Purification of aquaculture effluent using Picralima nitida seeds

Aquaculture effluent treatment is essential to eliminate the undesirable characteristics of water to ensure cleaner production and environmental sustainability. In an effort to develop green coagulant without compromising cost, this research investigated the feasibility of aquaculture effluent (AQEF) pollutant removal using Picralima nitida seeds extract (PNSC) and its bio-coagulation/adsorption kinetic characteristics with the substrate in water. The coagulative decrease was observed in terms of TD (turbidity), TSS (total suspended solids), COD (chemical oxygen demand), BOD (biochemical oxygen demand), and COLR (color) from AQEF. The active coagulant was extracted from the seeds and analyzed for its spectral and morphological characteristics through FTIR and SEM. The influence of PNSC dosage (0.10–0.50 g L−1), pH (2–10), settling time (0–60 min), and temperature (303–323 K) on the removal of contaminants were surveyed. The process kinetics of coagulation–flocculation were also explored. Maximal TD reduction of 90.35%, COD (82.11%), BOD (82.38%); TSS (88.84%), and COLR (65.77%) at 0.2 g PNSC L−1, pH 4, and 303 K was achieved. Analysis of variance (ANOVA) tests proved that pH, temperature, and settling time had a significant effect on pollutant removal. Results fitted Von Smoluchowski’s perikinetics theory at the optimum conditions, which gave R2 > 0.900. At perikinetics circumstances, the Kb (reaction rate) and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${t}_{f\frac{1}{2}}$$\end{document}tf12 (half-life) correspond to 0.0635 Lg−1 min−1 and 1.9 min. More so, sorption results fitted the Lagergren over the Ho model. Additionally, the net cost of using PNSC to handle 1 L of AQEF (including electricity, material, and labor costs) was evaluated to be €4.81. Overall, the PNSC appears reliable and useful in pretreating AQEF for improved biodegradability and superior effluent quality.

Preparation and characterization of the PNSC. The usage of plant extracts instead of the entire plant is advantageous because it tends to prevent the growth of pathogenic organics and other additional pollutants 36 . The active coagulant was isolated from the coagulant precursor, Akuamma (Picralima nitida) seeds, to maximize the efficiency of contaminants removal. The Akuamma (Picralima nitida) seeds consisted of 16.0% ash, 28.4% protein, 0.17 g mL −1 bulk density, 7.40% fat, 10.5% fiber, 31.1% carbohydrate, and 46.6% moisture,were obtained from Ihembosi environs, Anambra state, Nigeria. This study complies with Nnamdi Azikiwe University's insti-  www.nature.com/scientificreports/ tutional guidelines. The appropriate permission for the collection of plant specimens for experimentation was approved. The active coagulant was prepared according to the method described by Igwegbe et al. 46 . 30 g of the seeds was added to 250 mL of n-hexane at 70 °C for 6 h to extract the oil using an extractor. Prepared 250 ml salt solution of 4.0 g MgCl 2 , 25 g NaCl, 0.75 g KCL, and 1.0 g CaCL 2 in 1000 mL of distilled water were mixed and shaken with 10 g of the Picralima nitida seeds at 323 K for 60 min. The filtrate of the mixture was hardened at 27 ± 2 °C. The functional groups contained in the PNSC were identified to determine which chemical groups were present and participated in the coagulation experiment. PNSC (10 g) were freeze-dried for 24 h at 105 °C in a York Scientific Industries, India Lyophilizer, before their spectrum characteristics were investigated. The FTIR analysis was performed by combining the PNSC in a 1:100 ratio with dry finely powdered potassium bromide and collecting spectra from 4000 to 400 cm −1 at room temperature. The PNSC spectra were acquired using a Fourier transform infrared transmission system (Buck M520 Infrared spectrophotometer). SEM was performed via a Carl Zeis Analytical SEM Series. MA 10.EVO-10-09-49 to observe the surface morphology of the PNSC. The image was recorded for magnifications of 1000 × and 2000 × at a working distance of 15 mm and accelerating voltage of 15 kV using full BSD (backscattered electron dictator).
Coagulation-flocculation test procedure. The standard jar test procedure conducted for the coagulation-flocculation treatment of AQEF was carried out using a flocculator (model ZSI-2120). The pH was measured using a pH meter (Hanna pH meter). To ensure a uniform concentration of the effluent medium, AQEF was agitated before collecting the samples for the experimentations. The effects of PNSC dose (0.10-0.50 g L −1 ), pH (2 to 10), settling time (0-60 min) and temperature (303-323 K) on TSS, COD, TDS, COLR, and BOD changes were investigated. The TD, TSS, COD, and BOD were tested using EPA Method 180.1 51 , APHA 2540 D 50 , APHA 5220-D 50 , and APHA 5210-B 48 , respectively. The COLR was tested by measuring the absorbance at 275 nm, which is the maximum wavelength obtained for AQEF 46 .
In the course of the jar test experimentation, a 500 mL measuring cylinder was used to measure the AQEF sample, which was then poured into various 1000 mL beakers. The pH was attuned using 1 M HCL or NaOH solutions. A 0.10 g L −1 dose of PNSC was added to each beaker, and the mixture was then agitated for 5 min at a stirring speed of 120 rpm, supported by 20 min of shaking at a reduced speed of 30 rpm at a temperature of 30 °C. The stirring was stopped to allow for studying the floc formed at various settling times (3-30) minutes. A syringe was used to extract 20 mL of the samples at 0.02 m depth from each of the 1000 mL beakers and tested for the coagulation efficiencies (%CGE) in terms of TD, TSS, BOD, COD, and COLR. The same procedure was repeated using other dosages of PNSC. In each case, the corresponding efficiencies (%CGE) were also evaluated following Eq. (1).
CGE i and CGE f are the initial and final concentrations of TSS, TDS, BOD, COD, and COLR respectively.
A one-way analysis of variance (ANOVA) test was performed to understand the significance and the effect of the changes in process parameters (dosage, pH, temperature, and settling time with removal parameters (TSS, TD, COD, BOD, and COLR) at a 95% confidence level. Minitab version 17.0 software was used for the Brownian coagulation-flocculation kinetic theory. The coagulation data were fitted into Eqs. (2)(3), to determine whether the mechanism of the CF treatment of AQEF with PNSC adheres to the von Smoluchowski's perikinetics concept 52 using the regression coefficient (R 2 ) as the criterion 47 . The coagulation and aggregation kinetics were investigated by plotting 1/C t with time (t) (using Eq. (2)) and Ln C t with t (using Eq. (3)) 43,53,54 . TD particle concentration was derived by converting values of TD (in NTU) to particle concentration (TDSP-total dissolved and suspended solids) (in mg L −1 ) using a calibration factor of 1.0912 (Fig. 3).
where C o and C are the initial TD (in mg L −1 ) and TD (in mg L −1 ) at any given period, t; K b (the constant of reaction rate); α is the reaction order. β F is a function of flocculation transport for the mechanisms of shear, Brownian, and differential sedimentation which is described by Eq. 4 52 given as: where K BC is 1.38064852 × 10 -23 J K −1 (the Boltzmann's constant), η is 2.6 m Pa s (the viscosity of the AQEF), ε e is the efficiency of collision, and T = absolute temperature. The rate of decline in the concentration of AQEF particles ( −r p ) at early Brownian kinetic coagulation (i.e. 30 min) is determined through Eq. 5 55,56 : where α is the coagulation reaction order, K b is the Menkonu constant rate of coagulation, and C t is the concentration of the particles (the total suspended and dissolved particles (TDSP) at t). K b can be obtained through Eq. 7 53,57 : where K sb is the von Smoluchowski's rate constant for fast coagulation. The diffusivity ( D 1 ) can be evaluated through Eq. 8 58,59 : where r is the particle's radius, r and B ff (the friction factor) can be evaluated using Eqs. 9 and 10: In practice, the particle distribution plot for CF with time may be depicted as follows Eq. 11 60 : where the values for p corresponds to the singlets (p = 1), doublets (p = 2), and triplets (p = 3) class of particles; τ f is the fast coagulation period and half-life ( τ f 1/2 ) evaluated using Eqs. (12-13) 59,61-63 : Cost estimation and energy consumption. Models associated with estimating costs have to be specific, with detailed implementation methodology. Also, models need to analyze cost against important material specifications 64 . In this current work, the total cost ( T c ) for the treatment of 1 L of the AQEF was evaluated using the expression shown in Eq. (14): where C L is the cost of coagulant production and C ce is the cost of energy. The energy consumption (E) was evaluated using Eq. (15): where P m is the power consumption by the machine (40 kW), L is a load factor (in a full mode so L = 1), t is the time of usage of the machine (0.25 h), and C is the energy estimated cost (€0.14/KWh) in Nigeria as at September 9, 2022.
Coagulation-adsorption kinetics studies. Coagulation phenomena can be modeled theoretically considering as an adsorption-like process 65 . Polyelectrolytes may destabilize materials by a mechanism that combines the effects of charge and adsorption 40 . A polyelectrolyte must be able to eliminate organic COLR to be universally acceptable for use in the treatment of water 40,66 . In this study, the COLR outputs (mg L −1 ) were investigated for the study of the adsorptive constituent of the coagulation-flocculation process, considering the effectiveness of the adsorption technique for the decrease of COLR from the medium. To analyze the sorption kinetics of the treatment process, the nonlinear pseudo-first-order (PFO) Eq. 16 67,68 , pseudo-second-order (PSO) Eq. 17 [69][70][71] and Elovich (Chemisorption) Eq. 18 72 kinetic models were tested.
where q t is the amount of adsorbate adsorbed at time t (mg g-1 ), q e is the adsorption capacity in the equilibrium (mg g-1 ), K 1 is the pseudo-first-order rate constant (min −1 ), K 2 is the pseudo-first-order rate constant (g mg −1 min −1 ) and t is the contact time (min); α is a constant related to chemisorption rate and β is a constant which depicts the extent of surface coverage.

Validity of the kinetic models' fittings.
To validate the adsorption kinetics models used in the study, in addition to the fixed correlation coefficient (R 2 ), the parameters of Marquardt's percent standard deviation (MPSD), hybrid error function (HYBRID), and sum of the errors squared (ERRSQ) were also evaluated, which can be described as Eqs. (19)(20)(21) respectively: where q e,exp and q e,calc (mg/g) are the experimental and calculated amounts of color adsorbed, respectively; n is the number of measurements made and p is the number of the test elements.

Results and Discussion
SEM and FTIR characterization. The SEM image magnification (1000 × to 2000 ×) of the PNSC is shown in Fig. 4a,b. As seen in Fig. 4, high surface heterogeneity can be observed on the PNSC, suggesting the availability of highly active adsorption sites on the active coagulant for enhanced adsorption mechanisms 73,74 in the CF process. Also, irregular and rough granular structures can be observed in the PNSC. In particular, these irregular and rough granular surfaces are necessary features of coagulant types concerning the adsorption of dissolved solids and aggregation of suspended solids. These features will promote colloidal particles' attraction, agglomeration, capturing 75 and promote their sedimentation.
The FTIR spectrums of the PNSC are shown in Fig. 5. The presence of N-H stretching in the spectrum of the PNSC confirmed the existence of amino compounds (protein). The FTIR analysis indicates that the carboxyl (C=O), hydroxyl (O-H), and amino or amide (N-H) groups, in addition to hydrogen bonding, were present in the structure of PNSC. These were proven functional groups active in the CF process 76 . The presence of C=O groups will serve as an anion bridge for divalent metal cations such as Mg 2+ and Ca 2+ at the surface of the particle to induce coagulation activity 77 . The O-H stretch, and free hydroxyl of alcohols and phenols, are strong evidence indicated by the sharp band observed on the PNSC. This hydrogen bonding aids particle adsorption. The presence of -OH stretching indicates that the active PNSC samples are hygroscopic in nature 78 . The performance output indicates that introducing a dosage of PNSC higher than the optimum (0.2 g L −1 ) caused the particles' surface charges to reverse due to lots of adsorption sites per PNSC particle, which inhibited the effectiveness of removal and interparticle bridging 79,80 .The influence of the PNSC-induced flocculation process was observed to be linked to the optimal dosage and subsequent efficiency of removal 81,82 .The findings suggest that the optimum dose (0.2 g L −1 ) of the active PNSC was most effective for the clarification of TD and other contaminants present in the aquaculture effluent. The low optimum dosage of PNSC ≤ 0.2 g L −1 indicates that the main mechanisms behind the CF process are charge neutralization and adsorption 82,83 . Also, the lower dose of 0.2 g L −1 being optimal will minimize sludge generation while reducing expense and environmental impact 84 . The one-way ANOVA test was performed to understand the significance of the changes in dosage with removal parameters (TSS, TD, COD, BOD, and COLR) at a 95% confidence level. A p value of 0.568 was evaluated at ( p > 0.05 ). An F value (0.75) < 1 recorded implies the effect of dosage on the reduction of TSS, TD, COD, BOD and COLR from AQEF using PNSC is not significant statistically. .37%, respectively. The outcome established that the optimal reduction of the selected contaminants was significantly successful in acidic media. Moreover, it can be observed from the plot of Fig. 7 that the PNSC performed poorly at pH 6 which plots removal efficiency ≤ 65%. The authors reasoned that the impact of increasing pH > 4 on particle electrophoretic mobility resulted in little surface charge neutralization, thus resulting in low contaminant removal rates. The pH modification above neutral (pH 7) transcends to the occurrence of comparable charge with more availability of OH-, leading to a rise in electrostatic repulsion of the AQEF particles. The interpretation of Fig. 7 suggests that operating PNSC outside the optimum pH window resulted in poor stability of the suspended particles in the AQEF 40,64 . Furthermore, it can be inferred from the results that particle agglomeration was enhanced at low pH due to decreased in inter particle repulsions 85,86 and changing effluent chemistry. The outcome of the PNSC-driven pH modification of AQEF is an indication that charge neutralization and inter-particle bridging effect of the polymeric material aided adsorption, and agglomeration of the various contaminants under investigation, thereby converting the particles to flocs that settle easily. Also, a combination of the carbohydrate content (usually the C=O and -OH groups), and the active metals from the complex salt extraction solution (MgCl 2 + NaCl + KCl + CaCl 2 ) was deposited on the surface of the active PNSC allays the fear associated with the protein denature at optimum pH 4. This outcome accounted for the reduction of TD, TSS, BOD, COD, and COLR concentrations which transcends to removal efficiencies ≥ 85%. The result also confirmed the PNSCdriven coagulation-flocculation treatment favored the removal of TD and TSS compared to BOD, COD, and COLR present in AQEF under acidic media with a corresponding removal efficiency ≥ 90%. In a study conducted by Beltrán-Heredia et al. 87 , the proteinic-cationic characteristics of M. oleifera, according to the authors, may indicate improvement of the coagulant activity at a low pH of 4 where the efficiency was decreased from pH 4-10; they observed a similar result for a tannin-based coagulant were a swift decrease was seen. The optimum pH (4) reported for PNSC seems to support other findings reported about green biocoagulants in published works of Menkiti and Ejimofor 62 , Ejimofor et al. 88 , and Okolo et al. 89 . Considering the increase in PNSC-driven TD and TSS removal efficiency, the authors reasoned that the pH window had a significant antagonistic effect on the sorption capacity of PNSC compliance with AQEF. The finding is a reasonable agreement reported in published works in the literature 62,88,89 . The authors reasoned that PNSC like most chemicals (alum and ferric-based coagulants), is probably acidic in nature which produces a drop in pH in the water medium. The pH adjustment of the AQEF resulted from the changing effluent chemistry. The optimal pH (4) indicates that microbial and bacterial activity in the finished effluent will be reduced and is considered an advantage in this case. A p value of 0.013 ( p < 0.05 ) and F value = 4.19 > 1 were evaluated. This implies the effect of pH on the reduction of TSS, TD, COD, BOD and COLR from AQEF using PNSC is statistically significant.
Effect of temperature on TD, TSS, BOD, COD, and COLR reduction. Figure 8 was drawn to determine the optimum temperature for the coagulation-flocculation process. The outline of Fig. 8 shows the impact of temperature on TD, TSS, BOD, COD, and COLR reductions on AQEF compliance with PNSC. The analysis of the effect of temperature on the clarification efficacy of the active coagulant was tested at 303, 313, and 323 K under optimal operating conditions (dosage of 0.2 g L −1 and pH). The outcome confirmed that at 303 K, maximum TD, TSS, BOD, COD, and COLR elimination rates corresponding to 90.35, 89.57, 82.83, 82.34, and 61.53% were attained. The output corresponds to the height of the bar charts presented in Fig. 8. The findings established that the least TD, TSS, BOD, COD, and COLR removal rate was recorded operating the PNSC-driven CF treatment at the temperature of 323 K. The removal rate of the contaminants contained in the AQEF decreased intermittently as the temperature increased. The results obtained confirmed that the optimum temperature that produced the best removal rate was recorded at 303 K. The author reasoned that the decrease in TD, TSS, BOD, COD, and COLR removal rate with rising temperature can be attributed to the haphazard movement of pollutant particles produced by increased kinetic energy. The increase in temperature prevented the particle's trapping to the PNSC surface from forming flocs and led to a reduction in floc size 90 . The colloidal particles formed from the CF process spread widely apart rather than agglomerating together to create bigger flocs and prevent the  Most feasibly, increasing temperature above the optimum (303 K) would impair the bio-coagulant performance along with the adsorption potential of its active sites and functional groups. These findings are consistent with the observation reported from the influence of the optimum dosage, and pH on PNSC-driven coagulation treatment of the AQEF on TSS, and organics removal from AQEF. A p value of 0.001 ( p < 0.05 ) and F value = 12.34 > 1 were evaluated; this outcome implies that the effect of temperature on the reduction of TSS, TD, COD, BOD and COLR from AQEF using PNSC is significant. Figure 9 illustrates the influence of settling time on contaminant removal from AQEF compliance with PNSC. The impact of the coagulation-flocculation settling period on the reduction of the pollutants (TD, TSS, BOD, COD, and COLR) was investigated by varying the settling time from 0 to 60 min. The result showed that pollutants elimination improved substantially as settling time increased until equilibrium was reached at optimum operating conditions (pH 4, a dosage of 0.2 ml −1 , and 303 K). The result shows that, across all contaminants, the removal efficiency increased consistently from 10 to 92% as settling time increased from 0 to 60 min until stability was attained. The authors reasoned that the rapid settling time was aided by the formation of larger and denser flocs resulting from the biopolymer chain's adhesion to the particles in the effluent and the charge on the surface of the active coagulant 39 . Across all contaminants removed from the effluent, equilibrium was attained after 40 min. This outcome proved that the coagulant aligns with pollutants in the effluent, leading to the reduction of the contaminants in the AQEF. At the optimal dose (0.2 g L −1 ) and equilibrium settling time (40 min). The maximum TD, BOD, and COD removal corresponding to 90.35, 82.38, and 82.11% were realized. While removal rate of 65.77%, and 88.84% were recorded for COLR and TSS at 35 min. The performance output was consistent with the observations reported on green coagulants by several authors 46,93 . The experimentation indicates that settling time, pH, and the temperature had the most significant effect on the overall performance of the active PNSC as a green coagulant. The removal efficiency of PNSC was largely dependent on the antagonistic effect of pH, temperature, and settling time. The dosage had a ceiling effect on the clarification efficacy of the bio-coagulant, with a low tendency to form sludge 94 . Also, a p value of 0.001 ( p < 0.05 ) and F value = 24.23 > 1 were evaluated. This output confirmed the effect of settling time on the reduction of TSS, TD, COD, BOD, and COLR from AQEF using PNSC is significant statistically.   Fig. 10a,b. Figures 10a, b show the first and second-order CF kinetics at the optimum operating conditions (0.2 g PNSC L −1 and 303 K), obtained by comparing Eqs. 2 and 3, respectively. The summary of the kinetic parameters recorded at the optimum conditions are shown in Tables 2 and 3. The kinetics investigations proved that the PNSC-driven coagulation rate constant increased intermittently from2.4 × 10 -3 L mg min −1 to 4.1 × 10 -3 L mg min −1 with a range of coefficient of determination 0.9333 ≤ R 2 ≤ 0.9796 for first-order ( Table 2). The maximum rate constant (4.1 × 10 -3 L mg min −1 ) was recorded at pH 4 and transcends to optimum TD and TSS removal efficiency ≥ 90%. A similar outcome was observed with the second-order CF kinetics (Table 3), with the maximum value of the rate constant (6.35 × 10 -2 L mg min −1 ) with a corresponding R 2 (0.9679) recorded at pH of 4. The least flocculation rate recorded at pH 6 translates to contaminant reduction efficiency ≤ 75%. The kinetic data confirmed that, at the initial stage, the PNSC-driven coagulation dynamics were synergetic with the first-order and second-order kinetic models (where α = 1 and 2). The authors reasoned that the initial stages where the CF dynamics obeyed 1st-order kinetic is attributed to a shift from theoretical expectation but in line with empirical evidence 95,96 . As the coagulation reaction proceeded towards the optimum the dynamics of the PNSC in AQEF adjusted to the 2nd-order (perikinetics flocculation) with a corresponding R 2 ≤ 0.9796. Conse-   www.nature.com/scientificreports/ quently, the summary of the kinetic parameters (Table 3) was estimated for the second-order PNSC-driven CF kinetic parameters. The findings from the kinetic results established that the highest collision efficiency (ε e ) of 2.96 × 10 19 L mg −1 was recorded at pH 4. The performance of PNSC in AQEF proved that higher values of ε e resulted in high energy of kinetics, indicating a tendency to lower the zeta potential. The performance output of PNSC in AQEF is consistent with the pH results reported for green coagulants in published research works available in the literature 97,98 . The value of the half-life (τ f 1/2 ) for the CF treatment is an important parameter that is linked to optimal particle aggregation 99 . The values of half-life ( τ f 1/2 ) obtained from the PNSC-driven kinetics decreased intermittently as the pH of AQEF increased from 4 to 6. The outcome leads to colloidal destabilization 100 at τ f 1 2 ≤ 1.80 min, yielded optimum coag-flocculation efficiency ≥ 90.20%. The low values of τ f 1/2 recorded confirmed the theory of fast coagulation is prevalent on CF treatment of the AQEF 101 . The values ofthe PNSC-driven flocculation period ( τ f ) and half-life ( τ f 1/2 ) is shown in Table 3. The maximum value of Brownian collision factor (Β f ) ≥ 0.13 was recorded at pH 4.This output is connected to the collision efficiency 96,102 necessary to reduce the double layer compression or destabilize the particles to achieve low τ f 1/2 values necessary for rapid flocculation to occur.The optimum performance of PNSC in AQEF was prevalent on the minimum particle concentration (C 0 = 119.26 mg L −1 ) at flocculation period (t f = 3.78 min) and half-life ( t 1 2 = 1.90 min). These outputs were recorded at R 2 ≤ 1, confirming statistical fit of the CF data to the kinetic model were significant. The findings established that the best performance of the biocoagulant corresponds to the maximum coagulation-flocculation rate constant (K b ) ≥ 0.06 at pH 4 (Table 3), which transcend to TD, TSS, BOD, COD, and COLR efficiencies of removal corresponding to 90.35, 89.57, 82.83, 82.34, and 65.37%, respectively.

Settling time influence on TD, TSS, COD, BOD, and COLR reduction.
Particle distribution behavior of the process. The time-evolution and aggregating of the different classes of particles: singlets, doublets, and triplets based on size variation were employed to forecast their behavior as time changes. The time-evolution and particle aggregation parameters presented in Table 4 were evaluated following Eq. 11. The distribution pattern of the different aggregates formed in terms of the particle concentration per cubic meter in AQEF was investigated using values of τ f 1/2 , C 0 and K b derived from 2 nd order kinetics at optimum conditions ( Table 3). The time-evolution and particle distribution (Table 4) was drawn to illustrate the computed values of aggregates (C 1 -ƩC) of the triplet, doublet, and singlet particle counts, as well as the overall particle counts. Figure 11 was drawn to illustrate the time evolution and particle distribution for the impact of PNSC on AQEF. The summary of the distribution of the concentrations of the particles per cubic meter is presented in Table 5. The varying particles concentration number (C1-ƩC) with time (t) at the optimum operating conditions were evaluated by substituting values of τ f 1/2 =1.90 min, C 0 = 119.26 g L −1 , and K b = 0.0635 Lg −1 min −1 , into Eq. 11. Figure 11 shows the time evolution and aggregation of the particles that characterize the AQEF. The plot depicts the trajectory of the agglomeration of the particles and the settling characteristics of the various aggregates (C1-ƩC) with time (t). The rapid destabilization of singlets accompanied the formation of doublet and triplet counts. The singlet class particles declined more rapidly than the overall number of particles 93,103 .The authors reasoned that Brownian coagulation dominated the fundamental particles 104 . The mechanism of aggregation of the particles was described by a combination of charge neutralization and sweep flocculation 11 . The curvatures of the curves in Fig. 11 show that the estimate of singlet, doublet, and triplet aggregates dropped systematically throughout time. The protonated amine groups often destabilize the negative charges and the zeta potential, lowering or eliminating the DLVO energy barrier and allowing for more species interactions 98 . Conclusively, it can be inferred from the outcome of the time evolution and particles aggregation that charge neutralization mechanism occurred under the influence of the optimum dosage (0.2 g L −1 ), and most of the pollutant particulates were cleaned up from the effluent medium via gravity, settling after being entwined in the protein complex.  pseudo-second-order, and Elovich (chemisorption) kinetic models (Eqs. [16][17][18] were tested on the data and used to describe the mechanism of the adsorptive uptake of COLR from AQEF using PNSC at optimum conditions. Figure 12 illustrates the sorption kinetics plots obtain from values of q t plotted against time. The summary of the adsorption kinetics parametric values is presented in Table 5. The result shows that under the optimum conditions, the range of values of the correlation coefficient 0.9737 ≤ R 2 ≤ 0.9979, and values of estimated errorsquared in the range of 0.0001 ≤ ERRSQ ≤ 0.00002 were recorded for the coagulation-adsorption kinetics. The highest adjusted-R 2 ≤ 1.0 indicates that a particular model best fits the adsorption kinetic data 108 . The model with the lowest values of Marquardt's percent standard deviation (MPSD), hybrid error function (HYBRID), and the errors squared (ERRSQ) value best suits the data to the kinetic model. The use of error functions is statistically more accepted than the adjusted R 2 since it indicates low error obtained for the data. The findings from the adsorption kinetics established that the lowest model statistical metrics: HYBRID (0.0011), ERRSQ value (2 × 10 -5 ), and MPSD (7.28), were recorded for the PFO kinetic, as shown in Table 5. The results showed that the PFO kinetic model best describes the coagulation-adsorption mechanism. The outcome indicated that PNSC-driven adsorptive uptake of COLR from AQEF conformed to the PFO model. This outcome sug-  www.nature.com/scientificreports/ gests that the process of COLR reduction from AQEF is not chemically controlled. The models' adjusted-R 2 (0.9979) was closest to unity, confirming the goodness of fit of the kinetic data 94 . The maximum sorption capacity (q e = 0.2509 mg/g) recorded for the PFO model correspond to 70% COLR removal efficiency.
Comparative analysis of PNSC with other coagulants. The comparative analysis of PNSC and other green coagulants on AQEF expressed in terms of %TD, %TSS, %COLR, %BOD, and %COD measurements are presented in Table 6. The comparative analysis (Table 6) shows a consistency in the overall performance of PNSC with the published results reported by other researchers. The findings from the current research work established efficacy of PNSC is as a natural coagulant is favorable for clarification of TD from AQEF with a higher removal rate (90.23%) compared to 89% reported for Moringa oleifera seed extract 105 , 88% for Brachyura  www.nature.com/scientificreports/ shell waste 106 , 88% TD removal efficiency reported for bio-polymer chitosan 107 applied in the treatment of AQEF. However, the efficacy of PNSC in the removal of TSS (89.90%) was consistently higher than 78.82% reported for Serratia marcesens 108 , 62% for bio-polymer chitosan 107 , and 82.70% for Neem leaves 109 in AQEF. The COLR reduction rate of 65.70% was obviously higher than 54.77% reported for Parkia biglobosa seeds 10 in the AQEF system. The COD and BOD removal efficiency of 82.34 and 82.11% recorded for PNSC in the AQEF system was also reported higher than the 75% reported for Sesamum indica 12 , 62% and 52% efficiencies of removal reported for bi-polymer chitosan in AQEF system 107 . The authors reasoned that the circumstances under which these efficiencies were achieved are proven from the time evolution and aggregate distribution, coagulation, and adsorption kinetics. Generally, most writers focused primarily on the decrease of TD and TSS rather than on the organic components (BOD and COD) and colour removal, which was reported by the current study. The study demonstrated that the primary substances eliminated by the PNSC-driven coagulation-flocculation treatment of AQEF are turbidity and suspended particles. Consequently, instead of using chemical-based substances, natural coagulants such as PNSC can be employed.
Cost-energy considerations on the treatment process. Within a given environmental terrain where there are product concepts, alongside quality (and environmental) data, there is the cost data that has to be fed into the concept comparison matrix. This above-mentioned perspective was among the green quality function that Dong et al. 110 deployed in ascertaining key aspects of the cost estimation for environmentally conscious product development. More so, cost estimation practices continue to evolve over time, fundamentally aiming to increase by accuracy and effectiveness, largely dependent on data appropriateness as well as legitimacy, despite the fact that there is no complete theoretical model 64 . Further, besides exploring cost analysis as a decisionmaking tool to test for the feasibility of the active coagulant 11 , the cost of the CF operation has to be based on the efficacy of the active coagulant, energy consumption, and technology required to remove contaminants 43 . In this current work, the cost of treating 1 L of AQEF was calculated by taking into account the cost of the following: (a) preparing 0.2 g L −1 (optimal dose) of the PNSC; (b) energy, as well as (c) labor. To provide further explanation to these, the preparation of 0.2 g L −1 of PNSC would cost €0.17. However, the labor cost would be projected at €3.24, whereas the energy cost would be projected at €0.14/KWh. With these, computing the total cost appeared feasible, and generated €4.81, bearing in mind the considerations of costs of 0.2 g L −1 PNSC preparation, labor, and energy. Considering the above results, the techno-economic feasibility of wastewater and effluent handling would most likely necessitate the use of low-cost materials.

Conclusion
The aquaculture effluent pollutant removal using Picralima nitida seeds extract via coagulation-flocculation treatment was investigated. In the current research work, pilot scale experimentation conducted, and the results was focused on the bio-coagulation performance, coagulation-flocculation/adsorption kinetics, particle temporal evolution, and the cost-benefit analysis of employing PNSC to treat 1 L of AQEF. The results obtained show that TSS, TD, COD, BOD, and COLR concentration in aquaculture effluent (AQEF) were reduced using the novel PNSC. The effects of PNSC dosage, pH, temperature, and settling time on the reduction of pollutants were examined and their statistical significance were tested via ANOVA. The data were explored through the sorption and flocculation kinetics equations. The PNSC possessed the amino (N-H) and hydroxyl (O-H) groups-proven to be very beneficial for coagulation-flocculation. The efficacy of the active coagulant in the aquaculture system translates to the order of removal of the pollutants TD > TSS > organics (BOD and COD) > COLR. Maximal TD reduction = 90.35%, TSS = 88.84%, COD = 82.11%, BOD = 82.38% and COLR = 65.77% at 0.2 g L −1 dosage of PNSC, pH 4, and 303 K was achieved. The p values and F-values obtained from ANOVA analysis inferred that the pH, temperature and settling time had a significant effect on the pollutants removal. Von Smoluchowski's kinetics fit the results. At perikinetics condition, the K b (reaction rate) and t f 1 2 (half-life) correspond to 0.0635 L g −1 min −1 and 1.9 min under the ideal circumstances. The sorption data fitted the Lagergren more than the Ho adsorption model. The total cost of using PNSC to handle 1 L of AQEF was €4.81. The final BI (0.98 > 0.3) suggests that the pretreatment using the CF via PNSC improved the biodegradability of AQEF. The PNSC-driven coagulation process demonstrates a feasible solution for the clarification of TD, TSS, COD, COLR, and BOD removal from the aquaculture effluent system. Although an optimum pH (4) recorded vary slightly outside EPA standard for effluent discharge (5.5 ≤ pH ≤ 9), the outcome calls for further studies to be conducted to ascertain for the pH modification of PNSC in effluent system. Under such prevalence, PNSC can be classified as a flocculant or coagulant aid. Other challenges include experimental scale-up for industrial acceptance of PNSC and lack of research regarding the practical usage of PNSC with other effluent sources and coagulant aid. To overcome these shortcomings, the direction of future research could focus on undertaking modeling simulations to estimate the potential scale-up feasibility considering specifically contexts of different locations, testing PNSC with coagulant aids where pH modification is concerned, considering the availability of materials (to make PNSC), as a start.

Data availability
The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.