INTRODUCTION
Tremendous development in the field of green nanotechnology has been noted due to environmental benefits, less time duration, and cost-effectiveness. Biosynthesis of nanoparticles is applied to different scientific disciplines including material science and biomedical applications [1]. Different methods viz. physical, chemical, and hybrid methods are being used for the synthesis of nanoparticles. The high toxicity of chemicals employed in these conventional methods limits their applications. Moreover, these methods generate toxic wastes which are hazardous to the environment and harmful to humans. Therefore, eco-friendly and green approaches are now being used as a replacement for those expensive, outdated and inefficient methods. Comparing the cost-effectiveness and side effects of the conventional methods, researchers are mainly focusing on biological approaches [2]. The nanoparticles generated by the biological process show higher catalytic activity improved contact between metal salt and enzyme, greater surface area to volume ratio [3]. Nanoparticles exhibit their remarkable potential in the field of biology and medicine based on specific characteristics such as size, morphology, etc. [4].  
Green synthesis of various nanoparticles using plant extract has been reported by several workers. Sayadi et al. [5] reported the green biosynthesis of palladium nanoparticles using Spirulina platensis algal extract. In addition, Yazdani et al. [6] also reported the green synthesis of palladium oxide nanoparticles using  Dictyota indica seaweed for adsorption. 
Heavy metal pollution is becoming one of the most serious environmental problems globally [7]. The treatment of wastewater containing heavy metals is mandatory due to their persistence in the environment. In order to detoxify heavy metals, different techniques like photocatalytic oxidation, UV irradiation, electrochemical, bioremediation, ion-exchange, chemical coagulants, ozonation, reverse osmosis, and adsorption have been employed [8]. Among these technologies, the use of nano-based adsorbents is the more convenient method for the removal of heavy metals from the aqueous environment [9]. The application of iron oxide-based nanomaterial is more attractive for heavy metal removal from contaminated wastewater due to their important features like small size, high surface area, and magnetic properties [10]. Arsenic removal from mining effluent was reported using plant-mediated green synthesized iron nanoparticles [11]. Efficient removal of lead from aqueous solution using FeNi3@SiO2 magnetic nanocomposite was reported [12]. Removal of nickel and chromium from aqueous solution using Copper oxide nanoparticles was reported by Hosseini et al. [13].
Among the toxic metals, lead (Pb) is in the limelight due to its major impact on the environment. Chronic exposure to high amounts of Pb (II) can result in various and extensive damages to systems of the body, reproductive system, basic cellular processes, and brain functions. Furthermore, lead poisoning causes high blood pressure, anemia, insomnia, liver disease, bone cancer, headache, dizziness, irritability, weakness of muscles, hallucination and renal damage, coma and ultimately causing death [14]. Therefore, the removal of Pb (II) from the wastewater is mandatory.
Bioflocculants have gained tremendous attention owing to their unique flocculating characteristics, harmlessness, non-toxic and biodegradable nature which aids in creating an eco-friendly environment [15]. They serve as a replacement for chemical-based flocculants [16] and are widely used in various fields including synthesis of nanoparticles, removal of heavy metal and dyes, mineral processing, drinking water, and wastewater treatment, etc. [17, 18]. 
There are reports on the use of microbial bioflocculants for the removal of heavy metals from an aqueous environment [19]. Application of bioflocculant passivated copper nanoparticles on wastewater treatment has been reported recently [17]. But reports are scanty on bioflocculant mediated nanoparticles for the removal of heavy metals from an aqueous environment. 
Fe nanoparticles have attracted much attention due to their unique properties including superparamagnetism, surface volume ratio, greater surface area, and easy separation methodology. Fe nanoparticles have a great potential for their applications as wastewater treatment adsorbents, catalytic materials, pigments, gas sensors, coatings, ion exchangers, magnetic data storage devices, magnetic recording devices, magnetic resonance imaging, and bio-separation [20]. It is well-established fact that iron nanoparticles show greater catalytic activities for dye reduction and removal. There has been a great interest in using plant extract in synthesizing the Fe nanoparticles for industrial applications. The iron nanoparticles are safer and cheaper [21].
Therefore, the current study is focused on the green synthesis of iron nanoparticles (FeNPS) using bioflocculant extracted from Okra and its application towards the removal of Pb (II) from wastewater. The statistical optimization of the metal removal process was done using response surface methodology (RSM).

MATERIALS AND METHODS
Chemicals
All chemicals viz. ferric nitrate (Fe(NO3)3, ethanol, methanol, acetone, glucose, bovine serum albumin, and trifluoroacetic acid were purchased from Himedia company and used without further purification.

Plant Material 
The plant Okra (Abelmoschus esculentus (L) Moench) belonging to the family Malvaceae is widely grown through tropical and subtropical regions of the country. Fruits were purchased from the local vegetable market located at Vellore, Tamil Nadu, India, and used for the study. 

Collection and analysis of wastewater 
The wastewater sample was collected from a Common Effluent Treatment Plant (CETP) located in Ranipet, Vellore, Tamil Nadu, India. The Physico-chemical characteristics of the wastewater sample were analyzed using standard analytical methods [22]. The presence of heavy metals viz. Ni2+, Zn2+, Cd2+, Cu2+ and Pb2+ was determined using Atomic Absorption Spectrophotometer (AAS) (VARIAN SPECTRA 240 model). 

Extraction and Purification of Okra Bioflocculant 
The fresh fruits of okra were sliced, homogenized, and extracted with cold water containing 1 % w/v of metabisulphite. The crude mucilage was centrifuged at 3000 rpm for 5 minutes, precipitated from the supernatant with acetone, and washed several times. The cream-colored product was obtained and dried under a vacuum in a desiccator. A light brown colored powder was obtained after the removal of complete moisture. The dried powder was pulverized and stored in a bottle at 4°C till used [23].
The dried powder of bioflocculant (BF) was poured into two volumes of ice-cold solvents such as ethanol, methanol, and acetone (1:2 v/v) to precipitate the bioflocculant. It was allowed to stand overnight at 4°C in the refrigerator. The resulting precipitate was separated by centrifugation at 10,000 rpm for 15 min and washed by redissolving in the double-distilled water followed by reprecipitation. The process was repeated twice. The precipitated bioflocculant was dehydrated for 2 hours for the removal of water and solvent in a vacuum using a rotary evaporator.  The extract was subjected to dialysis against distilled water and kept overnight at 4°C [18].

Characterization of Bioflocculant
The purified bioflocculant was subjected to chemical characterization. The total carbohydrate content of bioflocculant was determined by the phenol-sulphuric acid method using glucose as a standard solution [24]. The total protein content of bioflocculant was determined using the Bradford method with bovine serum albumin as a standard [25]. The amino acid was estimated by the ninhydrin method [26]. After hydrolysis of the okra bioflocculant with 2 M, trifluoroacetic acid at 120°C for 2 h, neutral sugars, uronic acids, and amino sugars were determined with anthrone method, carbazole sulfate acid method, and the Elson-Morgan method, respectively [27]. Instrumental analysis viz. Fourier Transform Infra-red Spectroscopy (FT-IR), X-ray Diffraction (XRD), Scanning Electron Microscopy (SEM), Energy-dispersive X-ray spectroscopy (EDX), and Atomic Force Microscopy (AFM) analysis were also performed for characterization of purified bioflocculant.

FTIR and XRD analysis 
The functional groups of purified bioflocculant were investigated using Fourier transform infrared (FTIR, PerkinElmer MA, USA). The dried sample was blended with KBr powder and pressed into pellets for FTIR spectra measurement spectrum in the frequency range of 500-4000 cm-1 [18]. 
X-ray diffraction (XRD) patterns of bioflocculant were conducted with X’ Pert Pro P analytical diffractometer, Germany using monochromatic Cu Kα X-ray radiation (λ 0.154 nm) running at 40 kV and 30 mA [28].

SEM, EDX, and AFM analysis
Scanning electron microscopic studies were carried out to elucidate the surface morphology of the okra bioflocculant using HITACHI model (Carl Zeiss) high-resolution electron microscope to verify morphology and homogeneity of particles. Bioflocculant suspension was centrifuged at 10,000 rpm for 15 min. The pellets were resuspended in deionized double distilled water. A drop of the bioflocculant suspension was placed on a glass coverslip and air-dried. These samples were chemically fixed for 24 hours at 40°C using a final concentration of 2.5% glutaraldehyde. The sample was rinsed with distilled water 3 times to remove traces of glutaraldehyde. Then samples were dehydrated in grade series of ethanol (30, 50, 75, 85, 95, and 100%), and air-dried under vacuum [18]. 
The EDX analysis was done with an X-ray detector to examine the elemental constituent of the okra bioflocculant.
The surface topography of the bioflocculant was analyzed using Atomic force microscopy (AFM). About 5-10 mL of bioflocculant suspension was distributed on a mica disc (Pelco mica disc 10 mm) by a spin rotating plate and absolute ethanol was dropped over the sample to fix it on the mica disc. Then, the mica sheet was air-dried to eliminate the residual ethanol [29]. The AFM images were captured by Nanosurf Easyscan 2, Switzerland.  

Bioflocculant mediated green synthesis and characterization of iron (Fe) nanoparticles
The iron nanoparticles were synthesized using purified Okra bioflocculant.  The Okra fruits (5 g) were washed thoroughly with distilled water and cut into fine pieces and were subsequently macerated in a 50 mL mixture of double distilled water and methanol (1:1). This liquefied mixture was subjected to centrifugation at 10,000 rpm for 5 min at 4°C and the supernatant was separated. The supernatant obtained was used as a reducing agent for the synthesis of the iron nanoparticles. For the synthesis of Fe nanoparticles, bioflocculant (5 ml) was added dropwise into the solution of 10 mM ferric nitrate (Fe(NO₃)₃ under magnetic stirring conditions. Later the content was placed onto a rotary orbital shaker at 200 rpm for 72 h at 30°C temperature under dark conditions. The reduction of Fe ions was monitored at intervals of 24 h followed by measurement of the UV–Vis spectra using a spectrophotometer (JASCO V-670 PC). To find out the maximum absorption, a spectral scanning analysis was carried out by measuring the optical density of the sample after 72 h using a UV–Vis spectrophotometer from wavelength 300-700 nm. The synthesized Fe nanoparticles were extracted by centrifuging the sample at 13,000 rpm for 10 min after 72 h of the incubation period. For removing the organic impurities, the recovered pellets were then washed with 50 ml of 1:1, methanol-water solution, and pellets were freeze-dried using a lyophilizer. The dried pellet of synthesized nanoparticles was ground into powder form and subjected to X-ray diffraction (XRD), Fourier-transform infrared spectroscopy (FTIR) [30], scanning electron microscopy (SEM) analysis, Energy-dispersive X-ray spectrometer (EDX) analysis [31], Atomic Force Microscopy (AFM) analysis following the standard method as demonstrated by Vitta et al. [32], Transmission electron microscopy (TEM) under an accelerating voltage of 100 kV (TECNAI G20 TWIN), and dynamic light scattering (DLS) (ZEN 3600, Malvern, Worcestershire, UK) [11].

Flocculation assay using simulated wastewater containing heavy metals
The heavy metal removal efficiency of the bioflocculant and bioflocculant mediated iron nanoparticles (BFFeNPs) was determined using simulated wastewater containing heavy metals. For this purpose, aqueous solutions containing Ni2+, Zn2+, Pb2+, Cd2+, and Cu2+ at concentrations of 500 µg/L each were prepared. The biofocculation experiment was performed with purified bioflocculant and BFFeNPs. The samples were maintained at pH 7 and the final volume was made up to 50 mL. After the bioflocculant and BFFeNPs (0.2 g/L) were added, the samples were shaken on a rotary shaker at 150 rpm at 25º C, for 24 hours. After 24 hours the supernatant was filtered through a 0.45 µm syringe filter. The metal concentrations in the solution were measured with an Atomic Absorption Spectrometer (VARIAN SPECTRAA 240) as well [33].
The flocculating activity was calculated using the following equation : 

 Flocculation activity (%) 

Where F2 is the initial metal concentration and F1 is the final metal concentration.

Heavy metal removal from real wastewater using BFFeNPs
The experiment was conducted with a real wastewater sample collected from the Common Effluent Treatment Plant (CETP) and the procedure was followed as described above in the flocculation assay using simulated wastewater. Presence of heavy metals viz. Ni2+, Zn2+, Cd2+, Cu2+, and Pb2+ were detected in the wastewater sample. Additionally, the effect of different parameters on Pb (II) removal from wastewater is studied.

Effect of parameters on Pb (II) removal from real wastewater
The effects of different parameters on Pb (II) removal from real wastewater was tested using different pH value (2-10), BFFeNPs dosage (0.1-0.5 gm/L), contact time (10-50 min), and temperature (25-50°C ) keeping the other parameters at optimal conditions [34].

Process optimization for maximum Pb (II)removal
Response surface methodology using Box-Behnken design was used to optimize various parameters viz. pH, BFFeNPs dosage, contact time, and temperature for the maximum removal of Pb (II) metal ion. The quadratic model was used to analyze the data. Each factor in the design was studied at 3 different levels (−1, 0, +1) and the minimum and maximum ranges of pH, BFFeNPs dosage, contact time, and temperature variables were determined as shown in Table 1.
A design of 29 experiments was formulated and experiments were carried out in 250 mL Erlenmeyer flasks containing wastewater (100 mL) with different pH, BFFeNPs dosage, contact time, and temperature. Pb (II) removal percentages were evaluated as response 1. The 3D contour plots were prepared to evaluate the interaction between the optimized parameters, which influence the responses. The respective responses were analyzed by using a second-order polynomial equation, and the data were fitted to the equation by multiple regression procedures. 
Later, an experiment was conducted in triplicates using the optimum values for variables viz. pH, BFFeN dosage, contact time, and temperature given by response surface optimization to validate the predicted value and the experimental value of the responses. The results of the experimental design were analyzed and interpreted using Design-Expert version 12.0 (Minneapolis, MN, USA) statistical software.

RESULTS AND DISCUSSION 
Analysis of wastewater
The wastewater sample collected from a Common Effluent Treatment Plant (CETP) located in Ranipet, was black, giving a pungent smell with a temperature (28°C). The pH of wastewater was slightly alkaline (7.5). TDS and TSS in the wastewater sample were 5,534 mg/L and 2563 mg/L respectively. High values of COD were observed (1200 mg/L) compared to BOD (580.30 mg/L). Sulfate and nitrate were found to be 326 mg/L and 4.4 mg/L respectively. The average amount of heavy metal ions Cu2+, Cd2+, and Pb2+ were higher whereas Ni2+ and Zn2+ concentrations were observed as < 2 mg/L. 

Characterization of okra bioflocculant
The chemical composition of bioflocculants plays an important role in their flocculating activities. Bioflocculants contain proteins, glycoproteins, polysaccharides, lipids, and glycolipids [35].
In this study, a chemical analysis of the purified okra bioflocculant revealed that it was a sugar-protein derivative composed of carbohydrate (9.69%) and protein (19%) including neutral sugar, uronic acid, and amino sugar. Similar results were reported by Assi et al. and Acikgoz et al. where the total carbohydrate and protein content from Okra bioflocculant was found to be 9.75±7.69% and 19% respectively [36, 37].

FTIR and XRD analysis
The major functional groups present in okra bioflocculant were identified by Fourier transform infrared spectroscopy. The FTIR spectrum of bioflocculant extracted from okra was shown in Fig.1(a)which exhibited different peaks at different positions. The peak in the region of 1026.87 cm−1 refers to the stretching of the C-O-C group in polysaccharides. The band at 3279.34 cm−1is characteristics of stretching and deformations of the hydroxyl groups (OH). The absorption bands around 1200–800 cm−1 are considered characteristic polysaccharide bands. They indicate the presence of -C-O- bonding of alcohol and ether. Similar results were demonstrated by Freitas et al. [38]. The sharp peaks at 2923.25 cm−1and 2851.06 cm−1 were corresponding to the asymmetric and symmetric stretching of the group -CH-CH2 exists in fatty acids, a similar result was described by Swelam [39]. A significant band in the spectrum was observed between 1719 cm−1 to 1233 cm−1 resembling the presence of the carboxylic group. The absorption band at 1719 cm−1 attributed to the C=O stretching mode of either methylated or protonated (non-ionized) form of a carboxyl group. The band at 1412.57 cm−1 indicated the presence of C-H symmetrical deformation vibration. These functional groups serve as active sites for the attachment of colloidal particles and bioflocculant [38].
Fig. 1(b) shows the X-ray diffractogram of okra bioflocculant shows a broad peak and several halos with weak peaks and abundant noise in the band of 10°< 2θ<80°, which gives the characteristic patterns of an almost completely amorphous nature where the disordered state of molecules produces scattered bands. A similar XRD pattern was illustrated by Freitas et al. [38].

SEM, EDAX, and AFM analysis 
The SEM images of the okra bioflocculant, were studied to examine its morphology and microstructure as shown in Fig. 1(c). The surface micrograph of okra bioflocculant revealed irregular and amorphous structures. 
The EDX spectrum of okra bioflocculant was shown in Fig. 1(d), which indicated the presence of carbonaceous material with a carbon weight percentage of 51.08% and oxygen weight percentage of 42.66%. The presence of C revealed that the okra bioflocculant contained proteins which contributed to the efficiency of the material in the process of heavy metal removal [32]. Fig. 1(d) inset) shows the weight percentage of the inorganic element Mg, P, Cl, K and Ca which was detected from okra bioflocculant. Similar results were illustrated by Emeje et al. [40].
AFM is a high-resolution tool which was widely used in polysaccharide surface morphology characterization. As shown in Fig. 1(e), the 3D image of the okra bioflocculant displayed an irregular entanglement structure. These entangled appearances may be due to the Van der Waals force of attraction, intra and inter-molecular hydrogen bonds in acidic biopolymers [41].

Characterization of bioflocculant mediated FeNPs (BFFeNPs)
The instant change in color from yellow to black was observed after the mixing of ferric nitrate (Fe(NO₃)₃ with the Okra bioflocculant extracts at the ratio of 1:1. Fig. 2(a) inset), indicating the formation of FeNPs. UV spectrum shows the maximum absorption at three different wavelengths of 230 nm, 279 nm, and 305 nm which indicated the gradual formation of Fe nanoparticles as shown in Fig. 2(a). Huang et al. reported tea extract mediated synthesis of iron nanoparticles showed absorption maximum at 205 and 275 nm while Dhuper et al. demonstrated zero-valent Fe nanoparticles from leaf extract of Magnifera indica exhibited maximum absorption at 216-268 nm. Similar results on the biosynthesis of Fe nanoparticles using leaf extract were reported by Devatha et al. [42, 43, 44].
The XRD pattern of Fe nanoparticles exhibited well-defined peaks at 2θ values of 28.28 and 64.54 which correspond to the 311 and 564 and planes, respectively. As shown in Fig. 2(b), the intensity of Fe nanoparticles peaks at 2θ values of 28.28 and 64.54 reflected the high degree of crystallinity in Fe nanoparticles which were in good agreement with the results reported by Prasad et al. [30].
The FTIR spectrum of BFFeNPs is shown in Fig. 2(c). The peaks at 3304.06 cm-1 are due to the O–H stretching vibration of free hydroxyl groups of alcohol. The twin peaks at 2920.23 and 2850.79 cm-1 corresponded to the C-H tensile bond of alkanes [45, 46]. The absorption peaks at 1577.77 cm-1, 1384.89 cm-1, and 1014.56 cm-1 were attributed to C=C stretching vibration, C-H symmetric vibration, and C-H vibration respectively, which were characteristics functional group of bioflocculant. The small peak at 613.36 cm-1, 547.78 cm-1, and 480.28 cm-1 attributed to the Fe bond vibration of BFFeNPs. A similar result was demonstrated by Wang et al. [47].
Fig. 2(d) shows the SEM images of bioflocculant mediated iron nanoparticles. It can be seen that the obtained BFFeNPs have a spherical shape. Similar results have been reported for iron NPs prepared by different tea extracts and mint leaves extract [42, 45]. Moreover, the iron nanoparticles tended to aggregate probably due to the existence of polyphenols in the okra bioflocculant. Aggregation of the nanoparticles usually causes to increase in the size of nanoparticles, this is a common phenomenon during biosynthesis of metallic nanoparticles particularly when the plant extracts are used [48]. 
The elemental composition of the BFFeNPs was determined by the EDX. In the EDX spectrum (Fig. 2(e), three intense peaks of C, O, and Fe were found, which confirmed the formation of iron nanoparticles by okra-mediated bioflocculant. The presence of carbon and oxygen may arise from biomolecules of okra bioflocculant which was acted as a capping agent. Similar observations were also reported by Wang et al. for the synthesis of iron nanoparticles using eucalyptus leaves extracts. The presence of Fe (32.30%) demonstrated that iron nanoparticles have been synthesized. The weight composition of Fe synthesized by okra-mediated bioflocculant is relatively higher than that of eucalyptus leaf extract whose percentage reported to be 16.17% [47]. The weight percentages of C and O in the obtained product were 27.29 % and 40.41 %, respectively. 
AFM analysis was also performed to evaluate the nature of BFFeNPs and the response is shown in Fig. 2(f). The surface topology of BFFeNPs can be accessed from the grooves present in the dimensional views of iron nanoparticles. From the 3D view, it is observed that the size of the nanoparticle was variables in size. This behavior was noted in the case of BFFeNPs because of the presence of various bioactive molecules. Since bioactive molecules have different functional groups, which interact with one other by intermolecular forces, the particle may hold together and particle size may vary [49].
The size of the synthesized BFFeNPs was illustrated in the TEM images as shown in Figs. 2 (g) and (h). The TEM micrograph illustrated that the okra bioflocculant mediated Fe nanoparticles exhibited spherical nanostructures with an average diameter of 50 nm (Fig. 2(h) and the particles were seen to be agglomerated (Fig. 2 (g). Similar results of TEM analysis were reported where electron-dense spherical iron oxide particles having a diameter ranging from 30 to 40 nm were synthesized using Hordeum vulgare extracts and Rumex acetosa extracts [50].
The DLS analysis of BFFeNPs showed an average particle size of 52 nm. Some of the nanoparticles have irregular shapes because of agglomeration. It is observed that most of the BFFeNPs are spherical. Similar results are reported by Karimi et al. [11]. 

Flocculation assay using simulated wastewater containing heavy metals
The flocculation efficiency was tested for the purified bioflocculant and BFFeNPs using a simulated waster containing heavy metals. 
Fig. 3 showed that the removal of the heavy metals viz. Ni2+, Zn2 +, Pb2+, Cd2+ and Cu2+ was found to be 50, 43, 65, 27 and 20 %,  respectively using only bioflocculant. Compared to this, BFFeNPs was capable of removing Ni2+, Zn2+, Pb2+, Cd2+ and Cu2+ by 59, 48, 70, 29 and 23% respectively.

Heavy metal removal from real wastewater using BFFeNPs
The removal percentage of heavy metals viz. Pb2+, Ni2+, Zn2+, Cd2+ and Cu2+ from wastewater are summarized in Table 2. 
The Pb2+ removal percentage of 66.22 % and 70.14% was recorded highest among all metals using Okra bioflocculant and BFFeNPs respectively. The results on heavy metal removal signify the efficiency of BFFeNPs which was more compared to the application of bioflocculant alone.  Removal pattern followed was as follows : Pb2+> Ni2+> Zn2+> Cd2+> Cu2+  in both the cases.

Effect of parameters on Pb (II) removal
The effect of pH was monitored in the range 2–10, keeping all other factors constant, that is bioflocculant and BFFeNPs dosage of 0.2 g/L, contact time (30 min), and temperature of 25°C. Fig. 4(a) shows the removal percentage of Pb (II) ions depends significantly on pH. The percentage removal gradually increased with an increase in pH up to pH 6; above this value, insoluble lead hydroxides were seen to precipitate out. The low metal removal at pH 2 was due to the excess H+ ions and consequent protonation of BFFeNPs surface [12]. At pH 4-6 phosphate, carboxyl, and amino groups on the okra bioflocculant and BFFeNPs generate a negatively charged surface and electrostatic interactions between cationic species and this surface can be responsible for enhanced lead removal. The maximum Pb (II) removal of 67% and 72.09% was achieved at pH 6 using okra bioflocculant and BFFeNPs respectively.
The experiment was carried out to determine the optimum dosage for Pb (II) removal at optimum pH of 6, contact time of 30 min, and temperature of 25°C. The bioflocculant and BFFeNPs dosage were varied in the range from 0.1-0.5 g/L. Fig. 4(b) shows that the Pb (II) removal percentage increased with an increase in dosage of bioflocculant and BFFeNPs up to the optimum level of 0.3 g/L after which a gradual decrease in Pb (II) removal percentage was observed. The number of active sites on bioflocculant and BFFeNPs will increase with increasing dosage, so, the binding chances of Pb (II) ion with bioflocculant and BFFeNPs surface increases and Pb (II) ion removal efficiency also increases [12]. It was observed that the synthesized iron nanoparticles have a high affinity than the bioflocculant for removal of Pb (II) ion with removal efficiency above 75 % at the dosage of 0.3 g/L. Similar results were reported by Dlamini et al. [17]. 
The contact time is another important parameter for the removal of Pb (II) ions. Fig. 4(c) shows the variation in Pb (II) removal with different contact times ranging from 10-50 min using 0.3 g/L of bioflocculant and BFFeNPs dosage at pH 6 and the temperature was fixed at 25 °C. The results for the effect of contact time on Pb (II) removal are represented in Fig. 4(c). It was observed that removal efficiency increased initially after that it became steady, denoting the attainment of equilibrium [51]. The reason for the short time equilibrium is the entire active sites of the bioflocculant and BFFeNPs were vacant at the beginning and the binding affinity of  Pb (II) will increase towards these active sites [12]. At the equilibrium point, the highest Pb (II) removal efficiency of about 75.50% with BFFeNPs and 68.78 %with bioflocculant were obtained at 30 min of contact time.
Fig. 4(d) shows the effect of temperature on Pb (II) removal efficiency. The relationship between temperature and removal efficiency of BFFeNPs was examined at a temperature range from 25-50° C. As the temperature increased, the removal percentage of Pb (II) ion was also increased till it reached the optimum level. The highest removal efficiency of 76.14% was achieved at 35°C which was in good agreement with the result reported by Nharingo et al. [34].

Process optimization for maximum removal of lead (II)
Response surface methods were used to determine the effect of different variables such as pH (2, 4, 6, 8 and 10), BFFeNPs dosage (0.1, 0.2 and 0.3 g/L), contact time (10, 20, 30, 40 and 50 min) and temperature (10, 20, 30, 40 and 50 ºC) on Pb (II) removal percentage. A quadratic model of three factorial BBD was selected for the optimization of variables. A total of 29 runs were conducted to get the best levels of variables. Four parameters (pH, BFFeNPs dosage, contact time, and temperature) were taken as independent variables and Pb (II) removal percentage was taken as a response to the study. The responses of predicted and experimental values were computed using Analysis of variance (ANOVA) to check whether the polynomial expression is able to predict the responses statistically (Table 3). The ANOVA for the quadratic model of response (Pb (II) removal) was tabulated in Table  3.
P-value (< 0.0001) obtained revealed the regression model is significant. F value (4780.74) obtained further implies that the model is highly significant for the removal of Pb (II). There is only a 0.01% chance that a large F value could occur due to noise. Values of ‘‘Prob>F’’ less than 0.0500 indicate that model terms are significant [52]. In this case, A, B, C, D, AB, BC, BD, CD, A2, B2, C2, D2 are the significant model terms. Lack of fit F value-0.0446 implies a non-significant lack of fit. The goodness of the model was restricted by R2 (coefficient of determination). The high R2 value (0.9998) obtained is very close to 1.0 and advocates a great correlation between the predicted and actual values. The value of Predicted R2 (0.9996) and Adjusted R2 (0.9996) are in reasonable agreement. The regression model selected provides a clear explanation of the relationship between the variables and response. The relationship between the selected variables and response (Pb II removal) for the quadratic model was demonstrated by the second-order polynomial equation:  

Response: Pb II removal (YPb II) = + 91.46 + 0.8783 × A + 1.81 × B + 0.2833 × C – 0.8017 × D – 2.20 × AB – 0.1400 × AC – 2.41 × AD + 2.21 × BC + 2.76 × BD – 0.6950 × CD – 25.57 × A2 – 11.94 × B2 – 10.86 × C2 –13.34 × D2.

Where (YPb II) represents Pb (II) removal (%) as response and A, B, C  and D were coded terms for the four test variables i.e., pH, BFFeNPs dosage, contact time, and temperature respectively. The 3-dimensional (3D) and 2-dimensional (2D) contour plots showed a significant influence on the response using BFFeNPs. The 3D and 2D contour plots are helpful to investigate the interactive effect of two variables on Pb(II) removal percent in the experimental ranges as given in Figs. 5, 6, and 7.
3D and 2D Contour plots were shown in Fig. 5(a) illustrating the interactive effect of pH and BFFeNPs dosage on Pb (II) removal (%). As indicated in the figures., after optimum level Pb (II) removal percent was increased with an increase in the pH of the wastewater. Maximum Pb II removal percentage was achieved at pH 6 with a Pb (II) removal efficiency of 91.45%. Similar results regarding optimum pH of Pb (II) removal by fungal biomass were reported by Amin et al.[45]. According to Nharingo et al. at acidic pH, there is competition between hydrogen (H+) ions and the metal ions for the negatively charged sites on the anionic BFFeNPs. The minimum Pb (II) percentage removal was observed at acidic pH (1-5) due to repulsion between protonated BFFeNPs and the positive Pb (II) ions. The diminishing levels of H+ ions in solution, as pH was increased, gave rise to Pb (II) ions being the central cationic species in solution; hence, its percentage removal increased with pH. As pH increased the enhancement of metal flocculation was observed this is due to the high electrostatic attraction between negatively charged sites of BFFeNPs and the positively charged Pb (II)  ions [34]. At alkaline pH (7-10), the Pb (II) removal percentage remained constant. A similar trend of results was illustrated by Dlamini et al. [17]. At higher pH than the optimum level, flocculation may be hindered by the formation of hydroxo species of metal ion that do not bind to the BFFeNPs. Furthermore, the availability of the positively charged and negatively charged site fluctuates with pH, with the latter commonly decreased with increasing pH than the optimal level, thus resulting in a lower metal removal at higher pH through bioflocculation [33].
While in the case of BFFeNPs dosage, a gradual increase in Pb (II) removal (%) with an increase in BFFeNPs dosage was observed. An increase in dosage increased to negatively charged sites available for Pb (II) ion aggregation. A similar trend was observed for the biocoagulation-flocculation of Pb (II) ion from wastewater using cactus Opuntia ficus indica [34]. After the optimum level, a further increase in dosage causes the re-spreading of aggregated particles and also disturbs particle settling. Above the optimum dosage of 0.2 g/L, the trends showed a decrease in removal percentage due to some steric repulsion attributed to particle stabilization. Thus, removal efficiency decreases with increasing dosage beyond the optimum [34]. The BFFeNPs dosage of 0.2 g/L brought about complete charge neutralization of the Pb (II) ion and was therefore identified as the optimum dosage for the removal of Pb (II) ions from wastewater.
The interactive effect of pH and contact time on Pb (II) removal percentage was shown by 3D surface graphs and 2D contour plots Fig. 5(b). As contact time increases, the Pb (II) removal (%) also increases up to the optimum level. The highest Pb (II) removal (%) was observed at 30 min of contact time, after which no significant increase in removal percent was observed.
Fig. 5(c) shows the interactive effect of pH and temperature. Maximum Pb (II) removal was obtained at pH 6 and temperature 30°C. Coagulation-flocculation is affected by higher temperatures. Temperature beyond the optimum level results in rapid movement of particles that disturb their agglomeration; hence minimum removal of metal ions occurs [53]. 
As seen in Fig. 6(a), it is evident from the graphs that both the variables (BFFeNPs dosage and contact time) strongly influenced the Pb (II) removal response. Maximum Pb(II) removal was achieved at 0.2 g/L of BFFeNPs dosage and contact time 30 min. After the optimum level of dosage and contact time, Pb (II) removal percentages were slower because the active sites of BFFeNPs and bioflocculant are occupied by Pb (II) ions. Similar findings for Pb II removal were reported by Celebi& Gök [51].
BFFeNPs dosage and temperature also influenced the removal of Pb (II) ions. Fig. 6(b) depicts the interactive effect of BFFeNPs dosage and temperature. Highest Pb (II) removal percentage was observed at 0.2 g/L of BFFeN dosage and 30°C of temperature. Fig. 6(c) shows the effect of contact time and temperature on Pb (II) removal (%).
The expected values of Pb (II) removal percentage were calculated by regression analysis and correlated to the actual data which demonstrated that the experimental response values were in good agreement with the expected (predicted) response values (Table 4).
The normal plot for residuals and predicted vs actual plots of the Pb (II) removal (%) (response), were presented in Fig. 7 (a) and Fig. 7 (b) respectively.
A statistical design was confirmed by implementing a point prediction tool of RSM from an optimal value of all the four factors A, B, C, and D which were used for the experiment. The optimized factors of the Pb (II) removal percentage attained were A: pH (6), B: BFFeNPs dosage (0.2 g/L), C: contact time (30 min), and D: temperature (30°C). The experimental Pb (II) removal percentage (91.45±0.49%) was in good agreement with the predicted value (91.46±0.1%) demonstrating the rationality of the design. Thus, a significant increase in the Pb (II) removal percentage from 76.14 to 91.45% was attained using BFFeNPs under optimized conditions.
A comparison was carried out of the adsorption efficiency for Pb (II) ions removal by other adsorbents with current synthesized BFFeNPs nanoparticles and is shown in Table 5 [12, 52, 53, 54]. The adsorption efficiency within a short contact time and low BFFeNPs dosage of this study revealed that bioflocculant mediated iron nanoparticles (BFFeNPs) could be a promising adsorbent for Pb (II) ions removal.

Adsorbent recovery 
Regeneration of adsorbent is one of the key steps in the adsorbent economy for further uses to keep the process cost-effective. In the first step of the desorption process for Pb (II), the metal ions adsorbed iron nanoparticles were washed with distilled water to remove the unabsorbed Pb (II) ions which were loosely attached to the BFFeNPs adsorbent. Then, the BFFeNPs adsorbent was separated by centrifugation (at 6000 rpm for 5 min) and was shaken with 50 ml of 0.2M HCl. Thereafter, the remained solution was analyzed utilizing the atomic absorption spectrophotometer and the separated sorbents were washed with distilled water. Finally, the recovered BFFeNPs adsorbent was dried at 120 °C for 60 min. [57].
The metal recovery was calculated by the following equation :

Recovery (%) =  × 100

In the present study, BFFeNPs adsorbents desorbed ∼89% of the adsorbed Pb (II). Fig. 8. This showed that there was no significant change in desorption efficiencies, even after five cycles. These results showed that BFFeNPs is potential adsorbent with high adsorption and desorption capabilities. 

Mechanism of lead (II) adsorption
The proposed mechanism for Pb (II) ion adsorption onto BFFeNPs is illustrated in Fig. 9. It can be concluded that BFFeNPs removes Pb (II) ion by the following strategies 1) Pb (II) exists in wastewater as cations that can be removed via electrostatic force of attraction 2) Pb (II) interact more specifically with the functional groups on the surface of BFFeNPs [58]. 
The presence of functional groups such as the COOH group of conjugated carbonyls (mainly ketones and esters) and polar groups on the BFFeNPs surface that was shown by the FTIR analysis is responsible for the chemical adsorption of Pb (II) ion [59]. There are a large number of −OH functional groups on BFFeNPs. At low pH conditions, the number of H+ ions in the wastewater sample increases, and −OH groups become positively charged −OH2+, decreasing the adsorption capability of Pb(II) ions on the surface of the BFFeNPs adsorbent. At pH values higher than 6, there will be a decrease in adsorption of Pb (II) ion due to precipitation of Pb (II) in the form of Pb(OH)2 [60]. 

CONCLUSIONS
In this study, bioflocculant mediated iron nanoparticles (BFFeNPs) was utilized for the successful removal of Pb (II) ion from the wastewater within a short contact time. Continuous adsorption in the range of 200-350 nm in UV-Visible spectra indicated the existence of iron nanoparticles. The BFFeNPs was synthesized and characterized by different instrumental analysis (UV, XRD, FTIR, SEM, EDX, TEM, DLS, and AFM). SEM and TEM analysis exhibited that the synthesized Fe nanoparticles have a spherical shape. DLS analysis revealed an average particle diameter of 52 nm. In addition, response surface methodology was adopted for statistical optimization using BBD to enhance the Pb (II) removal efficiency. Optimum values obtained were pH 6.0, BFFeNPs dosage 0.2 g/L, contact time 30 min and temperature30°C. Flocculation activity was increased up to 91% after process optimization. This result showed that bioflocculant mediated iron nanoparticles (BFFeNPs) could be utilized as an efficient, cost-effective flocculant for bioremediation of Pb (II) ions from wastewater.

ACKNOWLEDGEMENTS 
This study was funded by University Grants Commission, India, grant (800/(CSIR-UGC NET DEC. 2017), New Delhi, India Authors are thankful to the School of Advanced Sciences (SAS), School of Bio-Medical Sciences (SBST), and DST-FIST/VIT SEM of Vellore Institute of Technology, India for providing laboratory facilities while conducting the experimental work.

CONFLICT OF INTEREST
The authors declare no conflict of interest.