance limit in water bodies (Zhang et al. 2017). In the MG concentration variation study, the highest MG concentration was 120 mg L⁻¹, which is significantly higher than that of the dyes present in textile effluents (Sirianuntapiboon et al. 2007).

To analyze contact time, the surfactant was dissolved in the effluent, pH adjusted (to 9) and CaCl₂ added. In assessing the effect of pH, the pH was adjusted after adding the surfactant and before adding CaCl₂. The influence of electrolytes was investigated because textile effluents have high electrolyte concentrations, mainly sodium chloride (NaCl) and sodium sulphate (Na₂SO₄) (Hunger 2003; Verma et al. 2012).

Four theoretical adsorption isotherm models – Freundlich, Langmuir, Temkin and Dubinin-Radushkevich (DR) – were used to describe the equilibrium between MG and the surfactants’ adsorbent surface.

Treated waters were analyzed to find suitable disposal destinations. Two types were analyzed, with (A1) and without (A2) NaCl added. The parameters determined were: pH, electrical conductivity (EC), potassium (K⁺), sodium (Na⁺), calcium (Ca²⁺), magnesium (Mg²⁺), free chlorine (Cl), carbonate (⁠⁠), bicarbonate (⁠⁠), sodium adsorption ratio (SAR) and hardness.

Figure 1 shows the effect of surfactant dosage on MG removal.

Within the surfactant dosage range studied, MG removal efficiency was not altered significantly by variation in surfactant concentration – i.e., ionic flocculation is independent of surfactant dosage. The high removal efficiency can be explained by the large number of adsorption sites available and interaction between MG and the hydrophobic part of the surfactant.

The MG-adsorption capacity declined significantly with increasing surfactant dosage. The greatest adsorption capacity (19.09 mg g⁻¹) was for a surfactant concentration of 500 mg L⁻¹ and the lowest (4.75 mg g⁻¹) for 2,000 mg L⁻¹. This decline may be due to the ready availability of adsorption sites leading to competition for MG molecules and the accumulation of unsaturated adsorbent particles reducing the surface area (Ozacar & Sengil 2005; Ramesh et al. 2008). The increased surfactant concentration is also insufficient to promote MG removal, as the latter's concentration in solution is very low, even when a small amount of surfactant is used (500 mg L⁻¹). Eltaweil et al. (2020) used a mesoporous magnetic biochar composite to adsorb MG, and Salamat et al. (2019) used nanochitosan prepared from shrimp shells, obtaining similar results.

The process is infeasible with surfactant concentrations below 500 mg L⁻¹, as there is not enough surfactant to produce flocs large enough for separation by centrifugation. The optimal surfactant concentration was 1,400 mg L⁻¹, with approximately 96% removal efficiency.

Figure 2 presents the effect of initial MG concentration (10–120 mg L⁻¹) on surfactant adsorption capacity at 10 minutes contact time.

As the initial MG concentration was increased from 10 to 120 mg L⁻¹, MG removal efficiency declined from 96 to 55%. The higher removal efficiency at low dye concentrations may be attributed to the greater proportional availability of active adsorbent sites. However, with increased dye concentration, active surfactant sites become saturated, resulting in decreased efficiency (Verma et al. 2020).

With respect to adsorption capacity, it rose from 9.6 mg g⁻¹ in 10 mg L⁻¹ dye solution to 65 mg g⁻¹ in 120 mg L⁻¹ of solution – i.e., as the initial MG concentration increased. This increase can be attributed to the larger number of MG molecules in solution. Thus, capacity rises until equilibrium time and maximum adsorption capacity are reached.

Figure 3 shows the effect of contact time on MG removal efficiency, and the influence of equilibrium time on MG removal.

Figure 3 shows that process equilibrium is achieved in only 10 minutes at 10 mg-MG L⁻¹ concentration. At higher MG concentrations, equilibrium is reached in 120 minutes. For example, for 60 mg-MG L⁻¹ concentration, MG removal is 67% and adsorption capacity 40 mg g⁻¹ at 10 minutes; after 120 minutes, MG removal and adsorption capacity become constant (98% and 59 mg g⁻¹).

The MG adsorption capacity increases over time due to the availability of active adsorption sites. However, when equilibrium is reached, adsorbent saturation keeps adsorption capacity and removal efficiency constant. Khawaja et al. (2021) used copper and graphene oxide nanoparticles with cellulose for MG adsorption, obtaining similar results.

Adsorption kinetics were studied to investigate the effects of contact time and obtain results for adsorption kinetic model parameters. The best fit among the kinetic models was assessed using the coefficient of determination (R²) and SSE. Figure 4 and Table 3 present the adsorption behavior results for different MG concentrations in the different kinetic models.

Table 3 shows that the pseudo-second-order model gave the best adsorption kinetics fit for the different MG concentrations – i.e., it exhibited the highest R² and lowest SSE values.

The pseudo-first-order and Elovich models yielded calculated values close to their theoretical counterparts, as well as high R² (Elovich model) and very high SSE (both). Similar results have been reported by others, e.g., in the use of nanocomposites for MG adsorption (Rajabi et al. 2019), MG adsorption by copper oxide-loaded, activated carbon (Sharifpour et al. 2019), MG removal by polyurethane functionalized by salicylate (El-Bouraie 2015) and MG adsorption in a polyamide nanocomposite containing ceric oxide as the adsorbent (Ghanbary & Jafarnejad 2017).

To assess the adsorption mechanism, the intraparticle diffusion model was analyzed based on the adsorption kinetic results. Figure 5 and Table 4 depict the results of the adsorption mechanism for different MG concentrations.

Figure 5 shows that MG adsorption onto the surfactant follows a two-stage mechanism (K_d,1 > K_d,2). The first is related to external surface adsorption, that is, rapid diffusion of MG molecules on the adsorbent's outer surface until saturation. In the second stage, the effect of intraparticle diffusion decreases significantly at all concentrations, meaning that adsorption equilibrium was reached at this stage through the adhesion of MG molecules inside the surfactant flocs (attached to the hydrophobic tail of the surfactant) (Wu et al. 2005; Mohammed et al. 2015).

The estimated constants associated with boundary-layer thickness (C) increased in both stages, indicating increases in the boundary-layers’ effects on the process at each stage. Values of C exceeding 0 also suggest that another mechanism, such as film diffusion, was involved as an adsorption rate-limiting step in conjunction with pore diffusion in the adsorption mechanism (Weber 1984).

The plot of q_tvs t^0.5 did not cross the origin at any MG concentration. Intraparticle diffusion was, thus, not the only adsorption rate-limiting step and different adsorption mechanisms may be occurring simultaneously (Kumar et al. 2010). Li et al. (2021) obtained similar results using polyurethane plastic residues in MG dye adsorption.

Adsorption isotherms are important for understanding the adsorption mechanism by correlating equilibrium data with theoretical models, on which basis it is possible to describe how MG molecules interact with active surfactant sites (Demiral & Güngör 2016). Batch experiments were performed under standard conditions – surfactant 1,400 mg L⁻¹, MG concentration range 10 to ⁻120 mg L⁻¹, agitation 20 rpm for 120 minutes, and pH 9. Figure 6 and Table 5 present the MG adsorption behavior results for the different isotherm models.

Table 5 shows that the Langmuir model offers the best fit for the experimental data, with the lowest SSE value (177) and the highest R² (0.98). This indicates that MG molecules are adsorbed homogeneously at the active surfactant sites, and maximum adsorption capacity, q_max, of 116 mg g⁻¹ is expected due to active site saturation. The separation factor, R_L, (0.13) indicates that MG separation by the surfactant is a favorable process (0 < RL < 1).

The low R² values for the Freundlich, Temkin and D-R models (R² < 0.95) and the highest SSE indicate that they do not fit the experimental data.

Thus, the Langmuir isotherm model represents this adsorption process – i.e., adsorption occurs in a monolayer and bond energies at the surfactant's active sites are equivalent, meaning that each MG molecule occupies only one adsorption site.

The pH range was between 7 and 12.7. At pHs 7 and 8 an oil emulsion forms in water because the low pH converts the surfactant into its respective fatty acids, precluding floc formation due to the lack of carboxyl anions and making the process inviable (Melo et al. 2015). Figure 7 shows the MG adsorption behavior results for different pH values.

Without pH adjustment, the pH of samples containing only diluted surfactant is between 10 and 10.5, because of the reaction between fatty acids and NaOH. Figure 7 shows that varying the pH alone, without adding surfactant and/or calcium, may produce false results, as MG displays different colors for different pH intervals. The results were determined by UV-vis spectrophotometry, so the pH ranges in which MG maintains its natural color and in which it becomes colorless need study.

For pHs between 7 and 9, MG maintains its natural green color. With the addition of calcium alone, about 5% of the MG is removed by flocculation. Dye removal efficiency depends directly on the interaction between MG molecules and the surfactant's nonpolar tail. At pH values greater than or equal to 11, MG starts to become colorless, whether removed from the solution or not.

Thus, for analyses of MG removal efficiency by ionic flocculation, pH was initially corrected to pH 9 (within the range that does not produce false values). At pH 9, MG removal was 96%.

Figure 8 shows MG adsorption behavior for different NaCl concentrations, with fixed surfactant and dye concentrations.

Figure 8 shows that the presence of electrolytes lowers MG removal efficiency slightly. Removal efficiency was approximately 96% for samples without electrolyte addition, while in those containing 1 mol L⁻¹ NaCl it was slightly more than 92%.

The electrolytes inhibit floc formation, as precipitation tolerance – i.e., the minimum calcium concentration capable of causing precipitation – rises when electrolyte concentration increases, because dissolved anionic monomers decline in the surfactant when NaCl is added (Noïk et al. 1987; Stellner & Scamehorn 1989).

As noted previously, two types of water were analyzed: NaCl was added to A1 but not to A2 (Table 6).

The ideal values – data rows 3 and 6 in Table 6 – are taken from the University of California Committee of Consultants, of 1974. They are used to compare water analyses with irrigation needs (UCCC 1974). Table 6 shows that A1 water is not suitable for irrigation, since it is only within the ideal range with respect to ⁠.

A2 water is fairly good for irrigation because it is only outside the ideal ranges for pH, Cl and hardness.

Ionic flocculation was used to remove MG from solution in this study.

• The surfactant-calcium combination has an MG removal efficiency of approximately 96% at 25 °C, with pH 9.

• With increasing dye concentration, surfactant adsorption efficiency tends to decline due to the saturation of active sites.

• Adding electrolytes to the effluent lowers adsorption efficiency a little.

• Equilibrium time is 10 minutes for 10 mg-MG L⁻¹, while, for concentrations above that, it is reached after 120 minutes.

• The Langmuir isotherm model best fits this adsorption process (R² = 0.98), with a separation factor (R_L) of 0.22.

• The pseudo-second-order kinetic model best fits the process.

The study's results demonstrate that ionic flocculation can be used to remove dyes from wastewater.

Not applicable.

The authors confirm consent to publication and authorship of the study. Informed consent was obtained from all individual participants included in the manuscript.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

All relevant data are included in the paper or its Supplementary Information.