Design, synthesis and application of a sponge-like nanocomposite ceramic for the treatment of Ni(II) and Co(II) wastewater in the zinc ingot industry

2.1 Material

Natural clay was prepared from Lalejin city, Hamadan province, Iran (The physicochemical properties of natural clay are presented in Table 1). Bovine bones were prepared and cleaned of meat, to remove the adipose tissue, the bones were immersed in boiling water at 120 ℃ for 3 h, the boiling was repeated 4 times, then the bones were crushed using a 5 $\mathrm{k}\mathrm{g}$ hammer to size of 5 mm. Human hair at 10 $\mathrm{m}\mathrm{m}$ long was collected from the hairdresser and washed twice with water at 50 ℃ and then dried at room temperature for 2 day. Hydrochloric acid 37%, Nitric acid 65%, sodium hydroxide, $\mathrm{N}\mathrm{i}{\mathrm{N}}_2{\mathrm{O}}_6.6{\mathrm{H}}_2\mathrm{O}$ and, $\mathrm{C}\mathrm{o}{\mathrm{N}}_2{\mathrm{O}}_6.6{\mathrm{H}}_2\mathrm{O}$ were purchased from Sigma-Aldrich company. Two samples of industrial wastewater containing nickel and cobalt ions, respectively, were collected from the zinc ingot factory.

2.2 Preparation of sponge-like nanocomposite ceramics

The crushed bovine bones were placed in a planetary ball mill (Fara Pajouhesh, Isfahan FP2, Iran) at 600 rpm for 2 h. The bovine bone powder (10–120 g), 250 g of natural clay, 50 g of human hair and, 200 g distilled water mixed in a steel container, using a high-powered laboratory mechanical stirrer for 30 min for uniform mixing. The mixture of ceramic compounds was transferred to a mold with dimensions of $20\mathrm{m}\mathrm{m}\times 20\mathrm{m}\mathrm{m}\times 4\mathrm{m}\mathrm{m}$ using a Plexiglas stencil (Fig. S1). Then composite ceramics dried at room temperature for 4 day (The mass of the ceramics was measured and recorded for 10 days, after 4 days no changes in the mass were observed, so 4 days were chosen to dry the ceramics). Human hair in the ceramic structure prevented the ceramic from cracking and breaking during the initial drying stage. The ceramics were calcined in an electric furnace at 850 ℃ for 12 h. Also, the rate of temperature rise is one of the important parameters in the formation of the uniform structure of nano-hydroxyapatite. Therefore, in this process, the rate of temperature increase was selected based on the results of previous researches of 5 ℃ min⁻¹ (Cui et al., 2018; Khadijah et al., 2020). Human hair burns and disappears in the building of ceramic substrates at high temperatures, creating a spongy structure in its void.

2.3 Batch adsorption studies

Adsorption experiments are performed on a fixed bed column with a circulation flow. Fig. 1 a shows a schematic of the adsorption process. According to Fig. 1 b the Fixed bed column with a square cross section was made of glass with a height of 25 $\mathrm{c}\mathrm{m}$ and sides with $1\mathrm{c}\mathrm{m}\times 6\mathrm{c}\mathrm{m}$ . Sponge-like nanocomposite ceramics (Fig. 1 c) were installed inside this column. The cooler pump (Model Alborz 220 V, Electrogen Co. Iran) was installed under the column to circulate the fluid. In each batch adsorption process, 10 $\mathrm{L}$ of wastewater entered the column. The effective factors in the adsorption process included: pH, adsorbent dose, temperature, initial ion concentration and retention time. The experiments were designed based on central composite design (CCD) in the standard response surface method (RSM). Table 2 shows the ranges as well as the coded and uncoded levels of the variables considered in this study. Design-Expert software suggested an experiments pattern with 50 runs, Table S1 shows runs and the result of each batch process. In each batch process, the heavy metal ions removal efficiency and adsorption capacity were calculated by Eq. (1) and (2), respectively:

Close

Fig. 1

Schematic of the adsorption process (a), Fixed bed column for installation of nanocomposite ceramics and tank equipped with a pump for Ni(II) wastewater circulation (b), Photographed image of sponge-like nanocomposite ceramics (c).

Export to PPT

2.4 Measurement and methods

FT-IR spectra (C88731 spectrophotometer, Perkin Elmer Co., Germany) in the range of 400–4000 ${\mathrm{c}\mathrm{m}}^{-1}$ was used to identify adsorbent functional groups. The morphology of sponge-like nanocomposite ceramics was investigated using field emission scanning electron microscopes (FE-SEM, TE-SCAN, Czech). The structure of sponge-like ceramic adsorbents was analyzed using X-ray diffraction (Philips PW3040). The pore structure is an important factor in the adsorption of heavy metal ions on the adsorbent, so BET and the pore structure of nanocomposite ceramic were calculated by BET analysis (Costech Sorptometer 1042). The concentration of metal ions was measured using atomic absorption (Varian Atomic Adsorption 240 spectrometer).

2.5 Mechanical strength and chemical resistance tests

Mechanical strength was calculated based on three-point bending tests method, using a manual hydraulic press device (Perkin Elmer Co., Germany). The sponge-like ceramic was placed on two steel rods with a diameter of 3 mm (The distance between the two bars was 10 cm.) and the pressure was applied from above on the center of the sponge-like ceramic at a loading rate of 1 $\mathrm{m}\mathrm{m}{\mathrm{m}\mathrm{i}\mathrm{n}}^{-1}$ . The pressure that caused the ceramic to break was recorded (Beni and Esmaeili, 2020). To evaluate the chemical resistance, the sponge-like nanocomposite ceramics were placed in an acidic ( $\mathrm{H}\mathrm{N}{\mathrm{O}}_3$ , pH 1) or alkaline ( $\mathrm{N}\mathrm{a}\mathrm{O}\mathrm{H}$ , pH 13) solution and the weight loss of the ceramic was calculated per hour (Shameli and Ameri, 2017; Torabi and Ameri, 2016) by Eq. (3):

2.6 Development of regression model equation

According to experimental results and its predicted (Table S1), the quadratic model was selected for response, that this model proposed by the software. This model with significant terms was expressed by Eq. (4) and Eq. (5) for removal of Ni(II) and Co(II), respectively:

A positive sign against each term of the Eq. (4) and Eq. (5) indicates a synergistic effect and the negative sign indicates a synergistic effect on the response surface $(\mathrm{R})$ .

The actual responses versus predicted responses in Fig. S2 a shown approximately a linear relationship with partial variation. The normal plot of residuals in Fig. S2 b is similar to a straight line indicating that the errors are evenly distributed and therefore support the least squares fit. According to Fig. S2 c-d, the residuals versus the predicted response and the residuals versus the experimental run exhibit the residuals has been distributed above and below the x-axis with unusual structure and no obvious pattern. Also, the ${\mathrm{R}}^2$ values for model equations were 0.981, the adjusted ${\mathrm{R}}^2$ and predicted ${\mathrm{R}}^2$ were 0.968 and 0.931, respectively with difference less than 0.2. The analysis of variance (ANOVA) helps to check the accuracy and validity of the proposed model. According to Table S2 and Table S3, based on ANOVA, the models for ${\mathrm{R}}_{\mathrm{N}\mathrm{i}}$ and ${\mathrm{R}}_{\mathrm{C}\mathrm{o}}$ were highly significant with F-value 76.10 and 76.07, respectively; p-value $<$ 0.0001 also, all of the factors ( ${\mathrm{X}}_1$ , ${\mathrm{X}}_2$ , ${\mathrm{X}}_3$ , ${\mathrm{X}}_4$ and ${\mathrm{X}}_5$ ) were significant. Therefore, these models are acceptable for the predicted results and optimization of the factors affecting the removal efficiency of Ni(II) and Co(II).

3.1 Mechanical properties

Human hair was added to the clay to create micro-channels (50 g of human hair to 450 g clay). The addition of hair also protected the sponge-like ceramic from disintegrating during the drying phase in the environment. Fig. 2a show the mechanical strengths of sintered sponge-like ceramics as functions of bone powder mass. The strength of sponge-like ceramics was based on the breaking stress of ceramics in the range of 1.44 to 3.5 bar. Sponge-like ceramics made of clay without bone powder had the highest mechanical strength because the clay particles were placed uniformly next to each other, but with the addition of bone powder, the clay particles were not uniformly placed next to each other, but nano-hydroxyapatite particles were placed between the clay particles. Beni et al. (Beni and Esmaeili, 2019) used nano-rubber in clay ceramic, their results showed a breaking pressure was 1.55–1.90 bar. Therefore, adding 100 g of bone powder to 450 g of clay and 50 g of human hair showed acceptable mechanical strength (Break pressure 2.4 bar).

Close

Fig. 2

Effect of bone powder mass (added to 450 g of clay and 50 g of hair) on the mechanical strength of sponge-like nanocomposite ceramics (a), Weight loss of the sintered sponge-like nanocomposite ceramic in nitric acid (pH 1) and sodium hydroxide (pH 13) solutions as a function of time at ambient temperature (b), FT-IR spectra of sponge-like nanocomposite (c), X-ray diffraction profiles of clay ceramic, hydroxyapatite and, sponge-like nanocomposite ceramic (d). The FE-SEM image of sponge-like nanocomposite ceramics (e-f), BET and the pore structure of nanocomposite ceramic (g).

Export to PPT

3.2 Chemical resistance

The stability of sponge-like ceramics in acidic and alkaline solution was investigated. Fig. 2 b shows the mass reduction of the ceramic, which was immersed in an acidic solution and a base in two separate experiments and its weight was measured at different times. The synthesized sponge-like nanocomposite ceramics showed 0.5% and 3.73% weight loss after immersion in the $\mathrm{N}\mathrm{a}\mathrm{O}\mathrm{H}$ solution (pH 13) and $\mathrm{H}\mathrm{N}{\mathrm{O}}_3$ solution (pH 1) respectively, after 24 h. In the concentrated $\mathrm{N}\mathrm{a}\mathrm{O}\mathrm{H}$ solution, the mass loss is negligible, but in the acidic solution weak resistance was observed and the sponge-like ceramic was corroded. According to Mouiya et al. (Mouiya et al., 2018), the results show that the nano-hydroxyapatite in ceramic tissue dissolves in acidic solution.

3.3 Characterization of sponge-like nanocomposite ceramics

According to Fig. 2 c was presented the FT-IR spectrum of sponge-like nanocomposite ceramic containing hydroxyapatite with a heat treatment of 750 ℃ before adsorption process. The characteristic signals belonging to the hydroxyl $\left(-\mathrm{O}\mathrm{H}\right)$ and phosphate ${\left({\mathrm{P}\mathrm{O}}_4\right)}^{-3}$ functional groups can be identified at 3573 and 1045 ${\mathrm{c}\mathrm{m}}^{-1}$ , respectively. XRD pattern of clay ceramic, hydroxyapatite and, sponge-like nanocomposite ceramic was shown in Fig. 2 d. It can be seen that the prepared hydroxyapatite belongs to the hexagonal structural type of hydroxyapatite with space group ${\mathrm{P}6}_3/\mathrm{m}$ . Based on BET analysis (Fig. 2 g), the average surface area of the sponge-like nanocomposite ceramic obtained was 172.46 ${\mathrm{m}}^2{\mathrm{g}}^{-1}$ and total pore volume was 913.73 ${\mathrm{c}\mathrm{m}}^3{\mathrm{g}}^{-1}$ . In order to study the morphology of hydroxyapatite nanoparticles and porosity of sponge-like nanocomposite ceramics, the FE-SEM micrographs were presented in Fig. 2 e-f. Hydroxyapatite nanoparticles were observed in clay ceramics around porous spaces with an average diameter of 100 nm.

3.4 Water permeability in sponge-like nanocomposite ceramic

An innovative method was used to investigate the distilled water permeability inside sponge-like nanocomposite ceramics. According to Fig. 3 a in this method a piece of sponge-like ceramic was installed at the bottom of a column 20 $\mathrm{c}\mathrm{m}$ high and 4 $\mathrm{c}\mathrm{m}$ in diameter. Fig. 3 b shows the relationship between the water level in the column and the average mass flow rate. The results showed that the physical structure of ceramic is very porous, similar to sponge. Water can seep into the sponge-like ceramic and drain from the other surface of the sponge-like ceramic. The sponge-like structure was created due to the use of human hair in the manufacture of nanocomposite ceramics.

Close

Fig. 3

Schematic of water permeability test in sponge-like nanocomposite ceramic adsorbent (a), The relationship between the water level in the column and the average mass flow rate from the sponge-like nanocomposite ceramic (b).

Export to PPT

3.5 Adsorption test

3.5.1

3.5.1 Effect of solution pH

The pH is an important factor in the adsorption of heavy metal ions on the adsorbent based on clay (Sáez et al., 2020), because pH affects the degree of ionization, the solubility of heavy metal ions changes with pH, also effective in developing the opposite electric charge in the adsorbent surface functional groups (Vahdat et al., 2019). At low pH (pH less than 4), large amounts of ${\mathrm{H}}^{+}$ ions surround the adsorbent surface and compete with Ni(II) ions for adsorption on adsorbent active sites (Beni, 2021). Hubadillah et al. (Hubadillah et al., 2020) synthesized a bio-ceramic based on hydroxyapatite using from waste cow bone to textile wastewater treatment; they reported that at a pH 7.3 bio-ceramic surfaces it has a negative charge and is more suitable for absorbing positive ions; while, Googerdchian et al. (Googerdchian et al., 2018) showed that the mechanism of Pb(II) adsorption on nano-hydroxyapatite is dissolution-precipitation and, at low pH 3 the dissolution of hydroxyapatite increases significantly and formed hydroxypyromorphite, ${\mathrm{P}\mathrm{b}}_{10}{({\mathrm{P}\mathrm{O}}_4)}_6{(\mathrm{O}\mathrm{H})}_2$ . Effect of solution pH on the Ni(II) and Co(II) removal efficiency (R %) presented in Fig. 4 a, where the mass of clay nanocomposite ceramic, temperature, initial concentration and retention time were fixed at 13.5 $\mathrm{g}$ , 40 ℃, 252.5 $\mathrm{m}\mathrm{g}{\mathrm{L}}^{-1}$ , and 125 $\mathrm{m}\mathrm{i}\mathrm{n}$ , respectively. The results show that by increasing the pH from 3 to 6, the adsorption efficiency increased, and then, after pH 6, the removal efficiency decreases. Similar to our results, Pb(II), Cr(VI), Ni(II), and Cd(II) from aqueous solution adsorbed at a pH 6 on calcined clay (Zeng et al., 2019), therefore, at high pH, hydroxyl groups have high efficiency (Gu et al., 2020) and also the stability of sponge-like nanocomposite ceramic improves the adsorption process.

Close

Fig. 4

Effect of pH (a) and temperature (b) on Ni(II) and Co(II) removal efficiency.

Export to PPT

3.5.3

3.5.3 Effect of adsorbent mass and initial concentration

The absorbent dose plays a role in making the process economical (Soliman and Moustafa, 2020). Where dose is the sum of the sponge-like nanocomposite ceramic substrates mass relative to the total volume of the wastewater. As the adsorbent mass increases, the removal efficiency of heavy metal ions increases (Mnasri-Ghnimi and Frini-Srasra, 2019). The presence of more than the optimal amount of adsorbent causes the active sites of adsorbent to remain unoccupied and the adsorption capacity is reduced (Dias Filho and Do Carmo, 2006). According to Fig. 5 a-b, in the fixed condition (pH 6, temperature 40 ℃, initial concentration 252.5 $\mathrm{m}\mathrm{g}{\mathrm{L}}^{-1}$ and retention time 125 $\mathrm{m}\mathrm{i}\mathrm{n}$ ), with increasing the adsorbent mass from 4.5 to 13.5 $\mathrm{g}$ , the removal efficiency of Ni(II) and Co(II) ions with a steep slope has increased and then the slope of the removal efficiency is fixed. The adsorption capacity curve shows that by adding more than 9–13.5 $\mathrm{g}$ of adsorbent mass, the adsorption capacity is sharply reduced.

Close

Fig. 5

Effect of adsorbent mass on adsorption capacity of sponge-like nanocomposite ceramic and removal efficiency: Ni(II) wastewater (a) and Co(II) wastewater (b). Effect of initial ion concentration on adsorption capacity of Sponge-like nanocomposite ceramic and removal efficiency: Ni(II) wastewater (c) and Co(II) wastewater (d).

Export to PPT

The initial concentration is one of the effective parameters in the adsorption process. According to Fig. 5 c-d, with increasing the concentration of Ni(II) and Co(II) ions from 5 to 180 $\mathrm{m}\mathrm{g}{\mathrm{g}}^{-1}$ , the removal efficiency increased; also, with increasing the concentration from 180 to 500 $\mathrm{m}\mathrm{g}{\mathrm{g}}^{-1}$ , the removal efficiency decreased. Therefore, the removal efficiency of heavy metal ions at any concentration corresponds to active sites that are readily available (Gu et al., 2020). However, increasing the initial concentration of heavy metal ions increases the adsorption capacity. Similar to the results of this study, Es-sahbany et al. (Es-Sahbany et al., 2019) reported that natural clay has the ability to adsorb Ni(II) ions at concentrations of 180 to 200 with a removal efficiency of about 75%.

3.5.4

3.5.4 Effect of retention time and kinetic models

To determine the equilibrium adsorption time, the change retention time was in the range of 10 to 240 $\mathrm{m}\mathrm{i}\mathrm{n}$ and the effective parameters in the adsorption were considered constant (pH 6, temperature 40 ℃, initial concentration 252.5 $\mathrm{m}\mathrm{g}{\mathrm{L}}^{-1}$ and adsorbent mass 13.5 $\mathrm{m}\mathrm{i}\mathrm{n}$ ).

According to Fig. 6 a-b, more than 50% of ions removed at 60 min. As increasing the retention time, the removal efficiency increases and, accordingly, the adsorption capacity also increases (El-Nagar et al., 2020). The Ni(II) and Co(II) removal efficiency during start process to 120 $\mathrm{m}\mathrm{i}\mathrm{n}$ increased from 5 to 73% and 62%, respectively. The adsorption rate decreased as the adsorption process approached equilibrium (at 120 to 180 $\mathrm{m}\mathrm{i}\mathrm{n}$ ) (Vahdat et al., 2019). Removal efficiency remained constant from 180 to 220 $\mathrm{m}\mathrm{i}\mathrm{n}$ ; therefore, adsorption entered the equilibrium stage after 180 $\mathrm{m}\mathrm{i}\mathrm{n}$ (Zhang et al., 2020). Googerdchian et al. (Googerdchian et al., 2018) reported a Pb(II) adsorption capacity of 200 $\mathrm{m}\mathrm{g}{\mathrm{g}}^{-1}$ in the adsorption process using hydroxyapatite nanoparticles at 60 $\mathrm{m}\mathrm{i}\mathrm{n}$ . This study showed that at a retention time of 60 $\mathrm{m}\mathrm{i}\mathrm{n}$ with an adsorbent mass of 9 $\mathrm{g}$ , the adsorption capacity is about 200 $\mathrm{m}\mathrm{g}{\mathrm{g}}^{-1}$ , while if the retention time is 240 $\mathrm{m}\mathrm{i}\mathrm{n}$ , the adsorption capacity will be approximately 300 $\mathrm{m}\mathrm{g}{\mathrm{g}}^{-1}$ .

Close

Fig. 6

Effect of retention time on adsorption capacity of sponge-like nanocomposite ceramic and removal efficiency: Ni(II) wastewater (a) and Co(II) wastewater (b).

Export to PPT

The pseudo first-order, second-order and, intraparticle diffusion models were used to fit the kinetic data according to Eq. (6), (7) and, (8) respectively.

(6)

$$\log (q_e-q_t)=\log q_e-(\frac{k_1}{2.303})t$$

(7)

$$\frac{1}{q_e-q_t}=\frac{1}{q_e}+k_{2}t$$

(8)

$$q_e=k_3{t}^{0.5}+C$$ where ${\mathrm{q}}_{\mathrm{e}}$ and ${\mathrm{q}}_{\mathrm{t}}(\mathrm{m}\mathrm{g}{\mathrm{g}}^{-1})$ are the adsorption capacity at equilibrium and $\mathrm{t}$ times $(\mathrm{m}\mathrm{i}\mathrm{n})$ , respectively. The ${\mathrm{k}}_1$ in the pseudo first-order model $({\mathrm{m}\mathrm{i}\mathrm{n}}^{-1})$ , $k_2$ in the second-order model $(\mathrm{m}\mathrm{g}{(\mathrm{g}\mathrm{m}\mathrm{i}\mathrm{n})}^{-1})$ and, $k_3$ in the intraparticle diffusion model $(\mathrm{m}\mathrm{g}{{\mathrm{g}}^{-1}\mathrm{m}\mathrm{i}\mathrm{n}}^{-0.5})$ are the rate constants.

The kinetic rate constants obtained from the kinetic models are given in Table 3. The kinetic data of Ni(II) and Co(II) adsorption with nanocomposite ceramic fitted well using pseudo first-order kinetic model ( $R^2>0.96$ ) also, the data of the intraparticle diffusion model with $R^2>0.95$ showed that the adsorption proceeds with intraparticle diffusion (diffusion within microchannels) instead of surface adsorption.

Table 3 Summary of sorption data evaluated by different kinetic models.

Wastewater sample	${\mathrm{q}}_{\mathrm{m}\mathrm{a}\mathrm{x}}(\mathrm{m}\mathrm{g}{\mathrm{g}}^{-1})$	Pseudo first – order model			Pseudo second – order model			Intraparticle diffusion model
		${\mathrm{q}}_{\mathrm{e}}(\mathrm{m}\mathrm{g}{\mathrm{g}}^{-1})$	${\mathrm{K}}_1({\mathrm{m}\mathrm{i}\mathrm{n}}^{-1})$	${\mathrm{R}}^2$	${\mathrm{q}}_{\mathrm{e}}(\mathrm{m}\mathrm{g}{\mathrm{g}}^{-1})$	${\mathrm{K}}_2$ (mg ${\mathrm{g}}^{-1}$ ${\mathrm{m}\mathrm{i}\mathrm{n}}^{-1}$ )	${\mathrm{R}}^2$	${\mathrm{K}}_3$ (mg ${\mathrm{g}}^{-1}$ ${\mathrm{m}\mathrm{i}\mathrm{n}}^{-0.5}$ )	C	${\mathrm{R}}^2$
Ni(II)	129.54	160.92	0.019	0.962	38.91	0.0009	0.659	9.725	5.21	0.957
Co(II)	100.84	133.32	0.018	0.967	89.44	0.0007	0.752	8.507	16.60	0.956

4.1 Thermodynamic parameters

Gibbs free energy $(\mathrm{\Delta }\mathrm{G})$ , enthalpy $(\mathrm{\Delta }\mathrm{H})$ , and entropy $(\mathrm{\Delta }\mathrm{S})$ as thermodynamic parameters are calculated from the following equations:

4.2 Adsorption isotherms

Adsorption isotherm is a valuable curve that describes the phenomenon governing the retention or mobility of a substance from an aqueous medium to a solid phase at constant temperature (Ajiboye et al., 2021). Equilibrium adsorption data were fitted based on different adsorption isotherm models, include: Langmuir, Freundlich and Dubinin-Radushkevich (D-R) isotherm models. In the Langmuir isotherm model assuming that all active sites are the same and adsorption is performed in a single layer in this system (Karimi et al., 2019); In the Friendlich isotherm model, adsorption is considered in several layers of the adsorbent surface. This model can be used for the adsorption process at the heterogeneous surface (Al-Ghouti and Da’ana, 2020) and D-R isotherm can be used to detect the physical adsorption of metal ions from chemical adsorption (Esmaeili and Beni, 2014). Based on the calculations of the parameters and constants of the isotherm models (Table 5), the data fit well with the Langmuir model ( ${\mathrm{R}}^2>0.98$ ). Also, the adsorption capacity obtained from the Langmuir model is closer to the actual adsorption capacity of the sponge-like nanocomposite adsorbent than Freundlich isotherm model. According to Table 4 the results of data fitting with the D-R model, the free adsorption energy was calculated to be less than $8\mathrm{k}\mathrm{J}{\mathrm{m}\mathrm{o}\mathrm{l}}^{-1}$ , so the adsorption process of Ni(II) and Co(II) on the nanocomposite surface indicate a physical ion exchange.

4.3 Adsorption mechanism

The sponge-like nanocomposite ceramic substrates synthesized in this study were prepared from two main components (natural clay base and hydroxyapatite additive component). Therefore, this soil-bone adsorbent has surface and structural properties of natural clay and hydroxyapatite. So that there are many functional groups in this adsorbent and show complex adsorption mechanisms relative to heavy metal ions (Zhang et al., 2021b). The montmorillonite clays used in this study have a permanent negative charge on the surface that balances the negative charge by exchangeable cations (Tajuddin et al., 2020). Adsorbable exchangeable cations are replaced by Ni(II) and Co(II) heavy metal ions. According to Fig. 2 c, which shows the FTIR spectrum before and after the adsorption process, ion exchange between the cation ions of the heavy metals Ni(II) and Co(II) takes place in functional groups such as $(-\mathrm{O}\mathrm{H})$ and ${\left({\mathrm{P}\mathrm{O}}_4\right)}^{-3}$ which is represented by hydroxyapatite. Therefore, ion exchange is one of the main mechanisms in this process.