Ethanol, nickel (II) nitrate, Pluronic P123, sodium hydroxide, and tetraethyl orthosilicate of analytical grade were bought from Merck. To prepare the nickel stock solution (1000.0 mg L^-1), suitable quantities of nickel (II) nitrate hexahydrate (99.9%) were dissolved in the ultrapure water. The stock solution was used in serial dilutions with ultrapure water to make working solutions. TAN was provided as a surface modifier (ligand) by Sigma Aldrich. To adjust pH and precipitate nanocomposite, a Metrohm 744 pH Meter and centrifuge were utilized, respectively. Ni (II) ions concentration was measured via a Thermo FAAS (Model: M5) (lamp current, 15 mA; monochromator spectral band pass, 0.1 nm; burner head = 50 mm; acetylene and air-flow rates: 0.8 and 10.0 L min^-1, respectively). The diameter of the fibers was determined using an HF2000 (Hitachi) Transmission Electron Microscope. The surface area was evaluated using the Brunauer–Emmett–Teller (BET) technique via Quanta Chrome (Chem BET 300 TPR/TPD). For elemental analysis of nanocomposites, energy-dispersive X-ray spectroscopy (EDS) was carried out using the Mira 3-XMU tool. To prove surface functionalization, elemental analysis CHNS (Costech ECS4010, Italy), Fourier-transform infrared (FT-IR) (Agilent Resolutions pro), and thermogravimetric analysis (TGA) devices (Bahr) were utilized. To produce SBA-15, Whatman #1 filter paper was utilized as a starting material.

After adding 2.0 g TAN and 4.7 g tosyl chloride to ethanol (50 mL), the mixture was put in an ice bath at 5°C and mixed for 45 min. Next, the mixture was stirred for 48 h at 25°C. Stripping off the ultimate precipitated sample using a paper filter, it was rinsed with distilled water and ethanol several times and dried for 1 week at 50°C to create TAN-tosylate. A measure of 1.0 g of TAN-tosylate, 40 mL of ethanol, and 3.0 g of SBA-15 were put under reflux conditions for 24 h at 80°C. The precipitate was then gathered using a paper filter and rinsed several times. Hence, a lack of unreacted reagent was ensured on the sample surface. Until the spectrometer showed no peak for TAN-tosylate absorption, the washing procedure was continued at its λ_max. Therefore, no unreacted TAN-tosylate was found on the sample surface. Next, the sample was dried for 1 week at 50°C. Figure 1 shows the reactions for the surface modification of SBA-15.

The mechanism of extraction involves the formation of a chemical bond between Ni (II) ion and SBA-15/TAN. The Ni (II) ion, on one hand, forms a coordinate covalent bond with a nitrogen atom that has a double bond, and on the other hand, the Ni (II) ion should be coordinated with a heteroatom that facilitates electron loss. Sulfur is a better heteroatom because, first, its electronegativity is lower, and second, while the valence electrons of nitrogen are in 2P orbitals, the valence electrons of sulfur are in 3P orbitals, which are farther from the nucleus and easier to lose.

After the addition of 0.05 g of the nanocomposite to a solution (100 mL) comprising 100 ng mL^-1 Ni (II) ions, extraction was performed. Here, a buffer solution (2.0 mL, acetic acid/sodium acetate) with pH = 6 was added to the mixture. The complete mixture was mixed in a shaker at room temperature for 10 min for the adsorption of Ni (II) ions on the nanocomposite. Then, centrifuging the solution was performed for 5 min at 4,500 rpm for phase separation. After overflowing the upper solution, the sample was eluted by dissolving Ni (II) ions adsorbed on the precipitated sorbent with 1.0 mL HNO₃ (0.2 mol L^-1). After adding HNO₃ and shaking the mixture for 5 min, the solution was re-centrifuged (for 10 min at 5,500 rpm). Finally, the Ni (II) ion concentration was measured by FAAS.

It is clear that the active sites and effective surface area are provided by the synthesis of SBA-15 mesoporous silica in nanoscale to functionalize by chemical reagents. According to BET analysis for SBA-15, the specific surface area of SBA-15 is 364.2 m²/g, which decreases to 327.5 m²/g, followed by surface modification. As stated in the former section, the solid-phase extraction orients the preconcentration of Ni (II) ions through adsorption over the modified SBA-15 active sites. Hence, high surface area and porous structure are satisfactory parameters for better adsorption. The details of the BET technique for the prepared modified SBA-15 represent a more precise surface area than for the modified SBA-15 samples in the literature (Khanjari et al., 2018).

Using elemental analysis CHNS, the types of elements in the nanocomposite were distinguished along with the element ratio within the sample. The existence of TAN ligand on SBA-15 was confirmed through CHNS analysis. Figure 2 shows that the nanocomposite contains the N, C, and S elements associated with TAN in the ratios mentioned. As Si, H, and O elements exist in SBA-15 along with N, C, and S in the nanocomposite, TAN is fixed over SBA-15.

Using FT-IR, the bond formation between TAN and SBA-15 is confirmed (Figure 3). Comparing the FT-IR spectra of SBA-15 with TAN/SBA-15, the immobilization of TAN ligand on SBA-15 is significantly recognized. The broad absorption band at 3,418 cm^-1 is associated with the OH group’s stretching vibration over the SBA-15 surface. The absorption band at 2,906 cm^-1 is allocated to the CH bond symmetric stretching vibration. The absorption band at 1451 cm^-1 is related to the C-H bond vibration of the CH₂ group, while the absorption band at 1084 cm^-1 represents the Si-O bond of SBA-15 (Figure 3A). The peaks of N=N bond at 1511 cm^-1, C–N bond at 1203 cm^-1, C=C bond at 1620 cm^-1, and C–O bond at 1374 cm^-1 can be found in the FT-IR spectra of TAN/SBA-15 (Figure 3B), leading to the TAN attachment to the SBA-15. After the extraction, absorption bands at 679 cm^-1 and 722 cm^-1 are related to the vibration of S-Ni and S-Ni bonds of SBA-15/TAN/Ni, respectively (Figure 3B).

The thermogravimetric study of TAN/SBA-15 was performed to assess the nanocomposite stability and the existence of organic groups in the material (Figures 4A, B). Approximately 10.65 mg of SBA-15 and 13.49 mg of TAN/SBA-15 were heated at 20 °C/min in an argon atmosphere. The existing water and TAN ligand are responsible for weight loss of 7.04% at approximately 95°C (Figure 4A) and weight loss (17.20%) between 480°C and 530°C (Figure 4B), respectively. A small weight and continuous loss are observed at T > 550°C, associated with the combustion of residual organic material. As observed, the TAN structure has less thermal stability than the structure of nanocomposites.

These findings are in line with the results of the EDS analysis. EDS analysis of TAN/SBA-15 (Figure 5) represents the presence of C, S, and N atoms, indicating that TAN is bonded to the external surface of SBA-15 and confirming the good purity of the sample. Moreover, Ni (II) ions are demonstrated in Figure 5 chelated by TAN ligand in the nanocomposite, followed by the adsorption experiment.

The SBA-15 meso cavity contains holes with a completely identical arrangement and a regular 6-sided 2D arrangement. A structural pattern is observed with a certain order (Figure 6), associated with the TEM images for SBA-15. The cavity size in the structure is approximately 25–30 nm. It is proved that the synthesized SBA-15 has nanoscale dimensions.

The effects of some parameters were also studied on preconcentration and extraction. In all experimental circumstances, the concentration of Ni (II) ions was 100 ng mL^-1.

The effects of pH on Ni (II) ion extraction were studied at pH values of 3.0–10.0 while keeping other parameters constant (Figure 7). As the pH of the solution increases, the recovery of Ni (II) improves, with effective recovery obtained at a pH of 5.0–7.0. At lower pH values, the nanocomposite’s surface seems to have a positive charge caused by protonation. Hence, the electrostatic adsorbent between the Ni (II) ions and nanocomposite decreased, resulting in the maximum amount of absorption in a mild acidic solution. At higher pH values, recovery is reduced, probably owing to the nickel hydroxide formation. Thus, by adding the acetate/acetic acid buffer solution, further extraction was conducted at pH 6.0.

The nanocomposite sorbents usually have a larger surface than usual sorbents (Mirabi et al., 2017b). Hence, a more desirable outcome could be obtained by utilizing fewer amounts of nanocomposite sorbents. The effect of several quantities of nanosorbents (SBA-15 and TAN/SBA-15) on the amount of nickel ion sorption was studied on the 10–70 mg scale. Figure 8 shows that using a higher amount than 50 mg of the modified SBA-15 nanocomposite surfaces could lead to a higher amount of nickel ion recovery. Hence, 50 mg was chosen as the optimum concentration of this sorbent. Moreover, good efficiency was not represented by purely SBA-15 when applied separately.

A good adsorbent achieves preconcentration in a short time. Hence, the rate of Ni (II) ion adsorption by modified nanocomposites was investigated with 0.05 g of the TAN/SBA-15 over a series of various shaking times (5–20 min). According to Figure 9, the absorbance of the Ni (II) ions showed no considerable variation after 10 min. Therefore, for further studies, the extraction time of 10 min was chosen. The ligand could create more complexes with nickel ions by increasing the time. Nevertheless, ion desorption is possible over a longer period, thus slightly reducing the extraction percentage.

A higher enrichment factor is obtained using an appropriate eluent. Hence, various experiments were used to select a proper solvent for the desorption of Ni (II) ions from the modified nanocomposites. The Ni (II) ions were desorbed with diverse concentrations (0.1–0.5 mol L^-1) of 1.0 mL of various acids. The results showed (Table 1) that HNO₃ 0.2 mol L^-1 achieved the quantitative elution of the target analyte chelated with TAN. Therefore, we have chosen 1.0 mL of HNO₃ (0.2 mol L^-1) as the solvent for Ni (II) ion desorption.

Adsorbent reusability is a key factor in analytical approaches. For absorption, the adsorption capability was assessed several times. It was indicated that TAN/SBA-15 can be utilized, followed by solvent recovery in the Ni (II) ion extraction process. The procedure was replicated consecutively, with a reduction in the extraction percentage (86.6%) from the order of 3 onwards and a maximum extraction percentage (96.3%–93.6%). The extraction reduction happens owing to nanocomposite particle loss during elution, irreversible/strong interactions between Ni (II) ions and TAN ligand, and releasing the TAN ligand grafts. Hence, the adsorbent can be utilized three times successively for extracting the Ni (II) ions.

Matrix interference is the main limitation to the FAAS method to determine heavy metal ions. The effects of some ions on the recovery of the analyte ions were studied at different concentrations. For Ni (II) ions, the extraction (100 mL) of solutions comprising 100 ng mL^-1 of Ni (II) ions and different quantities of interfering ions was studied (Table 2). According to the results, there is a tolerance limit of < ± 5%. Thus, the recovery of Ni (II) ion extraction is not affected by the existence of foreign ions via TAN/SBA-15. The reason is that the ligand functions as a selective sorbent for nickel ions, and a nickel lamp is utilized as an atomic absorption tool. Hence, it can be selective largely for nickel ions.

Using the suitability values for each technique, the efficiency is compared with other similar approaches (Table 3). The calibration curve has a linear dynamic range (LDR) of 5.0–500 ng mL^-1 and a limit of detection (LOD) of 1.8 ng mL^-1 (Figure 10). Moreover, to obtain a preconcentration factor (PCF) of 100, a sample volume ratio of 100 mL was considered, with a disintegration solvent volume (1 mL), and the enrichment factor (EF = 100) was derived from the concentration ratio of the Ni (II) ions, followed by the extraction to the pre-extraction concentration. The extraction percentage is approximately 100, and the values of EF and PCF are approximately equivalent in some cases, like the present work.

The capability of the presented technique for complex matrix intolerance was investigated using this technique for measuring Ni (II) ions in wastewater and natural water specimens. The accuracy of the technique was verified using the standard increase technique (Table 4). As observed, the method’s extraction efficiency was not highly affected by the complex matrices.