Apatite Ore-based Nanostructures: Novel and Eco-friendly Sorbent for Efficient Removal of Wastewater Containing Pb2+ and Fe3+

Global challenges in removing heavy metal ions from aquatic reservoirs require novel solutions, especially the application of environmentally friendly materials. This paper presented the efficient removal of Fe3+ and Pb2+ ions from wastewater by apatite ore-based nanostructures. The synthesized material exhibited a nanostructure with high thermal stability, high porosity, and negative surface potential, suitable for heavy metal removal in wastewater. The adsorption measurements performed in varying conditions (pH, mass of the adsorbent, and contact time onto the adsorbent) proved that even a few milligrams of the synthesized material could effectively absorb the lead and iron ions from the solution, reaching an effectiveness of about 90%. The maximum adsorption capacity was about 342.2 mg·g−1 for Pb2+ and 1089.59 mg·g−1 for Fe3+, while the adsorption isotherm determined with non-linear models undergoes the pseudo-first-order kinetics and Redlich-Peterson adsorption model. In addition, the thermodynamic parameters reveal the spontaneous process. Experiments conducted with industrial and craft-village wastewaters confirmed the high potential of the nanostructural hydroxyapatite as an efficient and affordable material for the removal of various pollutants from aqueous solutions in practical conditions.


Introduction
Due to growing industrialization, water reservoirs are increasingly and widely polluted. Besides chemical compounds, the most problematic hazards to remove are heavy metal ions. Their appearance in the environment is mainly caused by anthropogenic activities, such as the disposal of batteries or electronic devices (Zhou et al., 2020). This issue is becoming more severe with the use of portable electronics. Heavy metals can be found even in tap water, leading to adverse health effects, including endocrine disruption and cancer (Andrade et al., 2020;Husband & Boxall, 2011). Among heavy metals, Pb 2+ and Fe 3+ are the two most common contaminants found in wastewater in Vietnam (Ibrahim & Abdel-Hameed, 2022;Huong et al., 2010). While lead is often present in wastewater of heavy industries, e.g., plating, tanning, and recycling industry, iron can be found in most old water pipes, including domestic water. Removal of these contaminants from aquatic reservoirs requires immediate and novel solutions.
Given this, our study focuses on the preparation and application of nanostructural hydroxyapatite, one of the most effective materials that can be used for wastewater treatment using cost-effective and widely abundant apatite ores as a natural source of calcium and phosphates ions. The proposed material has a high removal capacity for Pb 2+ and Fe 3+ heavy metal ions, making it suitable to form less soluble mineral phases that are more stable in the environment than the Pb 2+ and Fe 3+ ions. Prior to the treatment of real wastewater, the adsorption studies were performed in the function of contact time, initial Pb 2+ and Fe 3+ concentration, pH, adsorbent mass, and temperature, as well as the kinetics and adsorption mechanism was studied.
The raw apatite ore was initially milled into the powder and dried at 105°C in an oven. Then, the dry mass of about 1 g of apatite ore was treated with 20 mL of 1 M HNO 3 solution and stirred within the magnetic stirring bar at 400 rpm for 30 min at room temperature. Then, the precipitate was separated by filtration, and the obtained solution was used as a source of Ca 2+ and PO 4 3− ions to prepare hydroxyapatite (HAp) with the use of the precipitating agent. The ammonia NH 3(aq) solution was added dropwise until pH 10, maintaining the hydroxyapatite formation instead of the calcium phosphate compounds. Then, the white precipitate was filtered and washed with 1 3 Vol.: (0123456789) distilled water until neutral pH. Next, the powder was dried at 105 °C overnight and had a phase structure of the hydroxyapatite (HAp) with a pH pzc of about 7.49 (Phuong et al., 2020).

Lead (Pb 2+ ) and Iron (III) (Fe 3+ ) Ions Adsorption Experiments
The adsorption experiments were performed using solutions containing Pb 2+ and Fe 3+ ions as follows: 50 mL of Pb(NO 3 ) 2 or FeCl 3 solutions at concentrations ranging from 5 to 400 mg·L −1 was prepared. The adsorption studies were performed using different adsorbent amounts, from 0.001 to 0.17 g (dry basis). Experiments were also conducted in the function of pH ranging from pH 1 to pH 6 adjusted with HCl and NaOH and in the function of time (5 min, 10 min, 20 min, 30 min, 40 min, 60 min, 80 min, 100 min, and 120 min). Pb 2+ and Fe 3+ adsorption efficiency H (%) and capacity (Q (mg·g −1 )) were determined based on Eqs. (1) and (2): where C o (mg·L −1 ) is the initial concentration of Pb 2+ and Fe 3+ ions in the solution, C e (mg·L −1 ) is the heavy metal ions concentration at equilibrium, V (L) is the solution volume (V = 50 mL), and m (g) is the amount of nanostructural hydroxyapatite.

Structural and Adsorption Analysis
The TGA-DTG analysis from room temperature to 800 °C at 5°C·min −1 in the air was performed on a Macro-thermogravimetric system (CIRAD -France) to determine the thermal behaviors of HAp. The surface morphology was studied using scanning electron microscopy (SEM), JSM-6510LV, JEOL Ltd. (Japan), and transmission electron microscopy (TEM) JEM-2100 JEOL Ltd. (Japan).
Before and after adsorption, the element composition of the material was determined by EDX analysis with X-Act (Oxford).
The surface functional groups were detected using an FTIR spectrometer (Nicolet iS10, Thermo Scientific) in a range of 400-4000 cm −1 .
The porous structures of hydroxyapatite were investigated using a Micromeritics 3Flex Adsorption Analyzer. The Brunauer-Emmett-Teller (BET) method was applied to estimate the total surface area and total pore volume of the material, while the micropore's and mesopore's surface area and volume were identified via the t-plot and BJH method, respectively. The surface potential onto HAp was determined using the dynamic light scattering method (DLS, SZ-100V2, Nanoparticle analyzer, Nanopartica, Horiba Scientific (Japan) in the charge range of about −200 to 200 mV at 25 °C).
The crystallinity of the hydroxyapatite before and after the adsorption was characterized by the phase component by X-ray diffraction (XRD), D8 ADVANCE-Bruker, CuK α radiation (λ = 1.54056 Å) with a step angle of 0.03°, the scanning rate of 0.04285° per second, and 2θ degree was measured in the range of 10-70°).

Morphology, Element Composition, and Surface Potential of Hydroxyapatite
The morphology of HAp was determined using scanning electron microscopy (SEM, Fig. 1) and transmission electron microscopy (TEM, Fig. 2). Figure 1 shows nanostructured grains distributed uniformly within the whole volume. The mapping image of different elements on the surface of the hydroxyapatite is measured by the EDX technique. It can be seen that the main surface elements, which are Ca, P, and O, are uniformly distributed on the whole surface of the sample. Based on the TEM analysis ( Fig. 2), it can be observed that HAp grains have a size of about 20 nm. The overlapping of the particles is caused by the drying of the sample from the aqueous solution pushing particles to move close to each other. Despite the aggregation, each individual particle possesses a nanostructural structure.
The surface charge of the nanostructural hydroxyapatite was determined by the zeta potential measurements (Fig. 3). The obtained value of the surface potential is about −34.4 mV for the presence of the hydroxyl groups. A negative value of the surface potential suggests a strong interaction of the material with the heavy metals cations, hence improving the effectiveness in the removal of these ions from aquatic systems (Ragab et al., 2019).

N 2 Adsorption-Desorption Isotherms
The porosity of the hydroxyapatite was analyzed from the N 2 adsorption-desorption isotherms (Fig. 4). The material exhibited type IV(a) isotherms in the IUPAC classification with a large-range hysteresis loop.   This phenomenon can be associated with capillary condensation (Satoshi et al., 2008), whereby a gas is condensed to a liquid-like phase in a pore at a pressure p < p o . This result suggests that mesopores greater than ~4 nm are dominant in the material. Moreover, the hysteresis loop can be classified as Type H3, where complex pore structures exist and networking effects are important. The presence of ink-bottle pores may exist, suggesting a good capturing potential of pollutants for this material when used as an adsorbent. The total surface area (S Total ) estimated by the BET method as well as the total pore volume (V Total ), the micropore volume (V Micro ), and mesopore volume (V Meso ) estimated by the t-plot and BJH methods are presented in Table 1.
The S Total and V Total of the hydroxyapatite are comparable to that of other highly porous HAp (Cedric et al., 2007), opening a great potential for the use of this material as a commercial product. The pore size distribution confirmed that the material was mostly composed of mesopores with pore widths in the range of 2.5 to 20 nm (Fig. 4b).

Thermal Behavior of Hydroxyapatite
The TGA-DTG curves of HAp from room temperature to 800 °C in air, at a heating rate of 5 °C min −1 , are presented in Fig. 5. It can be seen in Fig. 5 that the total weight loss of HAp is about only 0.05%, which is insignificant compared to some other HAp synthesized in previous studies (Marjan et al., 2020;Pramod et al., 2016;Slavica et al., 2001). The first peak at around 100 °C in the DTG corresponded to the remaining moisture released from the material. The main weight loss occurred between 100 and 550 °C with a very small rate of around 0.0001%·°C −1 . This result indicates that HAp is very stable and can maintain its chemical integrity, even under severe thermal conditions.

Effect of Adsorbent Mass and Contact Time
The effect of the adsorbent mass on the removal of Pb 2+ and Fe 3+ from aqueous solutions was investigated in this study. Therefore, 1-170 mg of adsorbent was added into different beakers for 60 min with the pH of Pb 2+ solution of 4 and 2.65 in the case of Fe 3+ . As shown in Fig. 6a, the adsorption efficiency 169.92 0.45 0 0.45 9.6 increases rapidly from 39 to 100%, corresponding to an increase in adsorbent mass (m) from 3 to 10 mg for C 0 = 50 mg·L −1 and contact time t = 60 min. Meanwhile, Fe 3+ ion removal efficiency increased dramatically from 35 to 93%, respectively (Fig. 6b).
The results also showed that the adsorption capacity of Fe 3+ ions by HAp is higher than that of Pb 2+ when the adsorbent mass increases. However, the amount of adsorbent continued to increase by more than 10 mg, and the ion removal efficiency increased insignificantly (in the case of Fe 3+ ) or slightly decreased (Pb 2+ case). It is concluded that the optimal amount of the nanostructural HAp is clearly shown to be around 10 mg, demonstrating its excellent efficacy in ion removal even with only a tiny amount of the adsorbent. It can be seen that the heavy metal ion (HMI) absorption was enhanced due to an increase in the adsorbent dosage and surface area available. After full adsorption of HMI in solution, the amount of HMI adsorbed per unit mass of HAp decreased due to excess accessible sorption sites. The splitting effect of the concentration gradient may possibly be contributing to the decline. The effect of different contact times t, from 5 to 120 min, on the heavy metal ions' adsorption was presented in Fig. 7. It can be seen that the adsorption trend is quite similar for both ions. For an increase of t from 5 to 40 min, the Pb 2+ adsorption efficiency rises from 27 to 97%, and the capacity rises from 71.87 to 269.52 mg·g −1 . However, after 40 min, these parameters remain stable and an adsorption equilibrium is reached. For the removal of Fe 3+ using hydroxyapatite, 80 min was selected as the optimal time, with an adsorption efficiency and capacity of 93% and 201.82 mg·g −1 , respectively. The fast firststage adsorption may be due to the presence of a large number of vacant adsorption sites. When heavy metal ions are adsorbed on HAp materials' adsorption sites, the number of these sites diminishes and the slope becomes flat as the adsorption rate lowers (Sugashini & Begum, 2013).

Effect of pH
The adsorption efficiency was also measured in the function of pH, ranging from 2.5 to 6 for Pb 2+ and 1 to 4 for Fe 3+ , assuming the pH point zero charge (pH pzc ) of 7.49 (Phuong et al., 2020). The surface charge was modified depending on the pH, with pH solution < pH pzc , the surface of the adsorbent having a positive charge, and vice versa. The adsorption efficacy as a function of pH may be described using the pH pzc value.
In the case of the Pb 2+ solution, precipitation was seen at pH values greater than 6, but for Fe 3+ , the critical pH value for precipitation was greater than 4, similar to the effect demonstrated by Mavropoulos et al. (2002) and Hashimoto et al. (2009). The surface of HAp has a positive charge when the pH is lower than pH pzc ; hence, electrostatic force occurs. As a result, these measurements were carried out in The rivalry between H + ions and heavy metal ions causes less efficient adsorption, as seen in Fig. 8. All the data reveal that the optimum adsorption efficiency and capacity at the initial pH for Pb 2+ is 4 and the initial pH for Fe 3+ is 2.65.

Effect of Initial Pb 2+ and Fe 3+ Concentration
To determine the influence of the initial concentration on the adsorption efficiency and capacity of the hydroxyapatite, the concentration C of Pb 2+ solution was varied from 20 to 150 mg·L −1 , and for the Fe 3+ solution, the concentration was changed from 5 to 400 mg·L −1 at t = 80 min. Figure 9 reveals a decrease in the efficiency and an increase of the capacity with the rise of the initial ions concentration, indicating that for particular values, adsorption reaches saturation. There is a specified number of adsorption pores for a given amount of hydroxyapatite. In the case of all the adsorption pores being filled by Pb 2+ and Fe 3+ ions, the adsorption capacity has nearly no change when the concentration of heavy metals increases.

Adsorption Kinetics
A comprehensive study of adsorption processes will be incomplete if the structure and dynamics of its many aspects, as well as how they interact, are not effectively represented. As a result, considerable research into adsorption thermodynamics in combination with adsorption kinetics is required.
The adsorption mechanism was determined based on the experimental results. To investigate the kinetics of that process, Lagergren's pseudo-firstorder law and McKay-Ho's pseudo-second-order law models were used, where both equations of the two models are given in Eqs. (3) and (4), respectively (Tran et al., 2017): where Q t (mg·g −1 ) is the adsorption capacity at time t, Q e (mg·g −1 ) is the adsorption capacity at equilibrium, and k 1 (min −1 ) is the pseudo-first-order adsorption rate constant.
Non-linear models were used to determine the amount adsorbed in the equilibrium, rate constant, and correlation coefficient. Figure 10 shows both nonlinear models for Pb 2+ and Fe 3+ ions. Based on the higher values of R 2 in the adsorption of metal ions, the pseudo-first-order of kinetics was proposed. The parameters estimated based on that model are presented in Table 2. The value Q e between from experimental and calculation is not remote. The non-linear form of kinetic models retains the distribution of the error and kinetic factor unchanged in both axes (Ganguli et al., 2020).

Adsorption Analysis
The equilibrium isotherm is described by the surface characteristics and affinity of the sorbent that is expressed in the values of the sorption isotherm's constants. The equilibrium is reached when the adsorbate concentration in the bulk solution equals the sorbent interface concentration. As follows, equilibrium data from the sorption process may be modeled using different commonly used equilibrium equations, so in our experiments, the Pb 2+ and Fe 3+ adsorption process on HAp was described by using the Langmuir, Freundlich, and Redlich-Peterson models. Tests on adsorption were carried out by mixing 10 mg of the hydroxyapatite with 50 mL of Pb 2+ or Fe 3+ having different concentrations in a 100-mL Duran flask and then stirring at 400 rpm for 80 min. The adsorption isotherms for Pb 2+ and Fe 3+ are presented in Fig. 11, where the simulations of the Langmuir, Freundlich, and Redlich-Peterson models were made based on Eqs. (5)-(7) as follows (Lima et al., 2015;Tran et al., 2017;Tran et al., 2021): where Q e stands for the amount of adsorbent in equilibrium (mg·g −1 ), Q max is the maximal amount adsorbed (mg·g −1 ), K L stands for the Langmuir  equilibrium constant (L·g −1 ), and C e stands for concentration of ions (mg·L −1 ).
where K F is the Freundlich constant with multilayer adsorption relating to the bond strength (mg·g −1 (L·g −1 ) n ) and n is the dimensionless parameter defining adsorption.
where K RP is the Redlich-Peterson constant (L·g −1 ), a RP stands for the Redlich-Peterson equilibrium constant (mg·L −1 ) −g , and g is a constant dimensionless parameter.
According to the Langmuir equation, a monolayer of adsorbate on the absorber's surface is required for maximal adsorption. For Freundlich's model, a broad variety of concentrations may be taken into account by taking into consideration surface heterogeneity, as well as the exponential distribution of active sites and their related energies. Following, the Redlich-Peterson model considers the limitations of the Langmuir and Freundlich models. It can be used to demonstrate adsorption equilibrium over a wide range of concentrations of adsorbate (Tran et al., 2017;Lima et al., 2021). Figure 11 shows the graphical representation of these models, where the values of the correlation coefficient R 2 obtained from particular models were used to determine the best fitting. In the case of the Pb 2+ adsorption, see Fig. 11a, it can be seen that the R 2 for Redlich-Peterson isotherm is higher (0.9996) than for Langmuir isotherm (0.9459) and Freundlich (0.7523) isotherm. Similar to these results, for the Fe 3+ , the value of the correlation coefficient was also higher for the Redlich-Peterson isotherm (0.9964) than for the Langmuir isotherm (0.9691) and Freundlich (0.9325) isotherm. Therefore, it can be seen that the adsorption undergoes the Redlich-Peterson model, which assumes that adsorbate is applicable either in the homogeneous or heterogeneous system. The parameters characteristic to all these models are presented in Table 3, showing that the Q max determined from the Langmuir isotherm is about 392.36 mg·g −1 for Pb 2+ and 1781.68 mg·g −1 for Fe 3+ .
According to the Pb 2+ removal, the literature shows that an increase in metal ion concentration leads to an increase in the number of metal ions present at the adsorbent surface, which in turn leads to an increase in the likelihood of adsorption (Sridevi et al., 2015). This outcome is compatible with S.T. Ramesh's research (Ramesh et al., 2013). They demonstrated that the maximal adsorption capacity of synthetic HAp is 357.14 mg Pb(II)·L −1 with 0.12 g HAp·L −1 optimum dosage and that the Pb 2+ adsorption kinetics match the pseudo-second-order reaction and the Langmuir adsorption isotherm model.  showed that the effectiveness of HAp over the Pb 2+ removal depends on experimental conditions and can vary from 84 to even 620 mg·g −1 (Minh et al., 2013). Another study presented by Meski et al. shows the results for carbonized HAp prepared from the domestic hen waste egg shells, where the Pb 2+ undergoes an adsorption Langmuir mechanism with a maximum adsorption capacity of 500 mg·g −1 at 25 °C and 35 °C (Meski et al., 2010). Furthermore, the Fe 3+ adsorption capacity measured in this study (1089.59 mg·g −1 ) was significantly higher than the maximum iron adsorption capacity in prior investigations (55.25 mg.g −1 ), where the adsorption follows pseudo-second-order kinetics (Brundavanam et al., 2015). Some works refer to the formation of many defects in the HAp structure that influence the effectiveness in the removal of particular compounds (Avakyan et al., 2021;Jiang et al., 2002). However, as mentioned above, the effectiveness of water purification with HAp depends on the experimental conditions. In work presented by Billah et al., the adsorption capacity with the fluorapatite is about 19.88 mg·g −1 , where the adsorption mechanism fits the Langmuir model and the kinetics follows the pseudo-second-order (Billah et al., 2021). Another work shows the 2.54 mg·g −1 adsorption capacity with cow bone-derived HAp (Olabiyi & Adekola, 2018). The kinetic and the isotherm models are proposed to be pseudo-second-order and Langmuir models, respectively.
The adsorption capacity for the Pb 2+ removal is similar to the data presented in the literature, while the values for Fe 3+ removal are much lower than the experimental data obtained in this work. The kinetic models differ from the literature in what can be caused by the application of linear models. This shows that cost-effective and widely available materials like apatite ore in Vietnam may be employed to manufacture extremely effective adsorbents.

Temperature Effect
The experiments were also performed in function of temperature, indicating that along with an increase in temperature, the adsorption efficiency and capacity also rise (Table 4). Then, the standard Gibbs free energy change (ΔG°), the enthalpy change (ΔH°), and the entropy change (ΔS°) of the adsorption of Pb 2+ and Fe 3+ onto the hydroxyapatite were estimated within Eqs. (8) and (9): where R is the universal gas constant (8.314 Jmol −1 K −1 ), T is the temperature (K), and K d is the equilibrium constant.
The intercept and slope of the plot of ln(K d ) versus (1/T) (Fig. 12) give the values of ΔG°, ΔH°, and ΔS°, which are summarized in Table 5. The positive values of ΔH° and the negative values of ΔG° confirm the spontaneous and endothermic character of the ions' adsorption. The values of ΔG° are smaller than the value of physical adsorption (−20 to 0 kJ·mol −1 ) and chemisorption range (−80 to −400 kJ·mol −1 ), suggesting that the adsorption of Pb 2+ and Fe 3+ onto that HAp has a physisorption character. The randomness at the adsorbate-adsorbant interface is confirmed with positive values of ΔS° (Thuy et al., 2021).

Wastewater Treatment
In complement to our detailed investigations of the heavy metal ions removal in the laboratory, real wastewater was tested to confirm the effectiveness of our proposed adsorbent in practical conditions. The wastewater was collected at a plating company in Noi Bai Industrial Zone and Trieu Khuc craft village, Hanoi, Vietnam. The sample had white-bluish color and a metallic odor. Analysis results (Table 6) show that the Fe 3+ concentration in the water supplied by the plating company is about 1.21 mg·L −1 , which is double that in the craft village, where the iron (III) ion concentration is about 0.06 mg·L −1 . These values exceed the maximum permitted limit concentration of domestic water following the QCVN 02:2009/ BYT norm (0.5 mg·L −1 ). The lead (II) ions were not detected in the plating wastewater. Experiments were performed in the following conditions: 0.03 g of hydroxyapatite was placed in 50 mL of wastewater primarily filtered to remove solid impurities suspended in the solution. After the filtration, the adsorbent was stirred for 45 min, likewise, the experiments described above. Then, the sample was centrifuged to separate the adsorbent from the suspension, and the solution was tested with the application of the AAS technique. Besides, the pH of the plating wastewater is very low, out of the range for domestic water (pH 6.0-8.5). After treatment, the pH of the solution increases, and the concentration of Fe 3+ is 0.38 mg·L −1 , smaller than 0.5 mg·L −1 ; the adsorption efficiency is 68.24% in the case of plating wastewater. The Fe 3+ concentration in craft-village wastewater after adsorption is below the detection limit, indicating an efficiency close to 100%. It can  be concluded that hydroxyapatite can be used as an effective absorbent for heavy metals removal from wastewater. Its application also can be used on a larger, industrial scale.
3.9 Crystallinity, FTIR, and EDX analysis Remember that the removal of metal ions by HAp adsorbent follows second-order kinetics and the Langmuir adsorption isotherm model. This allows us to predict that metal ion adsorption will occur mostly on the surface and form chemical linkages. For a better understanding, the adsorbent has been analyzed both before and after the adsorption process. Figure 13 shows the XRD patterns and FTIR spectra of the HAp before and after the heavy metal ions removal.
Characteristic peaks of hydroxyapatite were detected, with the highest intensity peak at around 32.5° and another peak at about 26.2° that corresponds to the (211) and (002) planes, respectively, while the other peaks are also in good agreement with the hydroxyapatite structure (Phuong et al., 2020). After Pb ion adsorption, the XRD pattern of the material has changed remarkably. The appearance of the maximum peak at 30.7°, which is indicative of lead hydroxypyromorphite -Pb 10 (PO 4 ) 6 (OH) 2 and various peaks were recorded (Ellis et al., 2006;Hopwood et al., 2016;Zhu et al., 2015), whereas the maximum peak of HAp was found a considerable decrease. This demonstrates that the exchange of ions in aqueous solution between Ca 10 (PO 4 ) 6 (OH) 2 and Pb 2+ results in a Ca 2+ substitution in the structure and the formation of Pb 10 (PO 4 ) 6 (OH) 2 (Wang et al., 2019).
Similar to lead ions, Fe 3+ adsorption also changes the structure of HAp, leading to the formation of the new phase -goethite -FeO(OH), where the peaks at about 12. 1°, 17.1°, 27.1°, 35.4°, 39.5°, 52.4°, 56.3°, and 68.2° confirm the goethite presence in the sample (Ghosh et al., 2012;Jiang et al., 2002;Kato et al., 2018;Zilm et al., 2016). However, not all calcium ions are substituted with iron ions. It is worth mentioning that post-treated HAp changes its color, proving effective in ion capturing (Qian et al., 2014). XRD results confirm the effectiveness of the HAp for the removal of heavy ions is due to an ion-exchange mechanism (metal ion ➔ Ca 2+ ) that is a main participant in the adsorption process. This result was confirmed by the desorption of Ca 2+ exchange cation in the solution after adsorption against the amount of Pb 2+ and Fe 3+ that were adsorbed (Table 7).
The FTIR spectra of hydroxyapatite before and after Pb 2+ and Fe 3+ adsorption are shown in Fig. 13b Fig. 13 XRD patterns (a) and IR spectra (b) of HAp before and after the adsorption of Pb 2+ and Fe 3+ and the asymmetric bending vibration of O-P-O. The bending variation of the OH − group is characterized by a peak at 1617 cm −1 . The peak at 1389 cm −1 is attributed to NO 3 − group in the modification process (Phuong et al., 2020). After Pb 2+ and Fe 3+ adsorption, the spectrum of material still does not change much because of the functional group of PO 4 3− and OH − in Pb 5 (PO 4 ) 3 (OH) and HAp.
The change in the distribution and content of Ca, Fe, and Pb present in HAp materials will be confirmed by SEM-EDX analysis results.
Furthermore, the Pb and Fe composition on the surface of the sorbents was also determined by energy-dispersive X-ray spectroscopy (EDX). The elemental composition of the HAp before and after the adsorption was examined at different points on the SEM images (described in Fig. 14). The average atomic and mass compositions are shown in Table 8.
The elemental mapping and spectra for Pb-and Fetreated samples have been recorded and reveal the uniform distribution of the elements on the material's surface after the removal of metal ions from water. It denotes the existence of metal ions such as Pb 2+ and Fe 3+ in the adsorbent after the adsorption. In addition, EDX analyses confirm the dosage of separated atomic and mass composition in the material. The amount of Ca decreases, and the amount of Pb and Fe increases after adsorption. The Ca/P ratio of hydroxyapatite is 1.60, approximately 1.67 of stoichiometry HAp, but after Pb and Fe adsorption, the Ca/P ratio is only 1.40 and 1.22, correspondingly.

Conclusions
In this work, HAp was successfully synthesized from apatite ore by chemical modification toward efficient and economical water purification. The obtained material exhibited nanostructures with high thermal stability and high porosity. Moreover, the presence of ink-bottle-shaped mesopores and negative surface potential encouraged the use of this material as an adsorbent for heavy metal removal in wastewater. The adsorption results showed that a small amount (around 10 mg) of hydroxyapatite could effectively remove Pb 2+ from the solution of 50 mg·L −1 ions concentration having 50 mL in volume at pH 0 after 40 min. Under the same conditions, the Fe 3+ ions were also removed effectively but after a much longer contact time of 80 min. The kinetics of Pb 2+ and Fe 3+ adsorption followed the pseudo-first-order model, where the non-linear equations were used. Based on the obtained results, it can be assumed that the occupation of the adsorption sites is proportional to the number of unoccupied sites on the adsorbent. In order to predict the best isotherm equation that describes the adsorption, non-linear Langmuir, Freundlich, and Redlich-Peterson models were used. The highest values of the correlation coefficient for adsorption of both Pb 2+ and Fe 3+ were for a Redlich-Peterson isotherm, assuming the mix of Langmuir and Freundlich models. Interestingly, the literature mainly proposes the Langmuir model, while in some works, the Redlich-Peterson isotherm fitting is not presented in the case of the Fe 3+ removal with HAp. The maximum adsorption capacity of 342.2 mg·g −1 for Pb 2+ is in good agreement with the literature. Interestingly, the adsorption capacity for Fe 3+ obtained by HAp proposed here is about 1089.59 mg·g −1 which is much higher than described elsewhere for the HAp made from different sources, e.g., bone-derived HAp. Thermodynamic parameters confirm that the adsorption process is spontaneous and feasible. Based on the optimization studies for the model pollutants, the studies were also performed for real wastewater. It is seen that the nanostructural HAp works effectively in water purification. It makes the HAp synthesized from natural apatite ore a cost-effective, affordable, and efficient solution for the removal of these contaminants from aqueous solutions in practical conditions.

Declarations
Ethical Approval Not applicable.

Consent to Participate
All authors consent to participate in this work.