Coagulation Activity of the Seed Extract from Zygia cauliflora (WILLD.) KILLIP applied in Water Treatment

The use of natural coagulants is a promising alternative to replace or assist chemical coagulants due to its numerous advantages. This paper evaluates the effectiveness of a natural coagulant in water treatment. The coagulant was extracted from the Zygia cauliflora (Willd.) Killip seed in saline solutions and defatted using 95% ethanol. The tests were conducted using different concentrations of the coagulant (0.1, 0.5, 1.0, 2.0, 3.0, and 4.0 g L-1) and of NaCl (0.0 M, 1.0 M, and 5.0 M). The results showed that the use of 2 g L-1 of coagulant and 1.0 M of the salt achieved an effectiveness of 20% and 70%, respectively, of color and turbidity removal. The analysis showed an effectiveness of 70% and 30%, respectively, when removing the UV254 compound and DOM (dissolved organic matter). The mechanism for turbidity removal by the coagulants prepared with aqueous solution and with saline solution at 1.0 M of NaCl is supposed to be adsorption and charge neutralization, respectively, following the Freundlich and Langmuir models. However, the coagulant prepared with saline solution at 5.0 M of NaCl tends to form a netlike structure followed by turbidity removal through a sweep coagulation mechanism. Future papers should therefore focus on the use of Zygia cauliflora (Willd.) Killip as an alternative for replacement or use as an auxiliary chemical coagulant.


INTRODUCTION
According to the World Health Organization (WHO) and its member countries, "every person, at any stage of development and socio-economic conditions, has the right to access an adequate supply of potable and safe water". In this context, 'safe' refers to a water supply which does not represent a health risk, with enough quantity to meet all household needs, that is continuously available, and is affordable. If the purpose is to improve public health, such conditions must be considered when defining and maintaining water quality and water supply programs (OPS, 2001).
During the water treatment process, one of the first steps is coagulation, which aims to reduce impurities to make the water potable. Since it is a delicate part of the treatment, a strict control of the dosage of these chemical products is necessary. However, an increase of these dosages is required considering the high demand of potable water and a considerable decrease in the quality of water resources. The decision of coagulant is often based on economic factors, related to suitability, raw water, treatment technology, and cost (Richter et al., 1991).
According to Richter (2009), the most commonly used coagulants are iron and aluminumbased coagulants, due to their reaction with the natural alkalinity of raw water. According to several studies, there is evidence of a relationship between the content of aluminum in treated water and the increasing incidents of neurological diseases. According to Flaten (2001), aluminum plays an important role in the etiology and pathogenesis of Alzheimer's, but this connection is still under debate. Bondy (2010) states that there are analyses that link Parkinson's disease and other chronic neurodegenerative diseases as well.
Moreover, the high sensitivity of inorganic coagulants to the water pH and the possibility of secondary contamination of drinking water with traces of toxic synthetic polymeric coagulants or residual iron and aluminum ions are the main challenges for flocculationcoagulation water treatment processes (Bratskaya et al., 2004).
In view of these disadvantages, it is desirable to replace chemical coagulants with natural coagulants. Using natural plants as coagulants has many advantages, among which are their profitability, biodegradability, and lack of requirement for a high pH for water treatment (Bratskaya et al., 2004).
Thus, the Moringa oleifera Lam (moringa) seed has been gaining prominence as a natural coagulant in water treatment since the species contains an edible oil, a water-soluble substance with protein in its organic composition, and can act as an effective coagulant Salazer Gámez et al., 2015). However, Moringa oleifera is an Indian seed, so its availability could be restrictive for water treatment worldwide. It is therefore important to research the applicability of regional leguminous seeds that could play the same role as Moringa oleifera Lam.
Many plants have recently been identified as sources of natural polymers for the treatment of various types of water, including Plantago ovata, Musa ABB, and Strychnos potatorum (Ramavandi, 2014;Tirado et al., 2017;Gaikwad and Munavalli, 2019).
The use of regional coagulants is studied to evaluate their applicability regarding the difference in coagulation capacity, cheapness, variety, and ease of access. This research has attempted to use Zygia cauliflora (Willd.) Killip. seeds as an abundantly available, low-cost, and renewable precursor for producing a coagulant for improving water quality. Zygia cauliflora (Willd.) Killip. was considered because it is one of the most abundant species in undisturbed forests and with the highest density throughout the Amazon rainforest, especially in permanent preservation areas of the Moju River Basin (Oliveira et al., 2016). Zygia cauliflora (Willd.) Killip. trees, flowers and seeds are shown in Figure 1. Therefore, this paper evaluates the coagulation potential of aqueous and saline solutions obtained from Zygia cauliflora (Willd.) Killip. seeds to determine water color, turbidity, and organic compounds absorbed at 254 nm (UV254 compounds) and to remove dissolved organic matter (DOM). The effect of the operational parameters including coagulant dosage and NaCl concentration was studied to better understand the coagulation process using Zygia cauliflora (Willd.) Killip. seeds.

MATERIALS AND METHODS
Raw water samples were collected from a public water treatment system (WTS), Paraná Sanitation Company (Companhia de Saneamento do Paraná, SANEPAR), located in Campo Mourão, State of Paraná, Brazil. The surface water originates from the Campo River Basin.
Characterization before and after the treatment was carried out using quality parameters such as color, turbidity, UV254, and dissolved organic matter (DOM).

Color
The apparent color was determined based on the procedures described in Standard Methods for the Examination of Water and Wastewater 10 using the Digimed colorimeter equipment, Model DM-COR.

Turbidity
The turbidity was determined based on the procedures described in Standard Methods for the Examination of Water and Wastewater (APHA et al., 2012) using a turbidimeter (PoliControl equipment, Model AP2000 iR).

Dissolved organic matter parameters
The UV254 and UV272 compounds were determined using the HACH spectrophotometer, Model DR5000.
Equation 1 was used to determine the concentration of dissolved organic matter from UV272 results, a method proposed previously (Khan et al., 2014). (1) The detection of dissolved organic matter (DOM) in water is necessary since it consists of a set of substances originated from the excretion, secretion, and intermediary process of the decomposition of terrestrial and aquatic organisms (Di Bernardo et al., 2002), which influences water appearance among other consequences.

Extraction and preparation of the coagulant
The method used for the extraction and preparation of the coagulant in saline solution was described by Sánchez-Martín et al. (2010).
The peels of the Zygia cauliflora (Willd.) Killip seeds were manually removed and ground in a domestic blender. The powder was defatted in 95% ethanol with a mass/volume ratio of 1:10 using a magnetic stirrer for 45 minutes at 150 rpm.
The supernatant was manually separated and placed in glass tubes so that the resulting powder would dry in an oven at 60 º C for three days.
A coagulant was prepared in distilled water, which was denominated aqueous solution. One gram of the dried seed was added in a glass beaker containing 0.1 L. The coagulant was extracted by turbolysis for 3 minutes and then stirred in a magnetic stirrer for 30 minutes at 150 rpm. The obtained solution was vacuum filtrated on quality filter paper and then on a 0.9 µm pore glass fiber membrane. The dried seeds and turbolysis process are shown in Figure 2.
The same procedure was performed for the saline solutions of 1 M and 5 M. All tests were performed in duplicate.

Coagulation/flocculation test
The coagulation/flocculation tests were performed on a Jar test device (Nova Ética) with a rotating regulator of mixing rods and jars with capacity of 1 L (Figure 3). The use of the Jar Test allows the determination of the ideal dosage of coagulant or coagulation aid necessary to cause raw water clarification (Di Bernardo et al., 2002).
Predetermined amounts of coagulant solution were added to each jar (Table 1), using an automatic pipette. The water used was at room temperature, without any pH correction or adjustment. Jar 1 Jar 2 Jar 3 Jar 4 Jar 5 Jar 6 The rapid mixing was set at 1 minute, with gradient speed of 120 rpm, while the slow mixing was set at 15 minutes at 60 rpm. The settling time was established at 15 minutes. The methodology used for the tests was described in Di Bernardo et al. (2002).
Subsequently, the color, turbidity, UV254 compounds, and DOM were measured to verify the efficiency of the process by comparing the results with raw water.

Mechanism evaluation
In order to evaluate the mechanism of coagulation/flocculation process using aqueous and saline solutions obtained from Zygia cauliflora (Willd.) Killip, Langmuir and Freundlich isotherms were employed to describe adsorption equilibrium. This evaluation was performed using turbidity removal data and was adapted from Fahmi et al. (2011), Mateus et al. (2020, Zhou et al. (2018) and Sia et al. (2020).
The theoretically derived Langmuir model assumes adsorption energies are uniform and independent of surface coverage and complete coverage of surface by a monolayer of adsorbate indicates maximum adsorption. The linear form of Langmuir isotherm is given in Equation 2: Where: qe = ratio of turbidity adsorbed to the dosage of coagulant employed Ce = equilibrium concentration of turbidity qm = maximum adsorption capacity of the coagulant employed Kl= Langmuir constant, related to the energy of adsorption The value of Langmuir constants (slope) and maximum adsorption capacity (intercept) were obtained from the linear correlation plots between 1/qe with 1/Ce.
On the other hand, the Freundlich model is empirical, encompassing the heterogeneity of adsorbent surface, the exponential distribution sites and their energies. The linear equation of the Freundlich isotherm is given in Equation 3: Where: qe = ratio of turbidity adsorbed to the dosage of coagulant employed Ce = equilibrium concentration of turbidity Kf = Freundlich constant indicative of the relative adsorption capacity of the adsorbent 1/n = constant that indicates the adsorption intensity The value of Freundlich constants (slope) and adsorption intensity (intercept) were obtained from the linear correlation plots between ln qe with ln Ce.

Characterization of the raw water
The results of the raw water characterizations are presented in Table 2 according to color, turbidity, UV254 compounds, and DOM. According to Gaikwad et al. (2019), the composition of typical raw water samples collected from a river has a turbidity of 5-50 NTU. The turbidity of the raw water collected in this study was 76.3 NTU, which can be considered high.
On the other hand, the raw water characteristics are similar to reports found in literature. Ramavandi (2014) evaluated a coagulant extracted from Platago ovata and used a raw water with turbidity of 76 NTU, TOC (total organic carbon) of 8.6 mg L -1 and DOC of 4.02 mg L -1 . 8 Lima et al. (2017) used raw water with color of 174 UC and turbidity of 97 NTU when studying Abelmoschus esculentus as a flocculation aid. Malik (2018) used raw water with higher levels of turbidity, 250 NTU. the color, turbidity, and UV254 were approximately 20%, 70%, and 70%, respectively, with no considerable differences between the dosage employed. Despite this, the coagulant prepared with 5.0 M of NaCl removed no DOM at any dosage, while the coagulant prepared with 1.0 M of NaCl removed approximately 21% of DOM at all dosages.

Results of the tests with Zygia cauliflora (Willd.) Killip as a potential natural coagulant
On the other hand, the coagulant obtained from the aqueous solution removed no color, while removing turbidity, UV254, and DOM at approximately 45%, 70%, and 30%, respectively. Considering these results, the coagulants prepared with a saline solution at 1.0 M of NaCl were more efficient than those prepared with a 5.0 M solution. Furthermore, better results were obtained for the coagulant concentration of 2 g L -1 , even when compared to concentrations of 3 g L -1 and 4 g L -1 , which also showed reasonable results of color and turbidity removal. Thus, a more satisfactory efficiency was obtained for the saline solution with 1.0 M concentration of NaCl for the analyzed parameters, presenting 30.4% and 74.5% of removal, respectively, for color and turbidity.
Moreover, the saline solutions presented greater efficiency in removing color and turbidity when compared with the aqueous solution, as observed by Madrona et al. (2010), Nkurunziza et al. (2009), andNishi et al. (2011). Nkurunziza et al. (2009) reached 99.8% of turbidity removal using Moringa oleifera coagulant extracted with saline solution at 1.0 M, while Nishi et al. (2011) obtained 96.4% of color removal in the same conditions. These results suggest that Zygia cauliflora (Willd.) Killip could be applied as a coagulation/flocculation aid, instead of a primary coagulant.
The aqueous solution presented better UV254 compound and DOM removal, removing 74.3% of UV254 compounds at the concentration of 3.0 mg.L -1 and 31.7% of DOM at the concentration of 4.0 mg L -1 .
The presence of NaCl in the water showed that increasing the ionic strength results in a greater amount of extracted compounds, but does not affect coagulation activity, which can be observed in the results of UV254 and DOM removal (Kukic et al. 2015).
Previous work found in the literature shows that the coagulant obtained from Plantago ovata at a dosage of 0.250 mg L -1 showed turbidity removal of 36% and 99%, respectively, when using crude and treated extract (Ramavandi, 2014). Tirado et al. (2017) applied higher dosages (5-200 mg L -1 ) of starch obtained from Musa ABB as a coagulant and achieved color and turbidity removals from 90 to 99%. 9 The difference in results when compared to this work could be due to the higher dosages used by the authors. Valverde et al. (2018) evaluated the synergistic effect between Moringa oleifera and APC (aluminum polychloride) for coagulation/flocculation and observed lower removal efficiencies at greater proportions of moringa, with 70.1%, 75.1%, and 56.7% for color, turbidity, and UV254 absorption, respectively, which suggests that natural coagulants present lower removal efficiencies. Furthermore, Valverde et al. (2018) obtained a turbidity removal close to that observed in this work with the coagulant prepared with saline solutions, from 65 to 70%. Moreover, the removal of UV254 compounds achieved in this paper are higher, at 70%, for aqueous and saline solutions than the results observed by Valverde et al. (2018), even though the authors employed a synthetic coagulant in association with Moringa oleifera.

Mechanism evaluation results
Figures 5 and 6 show the respective results of Langmuir and Freundlich adsorption equilibrium models for turbidity removal, and Table 3 presents the R 2 obtained for each model and coagulant condition.
It was observed that the coagulants prepared with aqueous solution and with a saline solution at 1.0 M of NaCl show positive correlation with Langmuir and Freundlich models. The coagulant prepared in aqueous solution presented R 2 of 0.77 and 0.98, respectively, for the Langmuir and Freundlich models, while the coagulant prepared with a saline solution at 1.0 M of NaCl showed R 2 of 0.63 for and 0.54, respectively, for the Langmuir and Freundlich models. This indicates that adsorption and charge neutralization mechanisms are most likely involved in the processes (Fahmi et al., 2011).   Furthermore, the mechanism for turbidity removal for the coagulant obtained from Zygia cauliflora (Willd.) Killip. Extract in aqueous solution was found to be due to adsorption, approximating to the Freundlich type (>R 2 ), which indicates that adsorption occurs on heterogeneous surface and the adsorption energy decreases logarithmically, as the surface of adsorbent becomes covered by the solute (Fahmi et al., 2011;Mateus et al., 2020;Zhou et al., 2018;Sia et al., 2020).
With the experimental data presented, it is possible to observe that the Langmuir model presented a better fit (>R 2 ) for the coagulant with a saline solution at 1.0 M of NaCl, suggesting homogeneous and monolayer adsorption characterizing a more selective adsorption process, as reported by Mateus et al. (2020). The decay in R 2 values when compared with aqueous solution suggests that the mechanisms involved tend to change when saline solution is used to extract the coagulant from Zygia cauliflora (Willd.) Killip. seeds.
However, these models were not found to fit for the coagulant prepared with a saline solution at 5.0 M of NaCl, which presented R 2 for the Langmuir model of 0.03 and for the Freundlich model of 0.15. This occurrence indicates that the coagulant obtained from Zygia cauliflora (Willd.) Killip. at higher saline concentrations more probably form a netlike structure followed by turbidity removal through a sweep coagulation mechanism, as previously described (Fahmi et al., 2011).

CONCLUSIONS
This study showed that the concentration of NaCl slightly influenced the removal of water parameters such as color, turbidity, UV254 compounds, and DOM using the coagulants obtained from Zygia cauliflora (Willd.) Killip. The optimum dosage of coagulant for color and turbidity removal was 2.0 mg L -1 with 1.0 M NaCl, which is very low and removed 21% of DOM from treated water. The removal of color, turbidity, UV254 compounds, and DOM were 30.4%, 74.5%, 70.1%, and 21.5%, respectively, in this optimum dosage.
Furthermore, aqueous solution in concentrations of 3.0 and 4.0 mg L -1 removed UV254 compounds and DOM more efficiently, with values of 74.3% and 31.7%, respectively.
Moreover, the mechanism for turbidity removal by the coagulants prepared with aqueous solution and with a saline solution at 1.0 M of NaCl is supposed to be adsorption and charge neutralization, with the adsorption isotherm following the Freundlich model for aqueous solution and the Langmuir model for saline solution at 1.0 M of NaCl. Therefore, the coagulant prepared with a saline solution at 5.0 M of NaCl tends to form a netlike structure followed by turbidity removal through sweep coagulation mechanism.
According to the data presented, the Zygia cauliflora (Willd.) Killip seed can be used as a natural coagulant and is a good alternative to improve water quality for human supply.