Treatment of Fly-Ash-Contaminated Wastewater Loaded with Heavy Metals by Using Fly-Ash-Synthesized Iron Oxide Nanoparticles

: Every year, a huge amount of water is polluted by various sources, out of which coal ﬂy ash (CFA) is one of the major pollutants. CFA has a large number of toxic metals, which reaches water bodies by coming in contact with water or rain. Due to heavy-metal contamination, water becomes unﬁt for drinking for human beings, which in long term may cause several disorders. Thus, iron oxide nanoparticles (IONPs) recovered from waste, such as CFA, could be the most promising material for treating wastewater, due to their low-cost, recyclable nature and magnetic property. The synthesis of IONPs from CFA involves three sequential steps. The ﬁrst step involves extraction of ferrous materials from CFA, followed by acidic treatment of ferrous materials to obtain acidic leachate, and lastly the precipitation of iron oxides by an alkali. The particle size of the synthesized IONPs varied from 30–70 nm and purity was about 90–93%, as conﬁrmed by transmission electron microscope (TEM) and electron diffraction spectroscopy (EDS). Further, the synthesized IONPs were used for the remediation of various heavy metals, especially Pb and Cr ions from 20% CFA aqueous solutions. The heavy-metal removal efﬁciency of IONPs varied from 40–70%. The developed method suggests heavy-metal removal from wastewater by using an economical and greener route.


Introduction
Every year, various potable water sources are polluted due to various anthropological activities. Industries discharge dyes, heavy metals, and other pollutants into drinkable water bodies, and make them unfit for drinking [1]. Heavy metal is one such inorganic pollutant from industrial activities and several byproducts, that causes toxicity in living beings.
In the present research work, coal fly ash was collected from the Gandhinagar thermal power plant. Further, ferrous fractions were isolated by a wet slurry method. Here, firstly, ferrous fractions were collected from the CFA slurry by using a strong neodymium magnet. This was later treated with strong mineral acid under sonication, to obtain ferrous leachate. Further, this ferrous leachate was used as a precursor material for the synthesis of highly pure IONPs by the co-precipitation method. The synthesized IONPs were then analyzed by sophisticated instruments for their detailed information, along with their purity. The objective was to use CFA as an alternative and economical source of ferrous materials. Another objective was to synthesize pure iron oxides from the extracted ferrous fraction, by using a chemical approach, such as the sonochemical and co-precipitation methods. Another objective was to assess the potential of the synthesized IONPs for the removal of heavy metals from wastewater. The synthesis, as well as applications of IONPs, as nano adsorbents, not only suggest a suitable substitute, but at the same, also minimizes pollution in the form of solid waste arising due to CFA. The synthesis of IONPs from CFA and their application as nano-adsorbents for the remediation of heavy metals or wastewater treatment, makes the process green and economical.

Synthesis of Iron Oxide Nanoparticles from Fly Ash
Firstly, CFA was added to distilled water in a plastic beaker to obtain a slurry. By using a strong external neodymium magnet, ferrous particles were extracted from the slurry, as shown in CFA (Figure 1a). The extracted ferrous particles ( Figure 1b) were dried in an oven at 60-70 • C. Further, a weighed amount of dried ferrous particles was added to 37% HCl in a round bottom flaks, keeping a solid-to-liquid ratio of 1:5. Further, the mixture was subjected to sonication by placement in an ultrasonicator (40 KHz), at 60-70 • C, in a round bottom flask (Figure 1c). Once the experiment was over, the reaction mixture was allowed to cool at room temperature. In order to obtain the acidic solution of ferrous material, the mixture was centrifuged at 5000 rpm for 5 min. Further, the acidic leachate was filtered by using Whatman Filter paper 42. About 20 mL of ferrous leachate was taken in a 100 mL RB, and there was continuous addition of 8M NaOH with continuous stirring at 200-300 rpm, along with heating at 60-70 • C (Figure 1d). The NaOH addition was stopped when a brown-to-black color precipitate started to appear, and at pH 11-13, and the reaction was carried out for 60-90 min (Figure 1e). When the mixture was cooled, the precipitate was washed several times with distilled water and ethanol, to remove any impurities. Finally, the obtained iron oxide particles were oven-dried at 40-50 • C and analyzed by various techniques, as shown in Figure 1f.

Preparation of 20% Fly Ash Aqueous Solution
The 20% aqueous solution of CFA was prepared by adding 200 g of CFA to one liter of double-distilled water (ddw) in a plastic container. The plastic container was sealed with paraffin and kept for shaking in a horizontal shaker at 150 rpm, at 25-26 • C. The shaking was carried out for 24 h. After that, the mixture was left undisturbed for 12 h, so that the solid CFA particles could settle down. When the solution appeared clean, leachate was collected [20,21]. After that, a few drops of nitric acid were added and kept in a refrigerator at 4 • C, to prevent any additional chemical reactions.

Preparation of 20% Fly Ash Aqueous Solution
The 20% aqueous solution of CFA was prepared by adding 200 g of CFA to one liter of double-distilled water (ddw) in a plastic container. The plastic container was sealed with paraffin and kept for shaking in a horizontal shaker at 150 rpm, at 25-26 °C. The shaking was carried out for 24 h. After that, the mixture was left undisturbed for 12 h, so that the solid CFA particles could settle down. When the solution appeared clean, leachate was collected [20,21]. After that, a few drops of nitric acid were added and kept in a refrigerator at 4 °C, to prevent any additional chemical reactions.

Characterization of Iron Oxide Nanoparticles
The UV-Vis measurement was done by dispersing the IONPs in deionized water, followed by sonication for 10 min in the range of 200-800 nm by using a UV-1800 Double Beam spectrophotometer, (Shimadzu, Japan), operated at a resolution of 1 nm. The FTIR spectra were obtained by preparing (IONPs (2mg)+ KBr (198 mg)) the KBr-based solid pellet, using the Spectra SP65, (Perkin Elmer, Waltham, USA) instrument. The FTIR measurement was done in the transmittance mode, in the mid-IR region of 400-4000 cm −1 , at a resolution of 2 cm −1 . The Raman spectroscopy measurement, in the region of 200-4000 cm −1 , was done by placing the sample on the glass coverslips and passing a beam of a laser at 560 nm and 650 nm for 5-10 s, with the help of a Witec Alpha300S+ SNOM Raman Module (Witec, Ulm, Germany) instrument. The particle size distribution of the synthesized IONPs was done by using a particle size analyzer (PSA) Nano S90 (Malvern

Characterization of Iron Oxide Nanoparticles
The UV-Vis measurement was done by dispersing the IONPs in deionized water, followed by sonication for 10 min in the range of 200-800 nm by using a UV-1800 Double Beam spectrophotometer, (Shimadzu, Japan), operated at a resolution of 1 nm. The FTIR spectra were obtained by preparing (IONPs (2mg) + KBr (198 mg)) the KBr-based solid pellet, using the Spectra SP65, (Perkin Elmer, Waltham, USA) instrument. The FTIR measurement was done in the transmittance mode, in the mid-IR region of 400-4000 cm −1 , at a resolution of 2 cm −1 . The Raman spectroscopy measurement, in the region of 200-4000 cm −1 , was done by placing the sample on the glass coverslips and passing a beam of a laser at 560 nm and 650 nm for 5-10 s, with the help of a Witec Alpha300S+ SNOM Raman Module (Witec, Ulm, Germany) instrument. The particle size distribution of the synthesized IONPs was done by using a particle size analyzer (PSA) Nano S90 (Malvern Zetasizer, Malvern WR14 1XZ, UK), at a temperature of 25 • C, by dispersing samples in the distilled water, followed by sonication for 10 min. The XRD patterns of IONPs were recorded using a D-8 Advance Bruker (Germany) instrument equipped with an X'celerometer, in the 2-theta range of 20-70, with a step size of 0.02 and a time of 5 s per step at, 40 kV voltage and a current of 30 mA. The surface morphology, purity, and elemental composition of the synthesized IONPs were analyzed by the FESEM, Nova, NanoSEM, and FEI-450 FESEM (Netherlands). The synthesized IONPs were loaded onto the carbon tape, which in turn was placed on the aluminum stub holder. The gold coating of the sample was done with 10 min of sputtering, to make the sample conductive. The electron diffraction spectroscopy (EDS) analysis of the IONPs was analyzed by an Oxford EDS analyzer attached to the FESEM, at variable magnifications and 20 kV. Transmission electron microscope (TEM) analysis was done for the internal size and shape determination of the synthesized IONPs. The analysis was done by using the FEI Model Tecnai G2 20 Twin (USA), operated at a voltage of 200 kV. For TEM and HR-TEM measurements, the sample was prepared by a drop-casting technique, where the IONPs were suspended in an aqueous medium and loaded onto carbon-coated copper grids. For obtaining the 3D structure of the synthesized IONPs, atomic force microscopy (AFM) was done. The IONPs dispersed in ddw for UV-Vis were sonicated, and a drop of the sample was loaded on a clean glass cover slip with the help of a micropipette. The cover slip was allowed to dry in the oven at 50-55 • C, till complete dryness. Further, the excess sample was removed by washing the slide, and dried again. Finally, the 3D and 2D measurements of the IONPs were taken by using the Perk-System model: XE-70 (Perkin Elmer, NY, USA) instrument. The magnetic measurements of the synthesized IONPs were analyzed by the physical property measurement system (PPMS) instrument, with model no. PPMS, at 300 k, i.e., room temperature, in the magnetic field of −7000 k to +7000 k. About 30 mg of powdered IONPs were taken, wrapped in paraffin tape and placed in the quartz tube.

Batch Experiment for Remediation of Heavy Metals
The remediation of heavy metals by the adsorption process in a multi-component system (Cd 2+ , Mn 2+ , Zn 2+ , Pb 2+ , Al, Cu 2+ , Co 2+ , Cr 6+ , Cr 4+ and Ni 2+ ) was performed by the shake-flask batch adsorption process. In a typical batch experiment, 50 mg and 100 mg of IONPs were dispersed in 150 mL of 20% w/v CFA solution in a 250 mL Erlenmeyer flask. The sample was shaken thoroughly in an incubator shaker at 150 pm, at 30 • C and at neutral pH (7). An aliquot of~10 mL sample was collected from the flask after every regular interval of time, i.e., 0, 10, 30, 60, 90, 120 minutes, and analyzed by inductively coupled plasma-optical emission spectroscopy (ICP-OES) (Model: 7300 DV, Make: Perkin Elmer, Netherlands), for the detection of heavy-metal concentration. Among all nine heavy metals selected for the removal from the CFA solution, the per cent removal of heavy metals was calculated by the following formula: where, C 0 = initial concentration of heavy metal C t = heavy-metal concentration at a particular time

Results and Discussion
In the present study, CFA-extracted ferrous particles have been used for the synthesis of pure IONPs. The HCl-based acidic treatment causes the dissolution of Fe from ferrous oxides into the HCl solution, and forms mixtures of ferrous chloride and ferric chloride. As Fe, Al, Na and other alkali metals have better solubility in an acidic medium, such as HCl, they also get dissolved into it, which was confirmed by the ICP-OES. The formation of IONPs takes place through the reactions given below. The CFA-extracted ferrospheres have different phases of iron oxides, such as magnetite (λ-Fe 3 O 4 ), maghemite (β-Fe 2 O 3 ), and hematite (α-Fe 2 O 3 ). Therefore, during the exposure of HCl to such ferrospheres, there is the dissolution of Fe from such iron oxides into the acidic medium, and the formation of ferric chloride and ferrous chloride according to reactions [2,3]. Moreover, Si, being insoluble in acid, remained in the ferrospheres. The ferrospheres were ruptured, along with the formation of the grooves and pits on their surface due to the Fe leaching into the medium, as confirmed by FESEM. Further, according to reaction [4], the NaOH precipitates these chlorides formed in the reactions (2&3) [23][24][25]. These synthesized IONPs were analyzed after drying for confirmation and purity by sophisticated instruments.

Reactions
Fe

Characterization
The preliminary characterization of synthesized nanoparticles by UV-Vis spectroscopy has proven to be a very useful technique for the analysis of nanoparticles [26].
The synthesized IONPs were dispersed in ddw water, followed by sonication for 10 min by an ultrasonicator (Sonar, 40 KHz). The well-dispersed sample was divided into two parts: one part was analyzed by UV-Vis, while the second part was analyzed by the PSA at 25 • C. The UV-Vis measurement peak at 210 nm was the characteristic feature of the presence of Fe 2+ ions and was a preliminary indication of the formation of IONPs, but did not distinguish among the phases of the iron oxides. Figure 2b exhibits the particle size distribution of IONPs, where the average particle size is 381.2 nm, with a polydispersity index (PDI) of 0.494. The PDI and size were in good agreement with size [27,28]. medium, as confirmed by FESEM. Further, according to reaction [4], the NaOH precipitates these chlorides formed in the reactions (2&3) [23][24][25]. These synthesized IONPs were analyzed after drying for confirmation and purity by sophisticated instruments. Reactions

Characterization
The preliminary characterization of synthesized nanoparticles by UV-Vis spectroscopy has proven to be a very useful technique for the analysis of nanoparticles [26].
The synthesized IONPs were dispersed in ddw water, followed by sonication for 10 min by an ultrasonicator (Sonar, 40 KHz). The well-dispersed sample was divided into two parts: one part was analyzed by UV-Vis, while the second part was analyzed by the PSA at 25 °C. The UV-Vis measurement peak at 210 nm was the characteristic feature of the presence of Fe 2+ ions and was a preliminary indication of the formation of IONPs, but did not distinguish among the phases of the iron oxides. Figure 2b exhibits the particle size distribution of IONPs, where the average particle size is 381.2 nm, with a polydispersity index (PDI) of 0.494. The PDI and size were in good agreement with size [27,28].    Figure 3a exhibits the FTIR spectrum of the as-synthesized and after-use IONPs. The as-synthesized IR spectra of IONPs exhibit characteristic bands of iron oxides [29,30]. The band at 422 cm −1 and 576 cm −1 is attributed to the vibrational modes of Fe-/Fe-O-Fe (magnetite and maghemite) of IONPs. The band at 3400 cm −1 and 1600 cm −1 indicates the presence of OH groups of either water molecules or ferric hydroxides (Fe-O-OH) [31]. Previously, Gupta and their team and Yadav et al. also obtained bands in the same region for the synthesized IONPs from ISA, and the results were in close agreement [14,32].
A typical XRD diffractogram is shown for the as-synthesized IONPs and 100 mg residue in Figure 3b. The as-synthesized IONPs show small intensity peaks at 33 • and 35 • , which are due to the hematite and magnetite, respectively. The XRD confirms the mixture of hematite and magnetite phases in the synthesized IONPs. The mixed phases could be because of either partial transformation of ferrous chlorides into a single phase during the synthesis. A team led by Yadav [14] obtained similar results for the synthesized IONPs from ISA. They also had a peak in the range of 33-36 • , attributed to the magnetite or maghemite [14].
(magnetite and maghemite) of IONPs. The band at 3400 cm −1 and 1600 cm −1 indicates the presence of OH groups of either water molecules or ferric hydroxides (Fe-O-OH) [31]. Previously, Gupta and their team and Yadav et al. also obtained bands in the same region for the synthesized IONPs from ISA, and the results were in close agreement [14,32]. A typical XRD diffractogram is shown for the as-synthesized IONPs and 100 mg residue in Figure 3b. The as-synthesized IONPs show small intensity peaks at 33° and 35°, which are due to the hematite and magnetite, respectively. The XRD confirms the mixture of hematite and magnetite phases in the synthesized IONPs. The mixed phases could be because of either partial transformation of ferrous chlorides into a single phase during the synthesis. A team led by Yadav [14] obtained similar results for the synthesized IONPs from ISA. They also had a peak in the range of 33-36°, attributed to the magnetite or maghemite [14].
A typical Raman spectrum of IONPs and residue of 100 mg is shown in Figure 4. It was applied to determine the nature of the IONPs. The bands at 293 cm −1 , 481 cm −1 and 633 cm −1 are attributed to the Fe-O-Fe bonds in the as-synthesized IONPs. Similar observations were also reported by Yadav and their team in 2020 [33], and in the year 2017 by a team led by Justin [29], which determined that IONPs are excited by the 560 nm wavelength laser excitation. These bands confirm the synthesis of crystalline IONPs from the CFA-extracted ferrospheres. A typical Raman spectrum of IONPs and residue of 100 mg is shown in Figure 4. It was applied to determine the nature of the IONPs. The bands at 293 cm −1 , 481 cm −1 and 633 cm −1 are attributed to the Fe-O-Fe bonds in the as-synthesized IONPs. Similar observations were also reported by Yadav and their team in 2020 [33], and in the year 2017 by a team led by Justin [29], which determined that IONPs are excited by the 560 nm wavelength laser excitation. These bands confirm the synthesis of crystalline IONPs from the CFA-extracted ferrospheres.  The TEM micrographs in Figure 5a,b reveals that the particle size varied from 10 to 30 nm, and the particles were showing aggregation. The particles were spherical in shape, along with a few rod-shaped particles. On the other hand, Figure 5 c,d shows HRTEM images of IONPs, which are mainly aggregated and spherical-shaped. Figure 5e shows the lattice fringes of IONPs, where the d-spacing of the IONPs is 0.24 nm. Figure 5f shows the scattering area diffraction pattern (SAED) pattern of the synthesized IONPs. The SAED pattern shows the polycrystalline nature of the IONPs. In 2018, Yadav and Fulekar [18], and a team led by Yadav in 2021 [14], showed similarly shaped IONPs synthesized The TEM micrographs in Figure 5a,b reveals that the particle size varied from 10 to 30 nm, and the particles were showing aggregation. The particles were spherical in shape, along with a few rod-shaped particles. On the other hand, Figure 5c,d shows HRTEM images of IONPs, which are mainly aggregated and spherical-shaped. Figure 5e shows the lattice fringes of IONPs, where the d-spacing of the IONPs is 0.24 nm. Figure 5f shows the scattering area diffraction pattern (SAED) pattern of the synthesized IONPs. The SAED pattern shows the polycrystalline nature of the IONPs. In 2018, Yadav and Fulekar [18], and a team led by Yadav in 2021 [14], showed similarly shaped IONPs synthesized from ferrous fractions, collected from iron precursors and incense sticks ash, respectively. The synthesized IONPs were amorphous in nature in both cases. The EDS spectra (Figure 5g) also show peaks for Fe and O only, which indicates the purity of the synthesized ferrous carbonate and Cu peaks, which are due to the copper grids. The AFM image (Figure 5h) reveals the 3D images of the synthesized IONPs, which was also justified by the FESEM and TEM images [14]. The EDS spectra (Figure 5g) also show peaks for Fe and O only, which indicates the purity of the synthesized ferrous carbonate and Cu peaks, which are due to the copper grids. The AFM image (Figure 5h) reveals the 3D images of the synthesized IONPs, which was also justified by the FESEM and TEM images [14].
While PPMS spectra ( Figure 6) reveal the magnetic strength of the IONPs. The magnetic moment was calculated against the magnetic field [34,35]. During the year 2020, a team led by Yadav also showed similar magnetic properties of the polyhedral-shaped IONPs from ISA [14]. While PPMS spectra ( Figure 6) reveal the magnetic strength of the IONPs. The magnetic moment was calculated against the magnetic field [34,35]. During the year 2020, a team led by Yadav also showed similar magnetic properties of the polyhedral-shaped IONPs from ISA [14]. The FESEM micrographs in Figure 7a,b show that the synthesized IONPs are generally spherical in shape, but they are aggregated to form lumps. In Figure 7b, an aggregated lump of IONPs is shown to have an aggregation of numerous rod-shaped particles inside a large structure. In addition to this, there are several spherical-shaped, white-color dep ositions of small particles on their surface. Figure 7d shows aggregations of numerous spherical-shaped particles at a 200 nm scale. The size of spherical particles varied from 40 to 70 nm, while rod-shaped particles were more than 70 nm in length and 10-20 nm in width. The EDS spectra in Figure 8a have peaks for Fe, O and C, where Fe was 63% and O was 31%, which indicates the high purity of the synthesized IONPs [33]. While carbon is present at up to 4%, it is the only impurity which may be associated with the ferrous particles, or due to the carbon tape used in the stub. The total Fe and O content was 94% which indicates all the trace elements have been removed during the acidic treatment of the ferrous material. In 2021, a team led by Yadav also obtained similar results with IONPs synthesized from incense sticks ash (ISA). The FESEM micrographs in Figure 7a,b show that the synthesized IONPs are generally spherical in shape, but they are aggregated to form lumps. In Figure 7b, an aggregated lump of IONPs is shown to have an aggregation of numerous rod-shaped particles inside a large structure. In addition to this, there are several spherical-shaped, white-color depositions of small particles on their surface. Figure 7d shows aggregations of numerous sphericalshaped particles at a 200 nm scale. The size of spherical particles varied from 40 to 70 nm, while rod-shaped particles were more than 70 nm in length and 10-20 nm in width. The EDS spectra in Figure 8a have peaks for Fe, O and C, where Fe was 63% and O was 31%, which indicates the high purity of the synthesized IONPs [33]. While carbon is present at up to 4%, it is the only impurity which may be associated with the ferrous particles, or due to the carbon tape used in the stub. The total Fe and O content was 94%, which indicates all the trace elements have been removed during the acidic treatment of the ferrous material. In 2021, a team led by Yadav also obtained similar results with IONPs synthesized from incense sticks ash (ISA).

Mineralogical Changes: XRD, FTIR and Raman
FTIR analysis of residue IONPs was done to determine the functional groups involved in the remediation of heavy metals. The FTIR spectra of IONPs, 50 and 100 mg residue in Figure 3a, reveal that there are no significant changes in the bands of IONPs and 100 mg residue, as all the bands remain more or less the same. The bands in the region of 400-700 cm −1 , in all the IONPs, are attributed to Fe-O or Fe-O-Fe (magnetite and maghemite) bonds. After heavy-metal removal in 50 mg IONPs, some differences in the locations of the absorbance peaks were observed. There was a formation of sharp and

Mineralogical Changes: XRD, FTIR and Raman
FTIR analysis of residue IONPs was done to determine the functional groups involved in the remediation of heavy metals. The FTIR spectra of IONPs, 50 and 100 mg residue in Figure 3a, reveal that there are no significant changes in the bands of IONPs and 100 mg residue, as all the bands remain more or less the same. The bands in the region of 400-700 cm −1 , in all the IONPs, are attributed to Fe-O or Fe-O-Fe (magnetite and maghemite) bonds. After heavy-metal removal in 50 mg IONPs, some differences in the locations of the absorbance peaks were observed. There was a formation of sharp and prominent new bands in the 50 mg residue at 1127, 1398 and 1625 cm −1 . These bands could be attributed to the carbonates of Ca and Mg in the sample, which is also evident in the EDS data. The band at 1127 cm −1 could also be attributed to silicate adsorption on the IONPs, after use of the aqueous solution [36].
This could be also due to the formation of hydro complexes (FeOH, Fe-OH-Fe) of metal hydroxides, which were previously not present in the sample. While the band at 1622 cm −1 may be due to the bending vibrations of water molecules H-O-H and Fe-OH complexes, which became sharper and more intense in the 50 mg residue after the adsorption of heavy metals. The band in 50 mg residue at 3400 cm −1 was attributed to water molecules or ferrous hydroxides, having become broader than the IONPs and 100 mg residue. Initially, the peaks for carbonates and silicates were not present in the IONPs, which were adsorbed from the CFA aqueous solution on the surface of IONPs after the reaction, also supported by the EDS data [37,38]. It is assumed that the formation of new absorption bands, the change in absorption intensity and the shift in the wavenumber of functional groups might be due to the complexation between metal ions and the binding sites of the nano-adsorbents. The binding mechanism involved electron pair-sharing between electron donor atoms (O) and metal ions [39]. The XRD pattern in Figure 3b reveals that there are no major changes in the mineralogy of the IONPs before and after the remediation of heavy metals. The peaks at 33 • and 35 • remains as such, which are the characteristic peaks of iron oxide. Additionally, there is one prominent and sharp peak at 37 • , which could be due to the CaO adsorption on the surface of 100 mg IONPs, which is also further supported by EDS and FTIR data [40,41].
The Raman spectra in Figure 4 reveal that there is a shift in the peaks of IONPs before and after used IONPs. There are changes in the peaks, i.e., the peak is shifted to a lower wavenumber in the residue. The bands of as-synthesized IONPs from 293 cm −1 and 674 cm −1 shifted to 287 cm −1 and 633 cm −1 , respectively. There was also a decrease in the intensity of the bands, while the sharp and major band at 481 cm −1 completely disappeared in the residue. This could be because of the oxidation of the sample, or because of the deposition of elements from the aqueous solutions, which may respond to a weak signal. There are also changes in the crystallinity of the 100 mg residual sample, where its crystallinity was affected after being used as an adsorbent. This could be due to the adsorption of heavy metals and carbonates, and the silicates on them [42,43].

Morphological and Elemental Changes: FESEM-EDS and ICP-OES
From the EDS spectra, it is revealed that there is adsorption of Si, Al 3+ , P, and Mg 2+ , in both the samples of IONPs, i.e., in 50 mg and 100 mg residues. However, 50 mg IONPs have adsorbed more elements than the 100 mg IONPs besides heavy metals on their surface, as it has Cl -, Ca 2+ and K + , in addition to the above-mentioned elements [44]. In the 50 mg IONP residue EDS spectra, Cl -, Mg 2+ , and Ca 2+ alone comprise about 68.77%, which indicates the formation and deposition of chlorides on the surface of the 50 mg residue. This is because the fresh aqueous solutions of CFA have higher Ca and Mg, and both have a high solubility in normal room-temperature water. The higher solubility and concentration of Ca 2+ and Mg 2+ in the 20% CFA solution is also supported by the ICP-OES data. This indicates that when the IONP dose was 50 mg, then the adsorption of heavy metals and other elements was more effective [45][46][47].
The FESEM micrographs of IONPs before and after remediation of heavy metals are shown in Figure 8a-d. Figure 8a,b shows the FESEM micrographs of as-synthesized IONPs, while Figure 8e,f are images of 100 mg IONPs after use. The FESEM micrographs reveal that there are no distinguishable morphological changes, i.e., either in their shape or size before and after the use of IONPs. However, the aggregation of the particles remained as such. There was hardly any positive impact of a higher dose of IONPs on the rate of adsorption. The data was also supported by FTIR and ICP-OES. These findings, as related to EDS analysis, showed the involvement of the ion exchange mechanism in the removal of metal ions by the IONPs [48]. About 3 mg of the 100 mg IONPs, after use, was digested in aqua regia at 220 • C, and after digestion, it was filtered and make-up volume was prepared by adding ddw. The sample was analyzed for the detection of trace elements present on its surface after adsorption, as shown in Table 1. The following elements were identified on its surface, which were initially not present in the IONP sample: Ca 2+ , Al 3+ , Si, P, K + , Li 3+ , Mg 2+ , Na + , Cu 2+ , Cd 2+ , Co 2+ , Cr 6+ , Cr 4+ Pb 2+ , Mn 3+ , and Zn 2+ , have adsorbed on the IONP surface, which confirms the adsorption of heavy metals. All these were not present initially in the original sample, i.e., IONPs, which indicates that the source of these elements was the aqueous solutions of CFA. All these elements were adsorbed on the surface of IONPs during the reaction. Among nonheavy metals, the maximum adsorption was of Al 3+ , Mg 2+ , Ca 2+ and Na + , as these elements were present in a large concentration in the CFA [49].

The Remediation of Heavy Metals by 50 mg IONPs
The removal efficiency of all the nine heavy metals [Al 3+ , Cu 2+ , Cd 2+ , Co 2+ , Cr 6+ , Cr 4+ , Pb 2+ , Mn 3+ , Ni 2+ and Zn 2+ ] continuously increased, in Figure 9a from 10 min up to 60 min after which there was a decrease in the removal efficiency for all the heavy metals at 90 min. A further decrease was noticed at 120 min in all the heavy metals except for Co and Zn. This indicates that the removal efficiency was maximum at 60 min for all the heavy metals, where they reached equilibrium and after which desorption occurred on the surface of IONPs. This desorption process continued until 120 min except for Co 2+ and Zn 2+ . While the Pb 2+ was not detected after 10 min, in the solution and remained as such for up to 120 min. This was because Pb 2+ was present in a much lesser concentration than the other heavy metals in the solution. Therefore, within 10 min only its value reached below the sensitivity level of ICP-OES. The desorption is due to the non-availability of free adsorption sites on the IONP surface [50,51].

The Remediation of Heavy Metals by 100 mg IONPs
The removal efficiency of Ni 2+ , Cr 6+ and Cr 4+ remained constant for up to 30 min, and a sudden increase was noticed at 60 min, after which it remained as such until 90 min, as shown in Figure 9b. A drastic decrease in the removal efficiency was seen in Ni 2+ at 120 min, while the Cr 6+ and Cr 4+ reached saturation after 90 min to 120 min. In the case of Cu 2+ , Co 2+ , Zn 2+ , Al 3+ and Mn 3+ , there was an initial decrease in the efficiency from 10 min to 30 min, after which an increase was noticed in all of them at 60 min. After this, a marginal decrease was noticed at 90 min, which continued to decrease until 120 min. The Cd 2+ value continuously decreased from 10 min and continued up to 90 min, and a minor increase was noticed at 120 min. Likewise, at 50 mg, the Pb 2+ value increased at 10 min, after which the Pb 2+ was not detected up to 120 min, due to the above reason. Initially, the Pb 2+ concentration was less in the solution, and after 10 min, its value reached below the sensitivity level of ICP-OES.
At the higher dose of IONPs, the remediation did not increase drastically, possibly due to agglomeration and precipitation of nano-photocatalyst, which may have increased the size of the IONPs, leading to a reduction in the specific surface area and ultimately lowering the number of surface-active sites [52,53].

The Remediation of Heavy Metals by 100 mg IONPs
The removal efficiency of Ni 2+ , Cr 6+ and Cr 4+ remained constant for up to 30 min, and a sudden increase was noticed at 60 min, after which it remained as such until 90 min, as shown in Figure 9b. A drastic decrease in the removal efficiency was seen in Ni 2+ at 120 min, while the Cr 6+ and Cr 4+ reached saturation after 90 min to 120 min. In the case of Cu 2+ , Co 2+ , Zn 2+ , Al 3+ and Mn 3+ , there was an initial decrease in the efficiency from 10 min to 30 min, after which an increase was noticed in all of them at 60 min. After this, a marginal decrease was noticed at 90 min, which continued to decrease until 120 min. The Cd 2+ value continuously decreased from 10 min and continued up to 90 min, and a minor increase was noticed at 120 min. Likewise, at 50 mg, the Pb 2+ value increased at 10 min, after which the Pb 2+ was not detected up to 120 min, due to the above reason. Initially, the Pb 2+ concentration was less in the solution, and after 10 min, its value reached below the sensitivity level of ICP-OES.
At the higher dose of IONPs, the remediation did not increase drastically, possibly due to agglomeration and precipitation of nano-photocatalyst, which may have increased the size of the IONPs, leading to a reduction in the specific surface area and ultimately lowering the number of surface-active sites [52,53].

Kinetics Study of Pb and Cr
To find out the remediation of Pb 2+ , Cr 6+ and Cr 4+ ions from the CFA 20% simulated wastewater by iron IONPs, three different kinetic models (Pseudo-first order (1), Pseudosecond order (2) and intra-particle diffusion (3)) were applied to perform the kinetic studies for the removal of Pb 2+ and Cr ions by IONPs. Figure 10a-d shows kinetic studies for Pb removal and Figure 10e-h shows kinetic studies for Cr ion removal by IONPs. Tables 2 and 3 show the kinetic parameters of Pb removal and Cr ions removal, respectively. From the above data, it can be summarized that both Pb removal, as well as Cr ions removal, follows a Pseudo-second order reaction.
1 Figure 10. Contact time study (a), Pseudo-first order (b), Pseudo-second order (c) Intra-particle diffusion study of Pb 2+ removal by IONPs (d). Contact time study (e) Pseudo -first order (f) Pseudosecond order (g) Intra-particle diffusion study of Cr 6+ and Cr 4+ ions by IONPs (h). The slandered equation is given below to calculate the kinetics parameters for the suitable kinetic orders: where Q e and Q t are the amounts of metal (Pb, Cr) adsorbed (mg/g) at equilibrium and at time; t (min), respectively, and k 1 (min −1 ) is the rate constant in the pseudo-first order kinetic model.
where k 2 is the rate constant (g/mg. min) in the pseudo-second order kinetic model for adsorption.
where k id is the intra-particle diffusion rate constant (mg/g. min 3/2 ), and the values of C (mg/g) depict the boundary thickness. Table 2 shows the kinetic parameters of the nano-adsorbent for Pb 2+ . The highest R 2 value for the pseudo-second order is closer to 1 compared to others (Figure 10c), having Q e = 1.01 (mg/g) and k 2 = 0.019 × 10 −1 (g/mg. min), which attribute to the pseudo-second order adsorption kinetics followed for Pb 2+ adsorption. However, the same trend (psuedosecond order) adsorption kinetics was obtained for Cr 2+ metal adsorption with the highest R 2 value ≈ 0.993 (Figure 10g), having Q e = 2.19 (mg/g) and k 2 = 0.0101 × 10 −1 (g/mg. min) ( Table 3).

Adsorption Mechanism
Here the adsorption of heavy metals on the surface of IONPs was via electrostatic attraction and coordination. At neutral or slightly acidic pH on the surface of IONPs has a positive charge, which produces PZC repulsion to cations, and the adsorption of cations on the IONPs is due to the coordination between CH on the surface of IONPs and the bivalence metal ions. Such a phenomenon of adsorption and deposition was seen because the CFA aqueous solutions are a multicomponent system, where several heavy metals and non-heavy metals compete for the limited adsorption sites. Once these binding sites are occupied, there is competition among ions and nonmetallic ions, which results in the desorption of the heavy metals. The concentration of Pb 2+ reached below the detection level of ICP-OES within 30 min only, in comparison to the other cations. This could be because Pb 2+ has a higher electronegativity, as electronegativity plays an important role in controlling the adsorption of metal ions. The higher the electronegativity, the stronger the covalent formation of the metals with oxygen atoms located on the surface of IONPs [18].

Conclusions
The present study showed that it is possible to synthesize pure IONPs from CFA, and they could be used as alternative to nano-adsorbents for the removal of heavy metals from wastewater or CFA aqueous solutions. The CFA-based synthesized IONPs can effectively and efficiently remediate the CFA heavy metals from a multi-component CFA solution, and only the remediation time of different heavy metals varies. Whether heavy metals reach equilibrium early or later depends on the intrinsic properties of the respective ions and the conditions for adsorption. After reaching equilibrium, there is desorption of heavy metals from the binding sites of IONPs. This is due to the reason that initially, most of the adsorption sites are occupied by the ions and there are no further free sites available for adsorption. The efficiency of heavy metal removal is better with a lower dose of IONPs than with a higher dose, i.e., 50 mg. When the dose of IONPs is 50 mg, then most of the heavy metals are being remediated more efficiently. This is the case because there is an agglomeration of the IONPs at higher doses. Moreover, at higher doses, there are no vacant adsorption sites. Besides heavy metals, IONPs also remove other elements, such as K + , P, Na + , Be, Ca 2+ , Mg 2+ , etc., that were present on the residue during EDS analysis. They can also act as an adsorbent for the removal of several non-heavy metals, as revealed by the residual analysis by ICP-OES. Therefore, IONPs can effectively remove heavy metals, as well as non-heavy metals, from the solution.