Efficient and Environmentally Friendly Adsorbent Based on β-Ketoenol-Pyrazole-Thiophene for Heavy-Metal Ion Removal from Aquatic Medium: A Combined Experimental and Theoretical Study.

A new sustainable and environmentally friendly adsorbent based on a β-ketoenol-pyrazole-thiophene receptor grafted onto a silica surface was developed and applied to the removal of heavy-metal ions (Pb(II), Cu(II), Zn(II), and Cd(II)) from aquatic medium. The new material SiNPz-Th was well characterized and confirms the success of covalent binding of the receptor on the silica surface. The effect of environmental parameters on adsorption including pH, contact time, temperature, and the initial concentration were investigated. The maximum adsorption capacities of SiNPz-Th for Pb(II), Cu(II), Zn(II), and Cd(II) ions were 102.20, 76.42, 68.95, and 32.68 mg/g, respectively, at 30 min and pH = 6. The adsorption isotherms, kinetics, and thermodynamic process were investigated and showed efficiency and selectivity toward Pb(II) and good regeneration performance. Density functional theory, noncovalent-interaction, and quantum theory of atoms in molecules calculations were used to study and to gain a deeper understanding of both the adsorption mechanism and selectivity of metal ions onto the adsorbent. Accordingly, metal ions such as Pb(II), Cu(II), and Zn(II) were bidentate coordinated with the adsorbent by nitrogen and oxygen atoms of the Schiff base C=N and hydroxyl group -OH, respectively, to form stable complexes. Whereas Cd(II) was coordinated in a monodentate fashion with oxygen atom of the hydroxyl group. Furthermore, the affinity of SiNPz-Th toward the metal ions was decreased in the order of Pb(II) > Cu(II) > Zn(II) > Cd(II), in good agreement with the experimental results. All these results highlight that SiNPz-Th has good potential to be an advanced adsorbent for the removal of lead ions from real water.

Introduction

In recent years, extenn class="Chemical">sive industrialization has generated severe environmental problems; water contamination by heavy metals is seriously hazardous to the aquatic environment. These metal ions can cause serious health hazard to humans, even at low concentrations.[1−4] Lead has been identified as one of the toxic elements which can cause serious diseases such as renal disturbances, hepatitis, anemia, cancer, and so forth.[5,6] A wide range of treatment technologies have been indicated for the removal of heavy metals from contaminated water in the last decades, starting from liquid–liquid extraction,[7] ion exchange,[8] chemical precipitation,[9] membrane filtration,[10] coagulation,[11] adsorption,[12−14] and ending with electrochemical treatment.[15] Among these approaches, adsorption is regarded as one of the very popular ones and captures enormous attention due to its high efficiency, simplicity, economy, and high environmental friendliness.[16,17] Hence, several of such adsorbents, such as graphene,[18] bentonite,[19] cellulose,[20] zeolite,[21] and so forth, were fabricated and applied to the elimination of heavy metals in aquatic systems.

Functional silica gel has attracted much interest due to its low-n class="Chemical">cost, feasibility, high thermal and mechanical stability, and large surface areas.[22,23] In addition, the possibility to introduce various organic ligands onto its surface to construct adsorbents that can effectively trap potential toxic metal ions[24−30] and its adsorption performance, which mainly depends on the nature of grafted ligands, are considered as the key success factors to better chelate with metal ions.[31−33]

In the past decades, several functional groups n class="Chemical">containing donor atoms, such as nitrogen, oxygen, and sulfur have been employed to construct adsorbents with high efficiency.[34−42]

In this context, n class="Chemical">pyrazole has received much attention for its highly interesting properties in coordination chemistry,[43,44] particularly for the synthesis of lead complexes.[45,47] On the other hand, β-ketoenol is an interesting compound which is known for its high complexing ability with most transition-metal ions.[48,50]

The reaction of an amine group with a ketone group leads to the formation of a Schiff base which is known for its excellent capacity to coordinate metal complexes.[51−53]

In the present study, an efficient adsorbent based on β-ketoenol-pyrazole-thiophene-functionalized silica gel (SiNPz-Th) was successfully fabricated and characterized.[46]

The adsorbent can be regarded as a low production n class="Chemical">cost, simple to prepare, and environmentally friendly means for potentially toxic-metal-ion removal. The impact of pH value, contact time, swelling kinetics, initial concentrations, adsorption isotherm, regeneration time, and selectivity of metal ion in the mixture were discussed and evaluated. On the other hand, density functional theory (DFT), noncovalent-interaction (NCI), and quantum theory of atoms in molecules (QTAIM) methods have been applied to gain a deeper understanding of the adsorption mechanism and selectivity of the metal ion toward the ligand structure.[49]

Results and Discussion

Linker Synthesis

The fabrication principle of our new adsorbent is shown in Scheme . The first stage n class="Chemical">concerns the preparation of the target (Z)-1-(1,5-dimethyl-1H-pyrazol-3-yl)-3-hydroxy-3-(thiophen-2-yl)prop-2-en-1-one ligand (L) in its stable conjugated enol tautomeric form. The structure of the synthesized ligand was suitable for X-ray, analytical, and spectroscopic studies. The second step involves the preparation of 3-aminopropyl-silica (SiNH) by condensation of silica with 3-aminopropyltrimethoxysilane. Finally, the target L molecule was anchored to NH2-groups to yield the new chelating adsorbent named SiNPz-Th (Scheme ).

Scheme 1

Synthetic Pathway to the Hybrid Material SiNPz-Th

L Crystal Structure

The L crystallizes in the orthorhombic system, space group P212121 (#19). The crystal structure shows a planar structure (Figure ). The molecule is planar with the O and S atoms located in the same semiplane and the N atom in an anti position. Despite possible conformational degrees of freedom, it is important to note the role of delocalization and conjugation. The β-ketoenol moiety has an inner hydrogen bond, which possibly shows dynamic features that may contribute to a disorder that affects the crystal parameters. The planar molecules are arranged in a zig-zag fashion to have a very compact packing.

Figure 1

X-ray molecular structure of L with atoms numbering scheme (L: a = 4.163(7) Å, b = 14.96(2) Å, c = 19.62(3) Å, V = 1222(3) Å3. Z = 4; T = 20 °C; Dcalc = 1.350 g/cm3; no. of reflections measured total: 7793, unique: 2775; R1 (I > 2.00σ(I)) 0.1744, R (all reflections) 0.3678, wR2 (all reflections) 0.5506; goodness of fit indicator 1.001).

X-ray molecular structure of L with atoms numbering scheme (L: a = 4.163(7) Å, b = 14.96(2) Å, c = 19.62(3) Å, V = 1222(3) Å3. Z = 4; T = 20 °C; Dcalc = 1.350 g/cm3; no. of refln class="Chemical">ections measured total: 7793, unique: 2775; R1 (I > 2.00σ(I)) 0.1744, R (all reflections) 0.3678, wR2 (all reflections) 0.5506; goodness of fit indicator 1.001).

Characterization Methods

The percentage of carbon, n class="Chemical">nitrogen, and sulfur contained in L covalently grafted onto silica surface was confirmed by elemental analysis. The hybrid material SiNPz-Th showed atomic percentage of C, N, and S of 4.6, 1.94, and 0.2%, respectively.

FT-IR spectra of free n class="Chemical">silica SiG, SiNH, and SiNPz-Th are presented in Figure . For SiG, signals at 3449, 1100, and 970 cm–1 correspond to Si–OH, Si–O–Si, and Si–O, which are characteristic adsorption peaks of silica. For SiNH, two adsorption peaks located at 2941 and 1560 cm–1 are assigned to CH and NH2 elongation vibrations, demonstrating the presence of propylamine groups onto the silica surface. The spectrum of SiNPz-Th exhibits two new peaks appearing at 1497 and 1583 cm–1 which are related to C=C and C=N stretching vibrations, respectively, indicating that L has been immobilized onto the silica surface successfully.

Figure 2

FTIR spectra of the SiG, SiNH, and SiNPz-Th.

Scanning electron min class="Chemical">croscopy (SEM) images of free silica (SiG) and hybrid material (SiPz-Th) are presented in Figure . It is clear that the surface morphology of SiNPz-Th is different from that of SiG. The surface of the hybrid material becomes rough, with a wide distribution of micrometer-sized particles. The distinct roughness of the surface indicates successfully functionalized silica with the β-ketoenol-pyrazole-thiophene ligand.

Figure 3

SEM micrographs of SiG (A), SiNH (B), and SiNPz-Th (C).

Nitrogen adsorption–desorption isotherms and the n class="Chemical">corresponding Barrett–Joyner–Halenda (BJH) pore-size distribution of SiG, SiNH, and SiNPz-Th were examined. The compounds under study exhibit typical IV isotherms with visible hysteresis loops (Figure ), indicating a mesoporous structure.[54] Furthermore, the SiNPz-Th displays a Brunauer–Emmett–Teller (BET) surface area, pore volume, and BJH adsorption average pore diameter of 310.11 m2/g, 0.53 cm3/g, and 45 Å, respectively, compared to SiNH which shows 283.08 m2/g BET surface area, 0.69 cm3/g pore volume, and 88.3 Å BJH adsorption average pore diameter, confirming a marked porosity in the structure. These results indicate the presence of active adsorption sites within the synthesized material.

Figure 4

Nitrogen adsorption–desorption isotherm plots of SiG, SiNH, and SiNPz-Th.

Thermogravimetric analyses (TGA) were recorded in the temperature ranging from 25 to 800 °C for n class="Chemical">SiG, SiNH, and SiNPz-Th (Figure ). SiG presents a weight loss of 3.15% from 25 to 110 °C that corresponds mainly to the loss of water.[55] The second stage of mass loss of 5.85% occurring from 110 to 800 °C is attributed to the condensation of silanol groups.[56] For SiNH there are two decomposition stages. The first one which starts in the range of 25–100 °C, presenting a mass loss of 1.56% was assigned to the dehydration process. The second one loses 7.22% mass at around 208–800 °C and was ascribed to the decomposition of propylamine immobilized onto silica. SiNPz-Th experienced a two-stage weight loss; the first in the range of 25–108 °C for about 2.77% which is due to water release. Ultimately, the highest weight loss of 11.55% occurring over the range 257–800 °C was ascribed to the decomposition of the β-ketoenol-pyrazole-thiophene ligand.

Figure 5

Thermogravimetric profiles of free silica SiG, SiNH, and SiNTh-Th.

Effect of pH on the Adsorption

The pH of a solution is the key factor that significantly influences adsorption characteristics. The impact of pH on the removal of n class="Chemical">Pb(II), Cu(II), Zn(II), and Cd(II) by SiNPz-Th is depicted in Figure . The adsorption capacity of Pb(II), Cu(II), Zn(II), and Cd(II) increased with an initial pH increase. At low pH, the adsorption capacity was fairly low because the functional groups of β-ketoenol-pyrazole-thiophene were protonated due to the presence of excess protons in the solution, which caused decrease in the adsorption capacities of metal ions as a result of the electrostatic repulsion between positive charge of metal ions and that of the protonated ligand. With the increase of pH, the number of H+ in the solution reduces and the protonation of functional groups decreases, leading to improved active binding sites being available for complexation of metal ions. Hence, the optimum adsorption is achieved at pH = 6. However, when pH is higher than 7, it leads to the precipitation of M(OH)2. This is consistent with the point of zero charge (PZC) which corresponds to the pH at which the surface charge is zero. Indeed, the PZC of initial silica was determined as pH = 2.3 by using a simple described method.[57] Surface coverage of the modified silica with an organic compound leads to an increase of the PZC of the silica surface from pH = 2.3 to pH = 7 due to the basicity of the ligand used. At this pHPZC, corresponding to the total deprotonation of the ligand, maximum sorption was observed.

Figure 6

Effect of solution pH on the removal of Pb(II), Cu(II), Zn(II), and Cd(II) by SiNPz-Th at optimum concentration (110 mg/L in each case) for 60 min at 25 °C.

Effect of Contact Time

The contact time is one of the most important factors to n class="Chemical">consider in the adsorption procedure. In this work, the contact time was evaluated over the range 0–35 min by mixing the adsorbent (10 mg) with 10 mL of solution at pH = 6 and 25 °C. As can be seen from Figure , when the contact time increased, the adsorption capacity increased gradually and then remained stable after about 20 min; in the first 5 min, more than 90% of the equilibrium adsorption was attained. These results indicate that the adsorption is very rapid and may mainly be due to the active site receptors which are sufficient to facilitate the quick sorption of Pb(II), Cu(II), Zn(II), and Cd(II) ions.

Figure 7

Pseudo-second-order model fit for the adsorption of Pb(II), Cu(II), Zn(II), and Cd(II) by SiNPz-Th. Adsorption conditions: V = 10 mL, m = 10 mg of adsorbent, pH = 6, and optimum concentration (110 ppm in each case) at 25 °C.

Kinetic Modeling

Experimental data were described by two kinetic models to explain the mechanism of control of metal-ion adsorption: the pseudo-first-order and pseudo-second-order models.[58,59]

The nonlinear form of the pseudo-first-order kinetic mathematical expression is expressed as followswhere qe (mg/g) and q (mg/g) are the equilibrium adsorption amount and the adsorption capacity over a period of time, and k1 (min–1) is the rate constant of the first-order-adsorption.

The nonlinear form pseudo-second-order model is given aswhere k2 (g/mg/min) is the pseudo-sn class="Chemical">econd-order adsorption rate constant.

The results were illustrated in Figure , whereas the fitting parameters of the two models are gathered in Table . A fine analysis suggests that the pseudo-sn class="Chemical">econd-order model is more suitable for describing the kinetic adsorption process of Pb(II), Cu(II), Zn(II), and Cd(II). Furthermore, the results obtained from this model indicate that a chemisorption mechanism adsorption, involving complexation reaction shared between the adsorbate and adsorbent, is effective.

Table 1

Kinetics Models Data for Metal Removal onto SiNPz-Th

	Pb(II)	Cu(II)	Zn(II)	Cd(II)
qexp (mg/g)	102.20	76.42	68.95	32.68
Pseudo-First-Order
qe (mg/g)	97.02 ± 2.23	73.73 ± 1.23	66.16 ± 1.21	31.58 ± 0.40
K1 (min–1)	0.34 ± 0.06	0.27 ± 0.02	0.27 ± 0.03	0.33 ± 0.03
R2	0.8	0.99	0.98	0.99
Pseudo-Second-Order
qe (mg/g)	104.16 ± 2.35	80.44 ± 0.76	72.31 ± 0.89	33.67 ± 0.28
K2 (g/mg/min)	0.006 ± 0.001	0.005 ± 4.69 × 10–4	0.006 ± 6.55 × 10–4	0.02 ± 0.001
R2	0.99	0.99	0.99	0.99

Effect of Concentration

The concentration effn class="Chemical">ect of the metal ions on adsorption is very important. It can provide basic information about interaction between adsorbate and adsorbent. The initial concentration of all metal ions varies from 10 to 300 mg/L using the batch method. The impact of concentration on adsorption of Pb(II), Cu(II), Zn(II), and Cd(II) on SiNPz-Th is displayed in Figure . It is clearly shown that the adsorbed quantity of metal ions increases in parallel with the increase in the concentration, which indicates that the adsorption promotes a high concentration which corresponds to a greater mass transfer driving force at a high concentration gradient. It is shown in Figure that the equilibrium adsorption of metal ions on SiNPz-Th was in the order of Pb(II) > Cu(II) > Zn(II) > Cd(II).

Figure 8

Langmuir and Freundlich adsorption model fits for Pb(II), Cu(II), Zn(II), and Cd(II) by SiNPz-Th. Adsorption conditions: 10 mg, V = 10 mL, 25 °C, and pH = 6 for Pb(II), Zn(II), Cu(II), and Cd(II) ions.

Adsorption Isotherms

Langmuir and Freundlich models are commonly employed to reveal the isotherm adsorption mn class="Chemical">echanism[60,61] in various fields. The Langmuir model assumes single molecular layer adsorption with the adsorption sites which are uniformly distributed evenly on the surface of the adsorbent. The nonlinear form of the Langmuir model can be expressed by the following equationwhere Ce (mg/L) is the equilibrium concentration of the solution; qe (mg/g) the amount adsorbed at equilibrium; q (mg/g) is the saturated adsorption capacity; and KL (L/mg) represents the Langmuir constant.

The Freundlich model is employed to describe the adsorption of an adsorbate on non-uniform surfaces with an adsorption multilayer. The Freundlich model is represented by the equationwhere n class="Chemical">qe (mg/g) represents the equilibrium adsorption capacity; KF (mg/L) and n are the Freundlich constants; and Ce (mg/L) is the equilibrium concentration of metal ions.

As displayed in Figure and Table , the adsorption capacities q calculated un class="Chemical">sing the Langmuir model 103.84 ± 2.44, 85.00 ± 1.07, 77.71 ± 2.28, and 35.91 ± 0.41 of Pb(II), Cu(II), Zn(II), and Cd(II), respectively, are close to the experimental data. As a result, the Langmuir model gave better fitting of Pb(II), Cu(II), Zn(II), and Cd(II) adsorption on SiNPz-Th compared to the Freundlich model, thus suggesting that the adsorption of metal ions corresponds to a one-layer adsorption process on the SiNPz-Th surface.

Table 2

Adsorption Isotherm Parameters for the Removal of Heavy Metals onto SiNPz-Th

	Pb(II)	Cu(II)	Zn(II)	Cd(II)
qexp (mg/g)	102.20	76.42	68.95	32.68
Freundlich Isotherm Model
n	4.51 ± 0.57	3.50 ± 0.49	4.53 ± 0.94	3.39 ± 0.29
KF (mg/L)	42.82 ± 3.82	25.33 ± 3.56	29.98 ± 4.741	9.66 ± 0.91
R2	0.90	0.90	0.84	0.97
Langmuir Isotherm Model
KL (L/mg)	0.437 ± 0.04	0.17664 ± 0.00943	0.22207 ± 0.03091	0.12128 ± 0.00654
q (mg/g)	103.84 ± 2.44	85.00 ± 1.07	77.71 ± 2.28	35.91 ± 0.41
R2	0.98	0.99	0.97	0.99

Adsorption Thermodynamics

To examine the impact of temperature on the adsorption efficiency of the n class="Chemical">SiNPz-Th adsorbent, Gibbs free energy (ΔG°), enthalpy change (ΔH°), and entropy change (ΔS°) can be calculated by the following equations[62]where R (8.314 J/mol/K) is the gas constant, M (g/mol) is the molar mass, T (K) is the temperature, Kd is the distribution coefficient, Kc is the equilibrium constant, C0 (mg/L) is the initial concentration of metal ion, and Ce (mg/L) is the equilibrium concentration of metal ion.

The results are given in Figure and in Table . The positive ΔH° value indicates the endothermic process of the adsorption. The pon class="Chemical">sitive ΔS° value reveals that entropy increases during the adsorption. The removal of Pb(II), Cu(II), Zn(II), and Cd(II) by SiNPz-Th is a spontaneous process according to the negative value of ΔG°.

Figure 9

Effect of temperature on the adsorption of metal ions onto SiNPz-Th.

Table 3

Adsorption Models Used in This Work and Their Associated Parameters

metal	ΔH° (kJ/mol)	ΔS° (J/K/mol)	T (K)	ΔG° (kJ/mol)
Pb(II)	08.16	133.59	299.15	–31.80
			309.15	–33.14
			319.15	–34.47
Cu(II)	08.93	126.38	299.15	–28.88
			309.15	–30.13
			319.15	–31.41
Zn(II)	08.31	125.99	299.15	–29.50
			309.15	–30.74
			319.15	–32.02
Cd (II)	03.67	103.96	299.15	–27.42
			309.15	–28.46
			319.15	–29.49

Adsorption Mechanism

DFT, NCI, and QTAIM caln class="Chemical">culations were applied to understand the adsorption selectivity and adsorption behavior difference between Pb(II), Cu(II), Zn(II), and Cd(II) in the presence of our hybrid material. Topology analysis of electron density in QTAIM was used to describe the electron density population for each atomic space of SiNPz-Th based on the delocalized and localized index measurements of electron density which is shared or exchanged between two atoms; the isosurface map of the dispersed electron density was also output. The uncommitted electron population (UEP), for example, the non-bonding region, is related to the atomic centers of SiNPz-Th that should be coordinated with an empty d orbital of the metal ion. Thus, it can be used as information to find the possible active sites of SiNPz-Th that can interact with a 3d metal ion, even if we do not pretend to fully describe the complexation which could involve solvent species. The Lewis structure, the UEP of the significant atoms, and the shape of electron density of the entire SiNPz-Th are displayed in Figure . From Figure , we note that the most preferred active interaction sites of SiNPz-Th are the oxygen atom (O34) of the hydroxyl group and the nitrogen atom (N18) of the imine function because their UEP are significant enough (>2.2 e) to establish orbital bonds with the metal ion. The situation differs for the two heteroatoms, such as sulfur S37 of the thiophene group and nitrogen N10 of the pyrazole, their UEP being not sufficient to create a novel orbital bond with the vacant d orbital (<1.3 e). In other words, despite the nitrogen atom N4 displaying an important UEP value (>3 e), a part of it should be delocalized in the pyrazole ring and continually this UEP will be renewed by its electro-donating methyl group substituent (Figure a). These findings were reinforced using one of the most conceptual DFT indices, namely the nucleophilic Parr functions (P–)[63] which is a powerful approach to evaluate the electron-donating ability of compound sites. These functions are obtained from the analysis of the Mulliken atomic spin density at the radical cation by removing an electron. These functions evidence again the most nucleophilic sites of SiNPz-Th which can interact with a metal ion (Figure b, right). It should be noted that a negative value of P– corresponds to a non-reactive site and consequently the most nucleophilic sites will correspond to the highest values of P–.

Figure 10

(a) (left) Schematic view of Lewis structure and significant values of UEP; (right) shape of electron density of the entire SiNPz-Th. (b) (left) NCI isosurface; (right) nucleophilic P—Parr functions of SiNPz-Th.

(a) (left) Schematic view of Lewis structure and significant values of n class="Chemical">UEP; (right) shape of electron density of the entire SiNPz-Th. (b) (left) NCI isosurface; (right) nucleophilic P—Parr functions of SiNPz-Th.

The NCI theory has been n class="Chemical">commonly used to characterize three types of inter/intramolecular interactions: strong attractive interactions (in blue), weak interactions (in green), and repulsive interactions (in red) as represented in Figure b (left). On one hand, both repulsive and attractive interactions are found between the two nitrogen atoms (N10···N18) and between hydrogen H and nitrogen N10 atoms (Ha···N10), respectively. On the other hand, a weak interaction exists between the thiophene sulfur atom S37 and hydrogen atom Hb (S37···Hb). In contrast, no interactions were found between the hydroxyl oxygen atom O34 and the imine nitrogen atom N18, which supports the results obtained from DFT and QTAIM analysis, which suggested that these two atoms are responsible for the binding ability of SiNPz-Th to the metal ion.

Based on these computations, the posn class="Chemical">sible optimized structures of mononuclear coordination complexes of M(II) (M = Pb, Cu, Zn, and Cd) with SiNPz-Th are presented in Figure . In brief, SiNPz-Th coordinates with Pb(II), Cu(II), and Zn(II) using an oxygen atom O34 and a nitrogen atom N18 to form stable complexes, whereas Cd(II) is only bonded by an oxygen atom O34 (Figure ). Thus, SiNPz-Th adopts a bidentate coordination mode for Pb(II), Cu(II), and Zn(II) ions, whereas a monodentate mode if found for Cd(II) ion. Interestingly, we note that when Cd(II) interacts with the oxygen atom O34 of SiNPz-Th, the hydrogen atom of the hydroxyl group (−OH) is easily captured by the nitrogen atom N18; and thus weakens the coordination of Cd(II) with the ligand. Thus, coordination is weaker for Cd(II) compared to other metal ions, a situation confirmed by experimental results which show the weakest adsorption for Cd(II) ions.

Figure 11

Optimized geometry of complexes formed between reactive atoms of the ligand/water and metal ions (water(2)···M(II)···ligand).

Figure displays the isodensity surface plots of the highest ocn class="Chemical">cupied molecular orbital (HOMO) analysis. This calculation reflects also the chemical reactivity related to the lone paired electron transport from the ligand to the empty d orbital of the metal ion. As shown in Figure , the HOMO is not localized at the pyrazole and thiophene rings, which suggests that the nitrogen atoms of pyrazole and sulfur atom thiophene are not reactive toward the metal ion. In other words, the electron density of HOMO for Pb(II), Cu(II), and Zn(II) complexes are mainly located on the oxygen atom O34, nitrogen atom N18, and the adjacent carbon atoms. Whereas, for the Cd(II) complex, the HOMO is nearly spread at the oxygen atoms of complexed water and not over O34 and N18 atoms of the ligand. This situation means that the coordination with Cd(II) is commonly generated from the water molecules and not from the ligand. In other words, the degree associated to the contribution of HOMO for the concerned complexes (Figure ) can be classified as follow: Pb(II)-complex > Cu(II)-complex > Zn(II)-complex > Cd(II)-complex. This could be a good indication for the coordination selectivity of the metal ion toward the studied ligand. This selectivity will be confirmed in the next part in terms of the natural bond orbitals analysis.

Figure 12

Plot of HOMO for our M(II) complexes.

As follows, a natural bond orbital (NBO) analysis[64] for the studied coordination complexes has been undertaken to inspect selectivity of adsorption for Pb(II), Cu(II), Zn(II), and Cd(II); hence, strengths of the studied interactions N18 → M(II) and O34 → M(II) have been further evaluated from second-order stabilization energy E(2) and the electronic configurations (EC) of M(II), O34, and N18. As a large value of E(2) means a more intensive donor–acceptor interaction, these are considered as a good representation of the bond strength. The NBO analysis calculations, listed in Table , indicate that the valence ECs of O34 and N18 which are associated to the formed complexes (Figure ) are ranged as follows: EC(Cd(II)) < EC(Zn(II)) < EC(Cu(II)) < EC(Pb(II)). This supports that the charge transfer of adsorption takes place from O34 and N18 to the empty orbitals of Pb(II), Cu(II), and Zn(II), whereas for the Cd(II) complex, the charge transfer proceeds from O34 and N18 to the empty orbitals of both Cd(II) and the transferred hydrogen atom of hydroxyl group of the ligand, respectively. The calculated stabilization energies E(2) associated to the studied interactions are listed in Table .

Table 4

NBO Analysis Calculations for the M(II) Complex (M: Pb, Cu, Zn and Cd)

		E(2) energy (kcal/mol)
	EC of M, O34 and N18	LP(N18) → LP*(M)	LP(O34) → LP*(M)
Pb(II) complex	Pb:[core]6s1.836p1.20	88.33	25.54
	O34:[core]2s0.212p1.113p0.00
	N18:[core]2s0.242p1.713p0.00
Cu(II) complex	Cu:[core]4s0.173d4.934p0.06	53.24	13.23
	O34:[core]2s0.842p2.583p0.01
	N18:[core]2s0.682p2.103p0.01
Zn(II) complex	Zn:[core]4s1.963d9.994p0.07	35.64	10.68
	O34:[core]2s1.582p5.013p0.01
	N18:[core]2s1.412p5.093p0.01
Cd(II) complex	Cd:[core]5s1.984d10.005p0.03		4.32
	O34:[core]2s1.742p5.123p0.01
	N18:[core]2s1.472p4.313p0.01

These results show that the interaction strength ligand → M(II) follows the tendency: n class="Chemical">Pb(II) > Cu(II) > Zn(II) > Cd(II). To this end, we note that the experimental selectivity (Figure ) is completely reproduced by the present computational study.[65−73]

Figure 13

Selectivity of SiNPz-Th for Pb(II) removal with initial concentration of 150 mg/L for metal ions.

Adsorption Selectivity of SiNPz-Th

Competitive adsorption of heavy-n class="Chemical">metal ions on SiNPz-Th was investigated using quaternary mixed solutions (Pb(II), Cu(II), Zn(II), and Cd(II)) using the batch method under optimum conditions (Figure ). The SiNPz-Th material presents a higher selectivity toward Pb(II) ions compared to other common divalent ions such as Cu(II), Zn(II), and Cd(II) ions. The high selectivity of SiNPz-Th for Pb(II) should be ascribed to the β-ketoenol-pyrazole-thiophene functional groups bonded onto the silica surface, which can coordinate with Pb(II) to form more stable complexes. The reason of the high selectivity adsorption of Pb(II) maybe due to the large ionic radius and the electronegativity of lead. It is clear that the SiNPz-Th is capable of selectively adsorbing Pb(II) ions from the mixture of several metal ions in wastewater.

Effect of Electrolyte

In the actual environmental pollution, there are a variety of foreign ions, which can reduce the adsorption efficiency. It is therefore essential to investigate the impact of existing ions in the solution onto the sorption of n class="Chemical">Pb(II) ions. The experiment of Pb(II) sorption was thus performed in the presence of four co-existing ions (K+, Na+, Ca2+, and Mg2+) examined under optimum conditions. Table represents the effect of adsorption capacity of Pb(II) (C0 = 0.05 μg/mL) onto SiNPz-Th in the presence of co-existing ions. It is clear that the adsorption efficiency was maintained more than 98%, thus suggesting that the adsorption of Pb(II) ions is not weakened by the presence of co-existing ions. This experimental study of co-existing ions confirms that the adsorbent SiNPz-Th presents higher adsorption capacity toward Pb(II) and can potentially be applied to a real sample that contains various ions.

Table 5

Effect of Interfering Ions on the Recovery of Pb(II) Ions Adsorbed on the SiNPz-Th Sorbent (Concentration of Pb(II) Ion is 0.05 μg/mL)

interfering ion	concentration (μg/mL)	recovery of Pb(II) (%)
K+	3000	97.69
Na+	3000	99.15
Ca2+	2000	96.32
Mg2+	2000	95.12

Desorption and Recycling

Reusability is one of the crucial indicators for assesn class="Chemical">sing whether an adsorbent is practicable or not. Desorption of the adsorbent was examined by adding 2 mol/L HCl to 10 mg of SiNPz-Th after having absorbed Pb(II). The suspension was stirred for 60 min. Then, the SiNPz-Th was neutralized by a NaOH solution and dried in vacuum for the next adsorption. Table represents the removal efficiency of SiNPz-Th after five cycles for Pb(II) ions. Interestingly, the adsorbent retained more than 96% of their adsorption capacity. Thus, SiNPz-Th presents outstanding recyclability and applicability which could be useful for the purification of lead contaminated water or the handling of wastewater.

Table 6

Adsorption/Regeneration of Hybrid Material Toward Pb(II)

cycle number	Pb(II)(mg/g)
1	102.27
2	102.19
3	101.21
4	100.49
5	100.07

Adsorption of Pb(II) from Real Water Samples

The feasibility of n class="Chemical">SiNPz-Th for elimination of Pb(II) ions from field water samples was tested by mixing 10 mg of the adsorbent for 10 mL of water followed by the addition of 0.5 mL of HNO3 at 25 °C. The choice was focused on two types of river water samples selected from the Oriental Morocco area: (i) Ghiss river (located next to Al Hoceima) where pH = 7.7, total dissolved solids (TDS) = 1297 mg/L, and conductivity (σ) = 1733 μS/cm. (ii) Toussit-Bou-Beker river (in the Jerada-Oujda region) where pH = 7.1, TDS = 2031 mg/L, and σ = 2301 μS/cm. The water samples were filtered with a 0.45 μm nylon membrane to remove large particle-size impurities. As the designated real water samples did not enclose any Pb(II) ions, which is the metal ion for which our SiNPz-Th adsorbent would be the most efficient, the samples had to be doped with Pb(II) ions. The choice was made on 5 and 10 mg/L of Pb(II). Clearly, the removal efficiency of Pb(II) ions was high up to 96% (Table ). This result clearly indicates that our developed SiNPz-Th adsorbent is more effective in purification of real water and could be used as an excellent candidate in water pollution treatment.

Table 7

Analysis of Pb(II) in Real Water Samples

water samples	added Pb(II) (mg/L)	removal efficiency (%)
Touissit-Bou-Beker water	5	96.34
	10	94.67
Ghiss water	5	96.65
	10	95.98

Comparison with Other Adsorbents

A comparison of the n class="Chemical">Pb(II) adsorption performance of SiNPz-Th with other adsorbents reported is useful to place our hybrid material’s properties in perspective (Table ). Indeed, the SiNPz-Th adsorbent shows greater maximum adsorption capacity for Pb(II) ions (102 mg/g) compared to the previous silica hybrids; it is also higher when comparison is made with known commercialized activated carbons. This clearly indicates that the SiNPz-Th adsorbent could be considered as a useful solution in Pb(II) remediation applications.

Table 8

Comparison of the Maximum Adsorption Capacities of Pb(II) by Different Adsorbents Reported in the Literature

support: silica gel/ligand	references	metal ion (mg/g)
SiNPz-Th	this work	102.27
free silica (SiG)		04.32
propylamine (SiNH2)		06.33
porphyrin	(38)	55.17
bispyrazole	(39)	35.26
TOES	(65)	75.60
dithiocarbamate	(66)	42.19
2-hydroxy-3-methoxy benzaldehyde	(67)	02.27
3-(2-aminoethylamino)propyl	(68)	19.61
3-mercaptopropyl	(69)	32.58
PMAEEDA	(70)	61.90
commercial activated carbon (CS-1501)	(71)	25.90
commercial activated carbon (RS-1301)	(71)	35.43
commercial activated carbon (NC-60)	(71)	25.30
commercial activated carbon (aktivkohle)	(72)	23.42
biochar	(73)	37.82

Conclusions

A new sustainable and environmentally friendly hybrid inorganic–organic n class="Chemical">adsorbent including a novel β-ketoenol-pyrazole-thiophene receptor has been prepared and used for the removal of heavy-metal ions (Pb(II), Cu(II), Zn(II), and Cd(II)) from water samples. Superior characteristics were observed for Pb(II) ions with an adsorption capacity not less than 102.20 mg/g at 30 min and pH = 6, coupled to an excellent recyclability up to minimum five cycles. The adsorbent obtained has extremely high adsorption capacity and selectivity of Pb(II) from water. Insights to the adsorption and selectivity mechanism were gained by a computational study which supported metal-ion coordination following a bidentate mode for Pb(II), Cu(II), and Zn(II) and a monodentate mode for Cd(II). In addition, the affinity of our hybrid material toward Pb(II) was confirmed by our computational study. All these results highlight that our hybrid material has a good potential to be an advanced adsorbent for lead ion removal in aquatic media.

Experimental Section

Materials and Methods

All reagents (Aldrich, purity 99.5%) were of analytical grade. Initial n class="Chemical">silica gel (60 Å, 70–230 mesh) (E. Merck) was activated at 120 °C for 24 h. The quantification of metal ions in aqueous solutions was determined by atomic absorption (Spectra Varian A.A. 400 spectrophotometer). The surfaces were characterized by a CHN analyzer (Microanalysis Center Service, CNRS), Fourier-transform infrared spectroscopy (FTIR, Perkin Elmer System 2000), SEM (FEI-Quanta 200), TGA (Perkin Elmer Diamond), 13C solid state nuclear magnetic resonance (NMR, CP MAX CXP 300 MHz), and BET (ThermoQuest Sorpsomatic 1990 analyzer).

Preparation of (Z)-1-(1,5-Dimethyl-1H-pyrazol-3-yl)-3-hydroxy-3-(thiophen-2-yl)Prop-2-en-1-one (L)

Ethyl 1,5-dimethyl-1H-pyrazole-3-carboxylate (2 g, 11.89 mmol) was dissolved in n class="Chemical">toluene (25 mL). Metallic sodium (0.4 g, 17.39 mmol) was added to this solution with stirring; then, a solution of 2-acetylthiopene (1.5 g, 11.89 mmol) in toluene (3.5 mL) was slowly added at 0 °C; the reaction mixture was kept under stirring for about 3 days at 25 °C. The mixture was filtered, the residue was washed in toluene (15 mL), dissolved in water (10 mL), and treated with acetic acid (1 mL) to pH = 5. The product obtained was removed with CH2Cl2 (3 × 5 mL). The resulting product was recrystallized from methanol (1 mL) leading to single crystals suitable for X-ray diffraction analysis. Yield: 0.89 g, 30%; mp (hot MeOH) = 110 °C; R 0.42 (CH2Cl2/MeOH 9:1) silica; FTIR (KBr, cm–1) ν(OH): 3434; ν(C=O): 1672; ν(C=C enolic): 1531; 1H NMR (DMSO, δ (ppm)): 2.30 (s, 3H, Pz-CH3); 3.84 (s, 3H, CH3–N); 4.56 (s, 0.1H, keto, CH2); 6.58 (s, 1H, Pz–H); 6.80 (s, 0.9H, enol, C–H); 7.13 (m, 1H, Th-Hβ); 7.60 (d, 1H, Th-Hγ); and 7.79 (d, 1H, Th-Hα). 13C NMR (DMSO, δ (ppm)): 11.41 (1C, Pz-CH3); 37.00 (1C, CH3–N); 46.76 (1C, keto CH2); 92.89 (1C, enol C–H); 106.11 (1C, CH-Pz); 128.32 (1C, Th-Hγ); 130.20 (1C, Th-Cβ); 132.16 (1C, Th-Cα); 140.46 (1C, Th-Cε); 177.87 (1C, C–OH); and 180.96 (1C, C=O). m/z: (M + H)+ 249. Anal. Calcd for C12H12N2O2S: C, 58.05; H, 4.87; N, 11.28. Found: C, 58.02; H, 4.99; N, 11.31.

Preparation of Amine-Functionalized Silica (SiNH)

Amine-functionalized n class="Chemical">silica SiNH was constructed following a similar procedure reported earlier.[31,32] A suspension of activated silica gel (30 g) and 10 mL of 3-aminopropyltrimethoxysilane were fully mixed in 200 mL dry toluene in the presence of N2. Then, the reaction mixture was refluxed for 24 h, the material SiNH was obtained after filtration and Soxhlet extraction by anhydrous methanol for 12 h, and finally dried under vacuum at 60 °C for 24 h.

Fabrication of the SiNPz-Th Adsorbent

A mixture of L (4 g) and SiNH (5 g) in methanol (30 mL) was stirred, refluxed for 12 h. Then, the solid was obtained after separation and Soxhlet extraction with the solvent mixture MeOH/CH2Cl2 for 12 h. Finally, the obtained modified silica was dried in an oven at 60 °C for 24 h.

Batch Adsorption Experiments

For studying the adsorption properties of SiNPz-Th, the batch experiment was evaluated following the impact of the pH, initial n class="Chemical">concentrations, contact time, and adsorption temperature. The adsorption experiment of SiNPz-Th toward Pb(II), Cu(II), Zn(II), and Cd(II) was investigated by stirring 10 mg of the adsorbent with 10 mL of a metal salt solution (Pb(NO3)2·6H2O, Cu(NO3)2·3H2O, Zn(NO3)2·6H2O, and Cd(NO3)2·6H2O) in 50 mL tubes at room temperature. The pH values of Pb(II), Cu(II), Zn(II), and Cd(II) solutions were adjusted from 1 to 7 using dilute HCl or NaOH solutions. Meanwhile, the adsorption isotherm experiments were studied with initial concentrations ranging from 10 to 300 mg/L for all metal ions with pH = 6 at 25 °C for 60 min. The effect of contact time was studied from 5 to 60 min at 25 °C. The adsorption thermodynamics were investigated at different temperatures (25, 35, and 45 °C) for 60 min. After that, the concentration of each metal ion was determined by atomic adsorption measurements, and the adsorption capacity was calculated according to the following equation.[74]where qe (mg/g) is the adsorption amount, C0 (mg/L) is the initial concentration, Ce (mg/L) is final equilibrium concentrations, V (L) is the volume of the solution, and W (g) represents the weight of adsorbent.

Computational Methods

The selectivity and mn class="Chemical">echanism of metal adsorption were further supported by DFT,[75] QTAIM,[76] and NCI[77] analysis using GAUSSIAN 09[78] and Multiwfn[79] software. Optimization of the studied entities such as the ligand, metal ions M(II) (Pb(II), Cu(II), Zn(II) and Cd(II)), and M(II)-complexes were performed using DFT based on Beck’s three parameter exchange functional and Lee–Yang–Parr nonlocal correlation functional (B3LYP),[80] combined to the 6-311G++(d,p) basis set,[81] for carbon, hydrogen, nitrogen, sulfur, and nitrogen atoms and to the basis LANL2DZ level[82] for the metal. The intramolecular interactions of SiPz-Th were evaluated in terms of NCI analysis. This approach is based on the relationship between the electron density ρ(r) and the reduced density gradient s and is given as follows

It allows to give isosurfaces of s at low densities and thus to visualize the pon class="Chemical">sition and nature of noncovalent interactions in the 3D space (whichever repulsive, van der Waals, attractive, or all). This method has also been recently used to afford a more comprehensive understanding of noncovalent bonding. Furthermore, the strength of bonding interactions has been estimated based on the second-order stabilization energy E(2) calculations,[83,84] which are related in this work to the delocalization of lone pair orbitals (LP) of SiNPz-Th to antibonding orbitals (LP*) of metal ions.

Single Crystal X-ray Crystallography

Single-crystal X-ray diffraction data for L were collected on a Rigaku Rapid Axis II-IP-plate, 5 kV, curved detector diffractometer with Mo Kα (λ = 0.71075 Å) radiation. The crystal structure was solved by direct methods,[85] expanded using Fourier techniques and refined by the full-matrix least-squares technique [least squares function minimized: (SHELXL97) w(Fo2 – Fc2)2 where w = least squares weights] on F2 using the SHELXL package.[86] All non-hydrogen atoms were refined anisotropically. Hydrogen atoms were refined using the riding model. All calculations were performed using the Crystal Structure crystallographic software package,[87] except for refinement, which was performed using the SHELXL program. Crystallographic data for the structure reported in this paper was deposited in the Cambridge Crystallographic Data Center with CCDC reference number 1984476. Identical results were obtained on a XtaLAB Synergy-Rigaku Oxford-4 circle diffractometer. The structure was measured and solved using CrysAlisPro software.