Utilization of Moringa oleifera seeds as a biosorbent for diclofenac removal in the contaminated aquatic systems

The presence of emerging contaminants in the aquatic environment has been subject of increasing concerning due to their effects on the environment and on the human health even at low concentrations ( ≤ μg L⁻¹). In this category, pharmaceuticals are the contaminants most frequently detected due to its widespread use (Izadi et al. 2020). Diclofenac (DCF) is a nonsteroidal anti-inflammatory drug that is widely prescribed in human and veterinary medicine with analgesic and anti-pyretic properties (Naghipour et al. 2018; Xu et al. 2020). Although the DCF's ecotoxicity is relatively low, if it interacts with other drugs present in water, the toxicological potential increases considerably (Xu et al. 2020). Surface and ground water contaminated by DCF can cause extensive damage to biota, e.g., feminization of fish (Medina et al. 2021); therefore, this compound has been recently included in the List of Priority Contaminants, regulated by 2013/39/EU Directive (Medina et al. 2021).

Previous studies demonstrated that DCF is not removed or is partially eliminated in conventional water treatment plants (Xu et al. 2020; Medina et al. 2021). Therefore, it is necessary innovative and alternative technologies to removal DCF and other drugs in contaminated water. Advanced oxidation process (AOP) is innovative alternative to removal organic contaminants in water (Alkhuraiji 2019). However, the efficiency of these processes is considerably improved by the addition of chemical species, and additional techniques such as ultrasound radiation (Zhuan & Wang 2020). In addition, such procedures should be aware to the formation of degradation by-products with unknown toxicity. Also, considerable attention has been given to use of activated carbon (AC) for the adsorption of pharmaceuticals from aqueous solutions (Naghipour et al. 2018). Despite the efficiency of these procedures, they involve the use of chemical reagents, which may make them economically unfeasible. A promising alternative for the treatment of effluents that contain pharmaceuticals are natural adsorbents as a substitute for AC. Studies involving the use of biomass for the adsorption of pharmaceuticals are still scarce in the literature.

Moringa oleifera is a fast-growing tree that can reach 12 meters high found mainly in semiarid, tropical, and subtropical areas. Several applications have been proposed for M. oleifera, as its use in malnutrition and animal feed (Gopalakrishnan et al. 2016). The seeds have been used for the treatment of turbid water due to their flocculation properties (Baptista et al. 2017) and applied to textile effluent treatment (Agarwal et al. 2019). Other studies describe their use for the treatment of aqueous solutions containing silver (Araújo et al. 2010a), nickel (Marques et al. 2012) and chromium (Matouq et al. 2021) ions. However, there are still few reports available on the use of M. oleifera seeds, in their entirety, without the need for any structural change, for the adsorption of pharmaceuticals as DCF.

In this study, M. oleifera seeds were evaluated as a new low-cost adsorbent to remove DCF from the public potable water supply, river water and groundwater. The parameters that affect DCF adsorption (adsorbent dosage, pH, extraction time) were evaluated. Several models were carried out to determine the most suitable kinetic model and isotherm to explain the results. The use of M. oleifera seeds has certain advantages from both economic and environmental points of view, such as availability, the renewable nature of the adsorbent material, easy operation and their low cost.

Determination of diclofenac was performed using high-performance liquid chromatography – HPLC (Shimadzu, Kyoto, Japan) equipped with C18 column (150 × 4.6 mm, 5 μm diameter) and UV/VIS SPD-10A detector set at 282 nm. Acetonitrile and 0.10% (v/v) acetic acid (1:1) mixture was used as mobile phase.

All solutions were prepared using ultrapure water (18 MΩ cm), analytical grade reagents and HPLC grade solvents. Diclofenac sodium (DCF) (molecular weight = 318.1 g mol⁻¹; chemical formula = C₁₄H₁₀C_l2NNaO₂; pKa = 4.2) was purchased from Sigma Aldrich (Rome, Italy). Working solutions (0.5–100.0 mg L⁻¹) were prepared from the dilution of the 1,000 mg L⁻¹ DCF. Acetonitrile used as mobile phase was purchased from Sigma Aldrich. Prior to use, all glassware was kept overnight in 10% (v/v) aqueous nitric acid solution, followed by ultrasonication for 1 hour and finally washed with deionized water. Samples and solvents were filtrated in cellulose acetate membranes (Chrom filter) with 0.45 μm pores.

The M. oleifera seeds used as adsorbent material were collected in the city of Uberlândia-Brazil (latitude 18 °55′07″ S and longitude 48 °16′38″ W). After collecting the material, the pods were shelled and the seeds were washed twice with deionized water and left in an ultrasound bath for 10 minutes. Then, the seeds were dried in the open air (24–28 °C) for 8 hours in a dust-free environment. The dried seeds were ground in a commercial blender. Seed powders were separated according to particle size, using a sieve shaker with defined granulometry meshes. A particle size range of 0.5–1.0 mm was selected for further experiments. The material characterization was presented in previous works (Araújo et al. 2010b).

The experiments to investigate the adsorption kinetics were carried out in the temperature range of 25–28 °C using 0.5 g of M. oleifera seeds powder, 25 mL of 10.0 mg L⁻¹ DCF solution at pH 5.0, and stirring time up to 30 min. These same conditions were applied to adsorption isotherm experiments, except that the concentration of DCF varied from 0.5 to 100.0 mg L⁻¹. In both experiments, these conditions were applied to ensure the achievement of the equilibrium state; mixtures were then filtered and DCF quantified by HPLC.

The method was applied using the biosorbent for to evaluated adsorption of diclofenac from potable water (public supply), river water, and groundwater. The sample of potable water treated by a local sanitation company was collected from a tap in a research laboratory in the city of Uberlândia (Brazil), and river samples were collected from three different locations along the Uberabinha River, which runs through an urban zone in the city of Uberlândia. In these samples, diclofenac was not detected. Thus, to assess the performance of the adsorbent material in these systems, the samples were spiked with 10.0 mg L⁻¹ DCF. The samples were filtered through filter paper and the pH was adjusted to 5.0 using 0.1 mol L⁻¹ HCl. The samples (25 mL) were then treated with 0.5 g of adsorbent and using a stirring time of 5 min. The DCF was quantified using HPLC before and after the removal experiments.

The adsorption efficiency of the biosorbent depends on parameters as pH, material amount and contact time. The initial concentration of DCF (10 mg L⁻¹) was selected in such a way that the lowest concentration expected after adsorption was within the limit of quantification of the equipment (0.37 mg L⁻¹). Regarding the particle size, the selected size range (0.5–1.0 mm) was for presenting the largest contact surface without the occurrence of agglomerations in the solution.

The pH of DCF solutions was set at 5.0 for all experiments based on characterization studies of the M. oleifera seeds presented by Araújo and collaborators (Araújo et al. 2010b). Its pH value of the point of zero charge is in the range of 6 to 7. Thus, below this pH range, the surface of the sorbent is positively charged and it is able to adsorb diclofenac as its anionic form that is predominant close to pH 5.0.

The study of adsorption kinetics provides important information regarding its mechanism and the limiting step that controls the adsorption rate (Ho & McKay 1999). The models used to assess the kinetics of the adsorption process that best fit the data are the pseudo-first-order models of Lagergren, the pseudo-second order of Ho & McKay (1999) and the intraparticle diffusion model (Weber & Morris 1963).

The parameter values obtained from the linear fit for the pseudo-first order and pseudo-second order models are summarized in Table 1.

As can be seen in Table 1, the coefficient of determination for the pseudo-second-order kinetic model is the one that best fit, indicating that the adsorption process fits the pseudo-second-order mechanism. The pseudo-second-order adsorption kinetics plot of the drug DCF on the moringa is shown in Figure 3(b). This model assumes that the determining step of the adsorption rate depends on the physicochemical interactions between the adsorbent and the available groups on the surface of the adsorbent in its entirety. In addition, the adsorption process on the surface of the adsorbent is proportional to the number of sites occupied (Ho & McKay 1999; Ahmad et al. 2005).

The intra-particle diffusion model was also tested to obtain more information about the mechanism involving the formation of a permeable film, which can diffuse into the pores of the adsorbent (Weber & Morris 1963).

The performance of moringa seeds as an alternative adsorbent has presented similar conditions for satisfactory removal of other drugs. In a similar work, pseudo-second-order kinetics was also shown to be adequate to describe the limiting step of paracetamol adsorption, as reported by Ogunmodede et al. (2021).

Figure 7(b) was plotted according to Equation (8) and the values of n, and K_F are 1.0785 and 0.2894 respectively. The determination coefficient (r²) obtained was 0.9943, thus showing that the experimental data provide a good fitted with the Freundlich model, as observed in Figure 4(b). The value of n was greater than 1, suggesting relatively strong adsorption of DCF onto the surface of moringa seeds.

The isotherm evaluation showed that the experimental data for adsorption of DCF onto moringa seeds provided a good fit to the two isotherm models. Despite that linear correlation (r²) value of the Langmuir model is slightly higher than that of the Freundlich model (Table 2), it is reasonable to assume that DCF adsorption behavior onto biosorbent may take place in monolayer on the surface of the adsorbent. However, based on satisfactory correlation coefficients, the data are also consistent with a model in which the multilayers adsorption takes place on the adsorbent materials. This result is somewhat expected, bearing in mind the complexity of the structure and composition of the biosorbent.

For low initial concentrations of adsorbate this model assumes the Freundlich form, while at high concentrations, the Equation (9) assumes a form similar to the Langmuir isotherm, in monolayers. In the Sips isotherm model, the constant K_s is associated with the adsorption energy and the parameter n which also represents the degree of heterogeneity. If n= 1 the system is homogeneous, equating to the Langmuir model, and for values of n increasingly smaller than 1, an increase in heterogeneity has been characterized. In the case of n > 1, it is indicative of the formation of more than one layer of adsorbate on the adsorbent (Foo & Hameed 2010). The results were plotted in the Figure 7(c).

The good agreement of the experimental data with the Langmuir model suggests that the mechanism of adsorption of DCF onto moringa seeds occurs in a way defined by specific interactions at individual active sites. The parameters calculated by the Langmuir model are reasonable to indicate the reach of the adsorption equilibrium by the saturation of the active sites. The influence of material heterogeneity is related to the variation of the spatial orientation of the organic adsorbate on the surface, which may cause π-π interactions between the DCF molecules (Viotti et al. 2019). This explains the value of parameter n obtained in the Sips isotherm, slightly greater than 1.0.

Organic molecules such as diclofenac are limited to diffusing inside micropores, due to the size of molecules being relatively larger than micropores. As solid surfaces are rarely homogeneous, in addition to chemical reactions between the adsorbent and the adsorbate, the effects of the transport phenomenon in the inner portion of the adsorbate must be considered (Ho & McKay 1999), which is consistent with the experimental results obtained in this work, taking into account that the Langmuir and Freundlich isotherms obtained very similar and satisfactory coefficient of determination values.

The method was applied using the biosorbent for evaluated adsorption of DCF in potable water (public supply) and river water. In these samples DCF was not detected, and thus, they were spiked at a concentration of 10.0 mg L⁻¹. The samples were subsequently treated with the adsorbent material and the diclofenac removal efficiency was estimated as shown in Table 3. The results achieved indicate the potentiality of the M. oleifera seeds as an effective tool for DCF removal from water samples.

DCF removal efficiencies were greater than 95% under the experimental employed conditions (25 mL of 10.0 mg L⁻¹ DCF, pH of 5.0, contact time 30 min and 2.0 mg of M. oleifera seeds). The results showed satisfactory reproducibility to removal low concentrations of DCF in the samples tested.

The adsorption efficiency of moringa seeds as a natural biosorbent, without any structural modification, as reported here, is comparable to other more expensive adsorption procedures. In a study involving the adsorption of DCF in fixed bed columns, using AC as a synthetic material, a maximum adsorptive capacity of 36.25 mg g⁻¹ (298 K) was observed (Franco et al. 2018). In a similar work, Viotti et al. (2019), achieved good adsorptive capacity (60.805 mg g⁻¹) using M. oleífera pods for DCF removal in distilled water. However, a pre-treatment step of the adsorbent with HCl 0.1 mol L⁻¹ was necessary. Satisfactory results could only be obtained by heating the system to 328 K. The authors concluded that under such conditions, the adsorption of DCF it is not spontaneous (Viotti et al. 2019).

Compared to other previously presented materials for DCF removal as hybrid inorganic materials (Thanhmingliana & Tiwari 2015) or amended iron micro particles (Ghauch et al. 2010), the biosorbent based on M. oleifera seeds has advantages in high availability, easy material preparation, good adsorption capacity, cheap cost and potential applicability to removal of other organic and inorganic contaminants.

The evaluation of M. oleifera seeds as a biosorbent for DCF removal from potable water supplies demonstrated the high potential of this material as a cheap and environmentally friendly alternative for pharmaceutical remediation in water. DCF removal efficiencies were greater than 95% from water samples under the optimum experimental conditions. The use of moringa seeds without any structural change in their natural composition represented an advantage of this procedure. The results indicated that the rate-determining step of adsorption is controlled by specific interactions of an electrostatic nature between the DCF and the surface of the adsorbent, through pseudo-second-order kinetics. The Langmuir model proved to be adequate for adjusting the adsorption isotherm. The maximum adsorptive capacity was estimated at 32.05 mg g⁻¹, which is consistent with the theoretical value. The satisfactory maximum adsorption capacity obtained and negligible cost of biosorbent means that moringa seeds can be considered as a reliable natural material for the removal of DCF from aqueous effluents.

The authors are grateful for financial support from the Brazilian governmental agencies Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), from the MG state governmental agency Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG) and GO state governmental agency Fundação de Amparo à Pesquisa do Estado de Goiás (FAPEG).

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

The authors declare there is no conflict.