Selective removal of Cd(II), As(III), Pb(II) and Cr(III) ions from water resources using novel 2-anthracene ammonium-based magnetic ionic liquids

3.1.1 ¹H NMR spectra of 2-AA and [2-AA]Cl intermediate

The 2-AA spectrum (Fig. 2) shows the amine N-H protons at the chemical shift of 5.5 ppm. After protonation, the N-H signal became very weak and was strongly shifted downfield to around δ = 8–9. Similarly, all the C-H protons signals were downfield shifted by δ = 0.3. Interesting, there were no additional peaks upon protonation of 2-AA to form [2-AA]Cl and the integral area peaks were directly proportional to the number of hydrogen atoms, which implies high purity of the [2-AA]Cl intermediate.

Close

Fig. 2

¹H NMR spectra of 2-aminoanthracene and anthracene ammonium chloride intermediate.

Export to PPT

3.1.2 FTIR of [2-AA]CoCl₃ and [2-AA]FeCl₄ MIL adsorbents

The functional groups present in both 2-anthracene ammonium tetrachloroferrate (III) and 2-anthracene ammonium trichlorocobaltate (II) MIL adsorbents were elucidated using the FTIR. As presented in Fig. 3, the broad band observed between 3100 and 3400 cm⁻¹ are characteristics of stretching vibrations of the N-H bond in the adsorbents. Two peaks, one at 1590 cm⁻¹ owing to the NH₂ scissoring mode and the other at 3300 cm⁻¹ due to the v(N-H) stretching mode, are indicative of the amine group. Peaks observed below 1500 cm⁻¹ in the low-frequency portion of the spectra are caused by complex formation. The band seen at 2969.9 cm⁻¹ are C-H stretching vibrations due to the anthracene C-Hs. The skeletal vibrations of the rings are found at 1423.3 cm⁻¹ to 1650 cm⁻¹, while the broad band at 3300 – 3500 cm⁻¹ in Fig. 3(c) and (d) is characteristic of OH stretching vibration due to the hygroscopic nature of the chlorides of cobalt and iron. The intermediate anthracene ammonium chloride's lack of sharp peaks at roughly 3300 cm⁻¹ is obviously caused by an acid-base complex formation of the NH³⁺. The peak also observed at around 1615 cm⁻¹ to 1620 cm⁻¹ is caused by the v(C = C) stretching mode and demonstrates that the C = C double bond is intact and not distorted. The characteristic behavior of the two MILs is considered similar and confirms the formation of the complex metal chlorides (Hassan et al., 2021; Yassin et al., 2015).

Close

Fig. 3

FTIR spectrum of: a) 2-aminoanthracene b) Intermediate anthracene ammonium chloride c) 2-anthracene ammonium tetrachloroferrate(III) d) 2-anthracene ammonium trichlorocobaltate(II); (e) TGA decomposition curve of [2-AA]FeCl₄ and [2-AA]CoCl₃; (f) UV–vis spectra of 2-aminoanthracene, intermediate anthracene ammonium chloride, 2-anthracene ammonium tetrachloroferrate (III) and 2-anthracene ammonium trichlorocobaltate (II); (g) Zeta potential of MIL, 2-anthracene ammonium tetrachloroferrate (III) and 2-anthracene ammonium trichlorocobaltate (II).

Export to PPT

3.1.3 Thermogravimetric (TGA) analysis of [2-AA]CoCl₃ and [2-AA]FeCl₄ MIL adsorbents

The TGA decomposition curve of both MILs show similar decomposition pattern (Fig. 3(e)), with gradual decomposition up to 400 °C for [2-AA]CoCl₃ and 500 °C for [2-AA]FeCl₄. The decomposition below these temperatures is in three steps: first at 70 °C, which is ascribed to the melting point of the respective MILs, then at 150 °C, which is assigned to the loss of hydrated water molecules, and the sharp decrease observed above 300 °C is due to the decomposition of the respective MILs to form oxides. In general, the synthesized MILs have shown remarkable thermal stability with less than 10% weight loss up to 300 °C. This observation indicates that the magnetic moment to magnetic mass ratio is connected to the magnetization curves of the TGA. A similar observation was reported when cobalt was used in the synthesis of some MILs (Medeiros et al., 2012).

3.1.4 UV/Vis spectra of [2-AA]CoCl₃ and [2-AA]FeCl₄ MIL adsorbents

Superimposed UV–vis spectra of the starting material 2-aminoanthracene, the intermediate ion anthracene ammonium chloride, 2-anthracene ammonium tetrachloroferrate (III), and 2-anthracene ammonium trichlorocobaltate (II) was presented in Fig. 3(f). It was observed that all the spectra have similar pattern, however, the [2-AA]CoCl₃ and [2-AA]FeCl₄ MIL adsorbents show sharp and well resolved hypochromic shift peaks from 400 nm to 362 nm (indicated by the arrow). This can be associated with the stiffness introduced into the rings and confirms the formation of the corresponding complexes. Similar trend has been observed in previously reported MILs (Abdi Hassan et al., 2020; Yassin et al., 2015).

3.1.5 Zeta potential of [2-AA]CoCl₃ and [2-AA]FeCl₄ MIL adsorbents

The Zeta potential was measured as a function of pH dispersed in a double-distilled water, and the isoelectric point (IEP) was determined as the pH level at which the zeta potential is zero. The result of zeta potential measurement at different pH shown in Fig. 3(g) indicates that by increasing the pH above isoelectric point of 3.8 and 5.0, the interaction of the cationic moiety anthracene-2-aminium with the alkaline medium decreases, which is due to the repulsive forces generated by the negative charges of the basic medium and FeCl₄^- and similar behavior is observed in the two MILs. This indicates that the particles have both positive and negative charge. The higher mV values obtained are due to the presence of the aromatic rings (Breite et al., 2016; Sarkar et al., 2021).

3.1.6 Magnetization properties of [2-AA]CoCl₃ and [2-AA]FeCl₄ MIL adsorbents

The magnetization study indicates that [2-AA]FeCl₄ is ferromagnetic while is [2-AA]CoCl₃ superparamagnetic as shown by the s-shaped curve in Fig. 4(a) and Fig. 4(b). The superparamagnetic property of [2-AA]CoCl₃ is attributed to the crystallite size reduction, which decreases the anisotropy energy, hence favoring the superparamagnetism of the material. In addition, high saturation magnetization and high two-curie temperature of cobalt-containing MIL can be due to allotropic characteristics of the element. In the case of [2-AA]FeCl₄ which has depicted ferromagnetic property can be associated with the presence of iron which has a curie point at 770 °C and changes its crystalline structure above the curie point. It is also noted that in both samples, the temperature is directly proportional to the magnetic moment (Fig. 4(c) and Fig. 4(d)), which agrees with Curie–Weiss law. Furthermore, a linear relationship between the magnetic field and the magnetization has been recorded and obtained at room temperature indicating that both MILs are magnetic. The magnetic susceptibility was found to be 1.17 and 1.16 for [2-AA]FeCl₄ and [2-AA]CoCl₃ respectively. This implies that both MILs have magnetic properties that can be attributed to iron and cobalt due to their structures which allow their electrons to line up more easily forming the magnetic field of the materials.

Close

Fig. 4

The magnetization of a) [2-AA]FeCl₄ as a function of the applied magnetic field at a temperature range of 10–300 K b) [2-AA]CoCl₃ as a function of the applied magnetic field at a temperature of 5–300 K c) [2-AA]CoCl₃ as a function of temperature in an applied magnetic field range of −0.2–1000 Oe and d) [2-AA]FeCl₄ as a function of temperature under an applied magnetic field of 50–1000 Oe.

Export to PPT

3.1.7 Morphological examination of [2-AA]CoCl₃ MIL adsorbent

The SEM image of [2-AA]CoCl₃ MIL before adsorption shows microparticle sizes with non-spherical, irregular shape, generally with fine texture, and non-uniform surface morphology (Fig. 5(a)). The thicker crystal-like structure observed after adsorption (Fig. 5(c)) is probably due to the presence of the adsorbed heavy metal ions on the surface of the [2-AA]CoCl₃ MIL adsorbent. In addition, the EDX spectrum shown in (Fig. 5(c)) confirms the presence of all the expected elements in their stoichiometric weight percentages in the [2-AA]CoCl₃ MIL adsorbent, while the EDX spectrum shown in (Fig. 5(d)) confirms the adsorption of the targeted heavy metal ions on the surface of [2-AA]CoCl₃ MIL adsorbent after the adsorption process.

Close

Fig. 5

SEM micrograph of [2-AA]CoCl₃ MIL adsorbent a) before and c) after absorption and EDX images b) before and d) after adsorption in 5 μm magnification.

Export to PPT

3.2.1 Effect of [2-AA]CoCl₃ dosage on its adsorption capacity

The effect of [2-AA]CoCl₃ dosage on its adsorption capacity for removal of Cd²⁺, As³⁺, Pb²⁺ and Cr³⁺ was studied by varying the adsorbent dosage from 3 to 30 mg. The adsorption experiment was carried out by dispersing the adsorbent in 10 mL solution of 100 ppm each of metal ions, then stirred for 24 h contact time. It was observed that the percent removal of the heavy metals increases with increase in the dosage up to 30 mg. However, the adsorption capacity (which factors in the mass of adsorbent, as shown in equation (1)) increases from 3 to 5 mg, then decreases up to 30 mg (Fig. 6b). Therefore, the optimized adsorbent dosage is 5 mg.

3.2.2 Effect of solution pH on the adsorption efficiency of [2-AA]CoCl₃

The pH plays a significant role in the adsorption capacity, as it can affect both the metal speciation through the hydrolysis reaction and the electrostatic interaction between the adsorbent and the adsorbate (Wang et al., 2013). In this regard, a wide pH range (2 – 9) was investigated in the adsorption experiments. As shown in Fig. SI-1, the adsorption efficiency of [2-AA]CoCl₃ increases with increase in the pH, especially for Pb²⁺ and Cr³⁺, which showed an excellent percent removal of more than 99% at pH 8 and 9 without major differences. However, by increasing the pH to beyond 9, it was observed that the metal ions precipitate and that is why higher pH was not considered. Therefore, pH 8 was chosen for the rest of the study. The pH result agrees with the isoelectric point of MIL, which shows a larger negative charge on the surface above pH 5, as shown in Fig. 3(g), and this favors the electrostatic interaction between the cationic species and the large negative adsorption sites (Ali et al., 2018). In contrast, the lower pH below the net surface charge of zero exhibits positive active sites, which produce unfavorable electrostatic interaction that causes repulsion between the cationic metal ions and the positively charged surface of MIL. Therefore, in general, the electrostatic interaction is the dominant factor affecting the adsorption behavior of the metal ions.

3.2.3 Effect of contact time on the adsorption efficiency of [2-AA]CoCl₃

The contact time between the [2-AA]CoCl₃ MIL and metal ions in the solution affects the adsorption efficiency of the adsorbent. As presented in Fig. SI-2, the adsorption capacity increases from 5 min to 30 min, then stabilizes up to 60 min. It was however, noticed that the adsorption capacity decreases slightly after the 60 min contact time for some of the heavy metal ions. This behavior can be attributed to the saturation of the pores of the adsorbent (Changmai et al., 2018; Darama et al., 2021; Hastuti and Siswanta, 2015) or gradual ionization leading to the formation of OH. Therefore, to save energy and time, 60 min was chosen as the equilibrium time for the subsequent adsorption studies.

3.2.4 Effect of metal ions concentration on the adsorption capacity of [2-AA]CoCl₃

Generally, the metal ions concentration affects the adsorption capacity of adsorbents (see equation (1)). The effect of metal ions concentration was studied by varying the initial concentrations of the ions from 50 to 350 mg/L. The results show that the adsorption capacity of [2-AA]CoCl₃ increases with increase in concentration of metal ions up to 250 ppm (Fig. SI-3), then decreases slightly. This phenomenon may be due to the saturation of the active sites of the MIL. Based on the optimized result, the adsorption capacity of between 171 and 265 mg/g was recorded for the heavy metal ions.

3.2.5 Kinetics studies

The adsorption processes of the Cd²⁺, As³⁺, Pb²⁺ and Cr³⁺ heavy metal ions on the [2-AA]CoCl₃ adsorbent can be better understood using adsorption kinetic parameters. These parameters are typically derived from the pseudo-first order (LAGERGREN, 1898) and pseudo-second order kinetic models (Ho and Mckay, 1999) that are mathematically expressed by Lagergren rate equation (2) and Ho and McKay rate equation (3) respectively.

The q_t (mg/g) and q_e (mg/g) denote the amount of metal ions adsorbed at time t (min) and at equilibrium respectively. Similarly, the k₁ (min⁻¹) and k₂ (g/mg·min) denote the pseudo-first order and pseudo-second order rate constants, respectively. If the adsorption processes for the heavy metal ions by [2-AA]CoCl₃ follows the pseudo-first order kinetic model, the plot of $\ln (q_e-q_t)$ versus t(min) shown in Fig. SI-4(a) is expected to give a straight line with an intercept of $\ln q_e$ and slope of $-k_1$ . Similarly, if the adsorption processes proceeded via the pseudo-second order kinetics, then the plot of $\frac{t}{q_t}$ versus t(min) presented in Fig. SI-4(b) is expected to give a straight line with an intercept of $\frac{1}{k_2{q}_e^2}$ and slope of $\frac{1}{q_e}$ . Therefore, the obtained slopes and intercepts from the respective pseudo-first order and pseudo-second order plots are used to calculate the kinetic parameters $k_1$ , $k_2$ and theoretical $q_e(cal)$ that are summarized in Table 1. Consequently, the higher R² values for the pseudo-second order plots for all the heavy metal ions’ adsorption on the [2-AA]CoCl₃ indicates that the adsorption processes proceeded via the pseudo-second order pathway, thus the process is dominated by chemical interactions (Ho and Mckay, 1999). Based on this, it may be inferred that the interaction between the heavy metal ions and [2-AA]CoCl₃ is electrostatic, between electron rich sites of the adsorbent and the metal ions. In addition, the closeness of experimental adsorption capacity ${(q}_e(expt))$ with the pseudo-second order calculated adsorption capacity ${(q}_e(cal))$ corroborate that the adsorption processes proceeded via the pseudo-second order pathway.

3.2.6 Adsorption isotherms

The adsorption isotherm studies is carried out to gain some insight into the nature of the interaction between the [2-AA]CoCl₃ adsorbent and the heavy metal ions. In this regard, the Langmuir and Freundlich isotherms models expressed by linear equations (4) and (5) respectively were used.

$C_e\left(\frac{\mathit{mg}}{L}\right)$ and $q_e(\frac{\mathit{mg}}{g})$ denote the equilibrium concentration of adsorbate and adsorption capacity of the adsorbent, respectively. $K_L$ and $q_m$ denote Langmuir equilibrium constant and the maximum adsorption capacity of adsorbent respectively. $K_F$ and $n$ are Freundlich constants for a given adsorbate and adsorbent.

The Langmuir isotherm model is based on the assumption that the adsorption of the heavy metal ions occurs homogeneously on the surface of the [2-AA]CoCl₃ adsorbent using specific adsorption sites and energies. This model considered that steric hindrances and lateral interactions between adsorbate (heavy metal ions in this case) is not significant (Deng et al., 2009). Contrarily, the Freundlich isotherm model presumes that the adsorption of the heavy metal ions occurs heterogeneously, resulting to uneven distribution of adsorption energies over the [2-AA]CoCl₃ adsorption surfaces (Haghseresht and Lu, 1998). Therefore, by plotting $\frac{C_e}{q_e}$ against $C_e$ using equation (4), to get slope equal to $\frac{1}{q_m}$ and intercept of $\frac{1}{K_L{q}_m}$ (Fig. SI-5(a)), the $K_L$ which relates directly with adsorption energy and $q_m$ that signifies the monolayer adsorption capacity can both be determined. Similarly, the plot of $\ln q_e$ against $\ln C_e$ (using equation (5)) gives a slope equal to $\frac{1}{n}$ and intercept of $\ln K_F$ (Fig. SI-5(b)), and the constants $n$ and $K_F$ for the heavy metal ions’ adsorption are determined accordingly (J. Zeldowitsch, 1934). The adsorption isotherm parameters are summarized in Table 2, and the correlation coefficient values (R²) of both models is used to evaluate the model that best fit the nature of the interaction between the [2-AA]CoCl₃ adsorbent and the heavy metal ions. Consequently, the R² values of Langmuir adsorption isotherm nears one more than the R² values of Freundlich adsorption isotherm, thus the heavy metal ions adsorption isotherm is best described by the Langmuir model which implies that the adsorption occurs homogeneously on the surface of the [2-AA]CoCl₃ adsorbent. In addition, Langmuir model has shown that the [2-AA]CoCl₃ adsorbent has demonstrated a remarkable performance in the removal of these heavy metal ions with maximum adsorption capacity ( $q_m$ ) in the range of 227 – 357 mg/g. The adsorption capacity of Cd²⁺, Pb²⁺, As³⁺, and Cr³⁺ onto [2-AA]CoCl₃ adsorbent was compared with other adsorbents in recent published works (Table 3), and the [2-AA]CoCl₃ adsorbent performs quite well in the adsorption of Cd²⁺, Pb²⁺, As³⁺, and Cr³⁺ compared to the adsorbents earlier reported.

3.2.7 Selectivity of [2-AA]CoCl₃ for Cd²⁺, Pb²⁺, As³⁺, and Cr³⁺

The selectivity of [2-AA]CoCl₃ for Cd²⁺, Pb²⁺, As³⁺, and Cr³⁺ was studied by introducing 150 ppm each of Ca²⁺, K⁺, Na⁺ and Mg²⁺ as competitive ions in the sample solution containing 150 ppm each of Cd²⁺, Pb²⁺, As³⁺, and Cr³⁺. It was observed that even in the presence of other competitive metal ions, the [2-AA]CoCl₃ still exhibited superior adsorption performance for Cd²⁺, Pb²⁺, As³⁺, and Cr³⁺ (Fig. SI-6), with adsorption capacity in the range of 206 – 286 mg/g. On the other hand, the adsorption capacity of Ca²⁺, K⁺, Na⁺ and Mg²⁺ is within the range of 20 – 34 mg/g (Table 4). The selectivity coefficient ( $K_{M/M^{\prime }}$ ) of [2-AA]CoCl₃ for the targeted heavy metal ions (M) over other competitive metal ions (M’) is calculated using the equation (6) (Ganiyu et al., 2016):

Where $K_d$ signifies the distribution coefficient of the metal ions and is calculated using equation (7):

Large value of $K_d$ implies large adsorption of metal ion by adsorbent and vice-versa. Table 4 shows the $K_d$ values of all the ions, and it is observed that the Cd²⁺, Pb²⁺, As³⁺, and Cr³⁺ have larger $K_d$ values, especially As³⁺ ( $K_d$  = 460 mL/g) than the Ca²⁺, K⁺, Na⁺ and Mg²⁺ ions. Similarly, the adsorption capacity of the targeted metal ions is in the range of 206–286 mg/g while the adsorption capacity of the interfering ions is 20–34 mg/g. In addition, the selectivity coefficient $(K_{M/M^{\prime }}$ ) of each of the targeted metal ions is 2–3 orders of magnitude higher than those of the interfering metal ions. This further confirms the high selectivity of [2-AA]CoCl₃ for the targeted heavy metal ions.

3.2.8 Regeneration studies

Recycling of the adsorbent was studied to evaluate the loss of activity and possible reusability of the [2-AA]CoCl₃ MIL after it has been used for heavy metal ions’ adsorption. Thus, after the adsorption, the adsorbent was recovered for reuse by centrifuging at 3500 rpm. Then washed thoroughly with 1 M nitric acid at pH 3 until no residue of the metal ions was detectable in the supernatant solution. Finally, the adsorbent was dried at 50 °C before reuse. It was discovered that the [2-AA]CoCl₃ MIL could be reused six times with excellent adsorption efficiency of more than 93% (Fig. 7(a)). Interestingly, similar results were obtained when real wastewater and ground water resources were spiked with 150 ppm each of the targeted metal ions as shown in Fig. 7(b). This shows the practicality of the [2-AA]CoCl₃ MIL adsorbent in the selective removal of Cd²⁺, Pb²⁺, As³⁺, and Cr³⁺ metal ions in a complex matrices.

Close

Fig. 7

(a) Regeneration and reuse of MIL, [2-AA]CoCl₃ for 6 consecutive times for the adsorption of the metal ions Cd (II), As (II), Pb (II) and Cr(III) by using 1 M HNO₃; (b) Adsorption performance of [2-AA]CoCl₃ MIL adsorbent using real raw underground and wastewater samples spiked with 150 ppm each of the targeted metal ions.

Export to PPT