Comparative Study of Synthesis Methods to Prepare New Functionalized Adsorbent Materials Based on MNPs–GO Coupling

In this work, the synthesis of new adsorbent nanomaterials based on the coupling of magnetic nanoparticles and graphene oxide (MNPs–GO) was addressed. Separately, MNPs and GO have adsorbent properties of great interest, but their use involves certain difficulties. The coupling seeks compensation for their disadvantages, while maintaining their excellent properties. Three different routes to synthesize coupled MNPs–GO were studied and are compared in this work. The three synthesized materials were functionalized with chelating groups: [1,5-bis (di-2-pyridyl) methylene] thiocarbonohydrazide (DPTH) and [1,5-bis(2-pyridyl)3-sulfophenylmethylene] thiocarbonohydrazide (PSTH). The new adsorbent nanomaterials were characterized adequately. Moreover, their capacities of adsorption toward heavy and noble metals were determined, in order to apply them as extractants in magnetic solid-phase extraction to preconcentrate metals in environmental samples. The results showed that one of the routes provided nanomaterials with better adsorbent characteristics and higher yields of functionalization.


Introduction
Heavy metals are a group of chemical agents widely studied in pollution [1]. Heavy metals arise from rocks weathering and oil and carbon combustion [2]. This group is found in nature in trace amounts (concentration lower than 100 mg/L). The heavy metals hazard lies in the difficulty to degrade them, producing bioaccumulation. Platinum-group elements (PGEs: Pt, Pd, Ir, Ru, Os and Rh) and other noble metals, such as Au and Ag, are known for being nonreactive elements. Their presence arises from their wide usage in jewellery, industry and catalysis, for example, in engines of vehicles. Nowadays, these catalysts are a new source of pollution [3]. In ecosystems, these metals can become toxic substances in the presence of certain substances, such as chlorides. Moreover, PGEs are transported to long distances in particles form, appearing in a huge variety of locations (Alpine snow, urban environments, etc.) [4,5]. The development of effective methods for elimination and determination of heavy metals [6] and noble metals is necessary. Moreover, many analytical methods use solid-phase extraction (SPE) as sample pretreatment because it offers high enrichment factors, complete recovery, rapid phase separation and possible automation with different detection techniques [7]. The SPE methods use less sample volume and reagents [8] and present three successive basic steps: First, the adsorbent is conditioned with a solvent. Then, the loading solution is introduced through the solid phase, to retain the analytes, and, finally, the analytes are collected with an appropriate eluent solution [9]. the concentrations for the samples were 0.5% (wt/wt) approximately. X-ray photoelectron spectroscopy (XPS) analysis were performed with a Physical Electronics ESCA 5701 instrument (Chanhassen, MN, USA); binding energies (BE) were observed, considering the position of the C 1s peak at 284.8 eV. The residual pressure in the analysis chamber was maintained below 3 × 10 −9 Torr during data acquisition. The microstructures of the new materials were observed and studied by transmission electron microscopy (TEM) JEOL, JEM-1400 (Peabody, MA, USA) and N 2

Preparation of Graphene Oxide
The exfoliation process described by Diagboya et al. was used to prepare GO from natural graphite, being purified by centrifugation cycles [28].

Method 1
In this method, first, MNPs were synthesized, covered with silica and functionalized; second, the coupling with GO was addressed.
Synthesis of MNPs with silica coating: FeCl 3 ·6H 2 O (11.68 g) and FeCl 2 ·4H 2 O (4.30 g) were mixed in 200 mL of deionized water at 80 • C, and then 50 mL of 30% ammonia solution was added quickly, for the coprecipitation of Fe 3 O 4 nanoparticles in N 2 atmosphere. The suspension was stirred with reflux for 75 min. The Fe 3 O 4 solution was cooled to room temperature and separated from the solution with the aid of external permanent magnet and washed with deionized water. The synthesis of MNPs was optimized by González Moreno et al. [29], the prepared MNPs were mixed with 8 mL of TEOS and 60 mL of glycerol in 200 mL of ethanol at 60 • C and stirred for 2 h in N 2 atmosphere. The resultant suspension was cooled to room temperature and separated from the solution with the aid of external permanent magnet, washed with deionized water and kept in ethanol for further functionalization ( Figure 1). permanent magnet, washed with deionized water and kept in ethanol for further functionalization ( Figure 1). Coupling MNPs-GO: This step was made after the functionalization of MNPs, and it was based on the work of Nodeh et al. [17]. The functionalization process was described elsewhere [26,30]. For the coupling, 500 mg of GO and 500 mg of functionalized MNPs were mixed in water and stirred for 10 min. The mix was incubated at 65 °C for 8 h. The coupling Fe3O4-GO suspension was cooled to room temperature and separated from the solution with the aid of external permanent magnet, and then washed with deionized water and ethanol. Finally, the precipitate was dried at 80 °C for 24 h. The obtained materials were called 1-DPTH and 1-PSTH.

Method 2
GO was made magnetic by slowly adding a 35 mL aqueous solution of mixed FeCl3·6H2O (1.10 g) and FeCl2·4H2O (0.43 g) into the GO suspension (300 mg of GO in 200 mL deionized water) at 80 °C, and then 5 mL of 30% ammonia solution was added quickly, for the co-precipitation of Fe3O4 nanoparticles on GO sheets in N2 atmosphere. After being rapidly stirred for 45 min, the resultant Fe3O4 decorated GO suspension was cooled to room temperature and separated from the solution with the aid of external permanent magnet, washed with deionized water and kept in ethanol for further functionalization [7].
MNPs-GO-EDA Coupling: The prepared material was suspended in 300 mL of deionized water and sonicated for 15 min; then, 3 g of NaOH and 3 g of Cl-CH2COONa were added, and the mix was stirred for 2 h [31]. The modified MNPs-GO was separated from the suspension by using an external permanent magnet and washed first with 1% HCl and finally with deionized water. This modification was included to increase the number of -COOH functional groups available over the surface, and, thus, the number of the active sites on the GO surface, by transforming the -OH and epoxide groups into functionalizable -O-CH2COOH groups, as is shown in Figure 2. The MNPs-GO-EDA was fabricated "in situ" by coupling of EDA with carboxylic groups of modified MNPs-GO using DCC as a coupling agent. The 1.2 mL of EDA was added into the modified MNPs-GO suspended ethanol solution, along with DCC. The suspension was stirred for 48 h at 50 °C. The -COOH functional groups of the modified MNPs-GO condense with one of the EDA amine groups to form amide bonds, while the second amine group was available to attach an organic group (-CONH-C2H4-NH2). Finally, MNPs-GO-EDA was separated from the suspension, using an external permanent magnet, and washed with deionized water. Coupling MNPs-GO: This step was made after the functionalization of MNPs, and it was based on the work of Nodeh et al. [17]. The functionalization process was described elsewhere [26,30]. For the coupling, 500 mg of GO and 500 mg of functionalized MNPs were mixed in water and stirred for 10 min. The mix was incubated at 65 • C for 8 h. The coupling Fe 3 O 4 -GO suspension was cooled to room temperature and separated from the solution with the aid of external permanent magnet, and then washed with deionized water and ethanol. Finally, the precipitate was dried at 80 • C for 24 h. The obtained materials were called 1-DPTH and 1-PSTH.

Method 2
GO was made magnetic by slowly adding a 35 mL aqueous solution of mixed FeCl 3 ·6H 2 O (1.10 g) and FeCl 2 ·4H 2 O (0.43 g) into the GO suspension (300 mg of GO in 200 mL deionized water) at 80 • C, and then 5 mL of 30% ammonia solution was added quickly, for the co-precipitation of Fe 3 O 4 nanoparticles on GO sheets in N 2 atmosphere. After being rapidly stirred for 45 min, the resultant Fe 3 O 4 decorated GO suspension was cooled to room temperature and separated from the solution with the aid of external permanent magnet, washed with deionized water and kept in ethanol for further functionalization [7].
MNPs-GO-EDA Coupling: The prepared material was suspended in 300 mL of deionized water and sonicated for 15 min; then, 3 g of NaOH and 3 g of Cl-CH 2 COONa were added, and the mix was stirred for 2 h [31]. The modified MNPs-GO was separated from the suspension by using an external permanent magnet and washed first with 1% HCl and finally with deionized water. This modification was included to increase the number of -COOH functional groups available over the surface, and, thus, the number of the active sites on the GO surface, by transforming the -OH and epoxide groups into functionalizable -O-CH 2 COOH groups, as is shown in Figure 2. The MNPs-GO-EDA was fabricated "in situ" by coupling of EDA with carboxylic groups of modified MNPs-GO using DCC as a coupling agent. The 1.2 mL of EDA was added into the modified MNPs-GO suspended ethanol solution, along with DCC. The suspension was stirred for 48 h at 50 • C. The -COOH functional groups of the modified MNPs-GO condense with one of the EDA amine groups to form amide bonds, while the second amine group was available to attach an organic group (-CONH-C 2 H 4 -NH 2 ). Finally, MNPs-GO-EDA was separated from the suspension, using an external permanent magnet, and washed with deionized water.

Method 3
In this method, double coupling and double functionalization were achieved by anchoring the functional groups with both MNPs and GO.
MNPs-GO Coupling: 500 mg of GO was suspended in 50 mL of ethanol, with 500 mg of coated MNPs (synthesized as described in the first step of Method 1) and 0.25 g of DCC in a 100 mL roundbottom flask. The mixture was sonicated for 10 min and kept at reflux at 50 °C for 48 h. In this way, part of the MNPs were covalently anchored to the GO sheet through an amide bond, which is formed by condensation between the acid groups of the GO surface and the amine groups of the MNPs. The rest of MNPs were physically adsorbed on GO surface due to two main factors: performance of the reaction of covalent MNP-GO coupling, which is supposed to be <100%, and the depletion of -COOH functional groups over the GO surface during this synthesis step. Therefore, the MNPs were coupled to the GO through two mechanisms: chemical interactions (covalent binding by condensation with the carboxylic acid groups) and physical interactions (electrostatic and Van der Waals forces) with the GO layer.
Modification of MNPs-GO: the solid MNPs-GO from the previous stage was suspended in 50 mL of deionized water. The mixture was sonicated for 15 min, and then 50 mL of NaOH 2.5 M was added. Next, 5 g of sodium acetate chloride (Cl-CH2COONa) in 50 mL of deionized water was introduced, and the mixture was maintained for 2 h, in ultrasound, at room temperature. The suspension and the matrix were separated with the aid of an external permanent magnet, and the solid was washed twice, first with 1% HCl and finally with deionized water. This modification increases the number of the active sites on the GO surface, by transformation of -OH and epoxide groups into functionalizable -O-CH2COOH groups.
EDA-MNPs-GO Coupling: The modified MNPs-GO solid from the previous step was suspended in 50 mL of ethanol, together with 4 mL of EDA and 0.25 g of DCC, in a 100 mL round-bottom flask, and stirred at 50 °C for 48 h. The free acid groups obtained in the previous step were condensed with one of the EDA amine groups to form amide bonds, while the second was available to attach an organic group to functionalize the solid. Thus, there were double-origin amino groups available for functionalization: the amino group of the coated MNPs dispersed on the sheet and the free noncondensed amino group of the EDA.

Method 3
In this method, double coupling and double functionalization were achieved by anchoring the functional groups with both MNPs and GO.
MNPs-GO Coupling: 500 mg of GO was suspended in 50 mL of ethanol, with 500 mg of coated MNPs (synthesized as described in the first step of Method 1) and 0.25 g of DCC in a 100 mL round-bottom flask. The mixture was sonicated for 10 min and kept at reflux at 50 • C for 48 h. In this way, part of the MNPs were covalently anchored to the GO sheet through an amide bond, which is formed by condensation between the acid groups of the GO surface and the amine groups of the MNPs. The rest of MNPs were physically adsorbed on GO surface due to two main factors: performance of the reaction of covalent MNP-GO coupling, which is supposed to be <100%, and the depletion of -COOH functional groups over the GO surface during this synthesis step. Therefore, the MNPs were coupled to the GO through two mechanisms: chemical interactions (covalent binding by condensation with the carboxylic acid groups) and physical interactions (electrostatic and Van der Waals forces) with the GO layer.
Modification of MNPs-GO: the solid MNPs-GO from the previous stage was suspended in 50 mL of deionized water. The mixture was sonicated for 15 min, and then 50 mL of NaOH 2.5 M was added. Next, 5 g of sodium acetate chloride (Cl-CH 2 COONa) in 50 mL of deionized water was introduced, and the mixture was maintained for 2 h, in ultrasound, at room temperature. The suspension and the matrix were separated with the aid of an external permanent magnet, and the solid was washed twice, first with 1% HCl and finally with deionized water. This modification increases the number of the active sites on the GO surface, by transformation of -OH and epoxide groups into functionalizable -O-CH 2 COOH groups.
EDA-MNPs-GO Coupling: The modified MNPs-GO solid from the previous step was suspended in 50 mL of ethanol, together with 4 mL of EDA and 0.25 g of DCC, in a 100 mL round-bottom flask, and stirred at 50 • C for 48 h. The free acid groups obtained in the previous step were condensed with one of the EDA amine groups to form amide bonds, while the second was available to attach an organic group to functionalize the solid. Thus, there were double-origin amino groups available for functionalization: the amino group of the coated MNPs dispersed on the sheet and the free non-condensed amino group of the EDA.

Functionalization of the Prepared Materials
The functionalization processes to attach DPTH or PSTH functional groups were carried out previously by the research group [26,30]. The magnetic materials synthesized following the Methods 1-3 were functionalized and called 1-DPTH, 1-PSTH, 2-DPTH, 2-PSTH, 3-DPTH and 3-PSTH, respectively, where the number is relating to the method. Their schematic structures were presented in Figure 3a-c. The structures of DPTH and PSTH are shown in Figure 4a,b, respectively. Nanomaterials 2020, 10, 304 6 of 18

Functionalization of the Prepared Materials
The functionalization processes to attach DPTH or PSTH functional groups were carried out previously by the research group [26,30]. The magnetic materials synthesized following the Methods 1-3 were functionalized and called 1-DPTH, 1-PSTH, 2-DPTH, 2-PSTH, 3-DPTH and 3-PSTH, respectively, where the number is relating to the method. Their schematic structures were presented in Figure 3a-c. The structures of DPTH and PSTH are shown in Figure 4a,b, respectively.

Functionalization of the Prepared Materials
The functionalization processes to attach DPTH or PSTH functional groups were carried out previously by the research group [26,30]. The magnetic materials synthesized following the Methods 1-3 were functionalized and called 1-DPTH, 1-PSTH, 2-DPTH, 2-PSTH, 3-DPTH and 3-PSTH, respectively, where the number is relating to the method. Their schematic structures were presented in Figure 3a-c. The structures of DPTH and PSTH are shown in Figure 4a,b, respectively.

Adsorption of Heavy and Noble Metals from Aqueous Solution
For the study of the capacity of adsorption, samples were prepared by mixing 5 mg of the magnetic material and 50 mL of 50 µg/L of Cd, Co, Cr, Cu, Ni, As, Hg, Pb, Pt, Pd, Ag and Au. The metals were divided in two groups, to avoid the saturation of the resins Cd, Co, Cr, Cu, Ni, As, Hg and Pb (group 1) and Pt, Pd, Ag and Au (group 2). Therefore, there was a total of twelve samples (1-PSTH groups 1 and 2; 2-PSTH groups 1 and 2; 3-PSTH groups 1 and 2; 1-DPTH groups 1 and 2; 2-DPTH groups 1 and 2; 3-DPTH groups 1 and 2). The capacities of adsorption of the materials were studied at acid pH (between 3 and 5). The pH was adjusted by using acetic acid/sodium acetate buffer. The suspensions were stirred for 10 min, and the materials were separated from the suspension by using an external permanent magnet. The decanted solutions were analyzed by ICP-MS. The highest capacity of adsorption that could be obtained was 0.5 mg/g material.
For the materials achieving the maximum capacity of adsorption, a second experiment with higher concentrations of metals was performed. In this case, samples were prepared by mixing 25 mg of the magnetic material and 50 mL of 10 mg/L of Cd, Co, Cr, Cu, Ni, As, Hg, Pb, Pt, Pd, Ag and Au. The metals were divided into three groups, to avoid the saturation of the resins Co, Cr, Cu, Ag and Ni (group 1); As, Hg, Pb and Cd (group 2); and Pt, Pd, and Au (group 3). Therefore, there was a total of six samples (3-PSTH groups 1-3 and 3-DPTH groups 1-3). The capacities of adsorption of the materials were studied at acid pH (between 3 and 5), and the pH was adjusted by using acetic acid/sodium acetate buffer. The suspensions were stirred for 10 min, and the materials were separated from the suspension by using an external permanent magnet. The decanted solutions were analyzed by ICP-OES. The highest capacity of adsorption that could be obtained was 20 mg/g material.

Results and Discussion
The materials obtained, using the three synthesis methods and the two functional groups, were characterized and tested as adsorbent forward to heavy and noble metals (12 elements). Different characterization techniques were employed.

Transmission Electron Microscopy (TEM)
The surface morphology of the synthesized materials was characterized by TEM and nitrogen adsorption isotherms. As an example, the TEM images of the materials functionalized with PSTH groups are shown in Figure 5; the TEM images of the DPTH materials were very similar. In these images, we can clearly observe that Fe 3 O 4 nanoparticles were disorderly coupled over the GO sheets with a diameter between 12 and 20 nm. Nanoparticles' size was selected intentionally, since smaller particles (<6 nm) show reduced saturation magnetization and magnetic susceptibility, due to surface effects that lower MRI relaxivity values, whereas larger particles (>20 nm) are difficult to disperse, owing to the presence of permanent magnetization at zero field [32]. The synthesis of adequate-sized MNPs was previously optimized by us [29].

N2 Adsorption/Desorption Isotherms
The nitrogen adsorption-desorption experiments for all materials presented isotherms similar to those type IV, typical of mesoporous materials (pore size between 20 and 500 Å ) [33]. The isotherms for the PSTH materials are shown in Figure 6. Some physics characteristics of the materials are shown in Table 1. As can be observed, the materials are mesoporous and present a surface area higher than non-coupled GO (2.63 m 2 /g). Method 3 materials presented lower surface areas than Methods 1 and 2 materials.

N 2 Adsorption/Desorption Isotherms
The nitrogen adsorption-desorption experiments for all materials presented isotherms similar to those type IV, typical of mesoporous materials (pore size between 20 and 500 Å) [33]. The isotherms for the PSTH materials are shown in Figure 6. Some physics characteristics of the materials are shown in Table 1. As can be observed, the materials are mesoporous and present a surface area higher than non-coupled GO (2.63 m 2 /g). Method 3 materials presented lower surface areas than Methods 1 and 2 materials.

X-Ray Photoelectronic Spectroscopy (XPS)
The XPS spectra in the sulfur region can be seen in Figure 7. In 1-DPTH, 2-DPTH and 3-DPTH spectra (Figure 7a,c,e, respectively), two oxidation states for sulfur are shown, being assigned to the tautomers of C=S (around 167 eV) and C-SH bond (around 163 eV). In 1-PSTH, 2-PSTH and 3-PSTH spectra (Figure 7b,d,f, respectively), the most energetic peak (more oxidized) is much higher due to the presence of a sulfonic group in the structure (Figure 4b,d). In the case of 2-DPTH, 2-PSTH, 3-DPTH and 3-PSTH (Figure 7c-f), the peak situated in a lower bind energy (C-SH bond) clearly presents a higher intensity in comparison with the equivalent peak in 1-DPTH and 1-PSTH spectra (Figure 7a,b). It can be concluded that the tautomeric equilibrium was displaced to the reduced tautomeric form (Figure 4c,d) when the materials were synthesized by Methods 2 and 3. The main explanation could be the presence of electronic interactions between the π-π system of GO and the aromatic system of the functional groups, favoring the tautomers with the most extended electronic system. This effect is stronger when the DPTH and PSTH are close to the GO sheet, meaning the functionalization through the -COOH groups of the GO sheet (Methods 2 and 3). In Method 1, the functional groups are attached only to the MNPs, resulting in a greater distance between the electronic systems.

X-Ray Photoelectronic Spectroscopy (XPS)
The XPS spectra in the sulfur region can be seen in Figure 7. In 1-DPTH, 2-DPTH and 3-DPTH spectra (Figure 7a,c,e, respectively), two oxidation states for sulfur are shown, being assigned to the tautomers of C=S (around 167 eV) and C-SH bond (around 163 eV). In 1-PSTH, 2-PSTH and 3-PSTH spectra (Figure 7b,d,f, respectively), the most energetic peak (more oxidized) is much higher due to the presence of a sulfonic group in the structure (Figure 4b,d). In the case of 2-DPTH, 2-PSTH, 3-DPTH and 3-PSTH (Figure 7c-f), the peak situated in a lower bind energy (C-SH bond) clearly presents a higher intensity in comparison with the equivalent peak in 1-DPTH and 1-PSTH spectra (Figure 7a,b). It can be concluded that the tautomeric equilibrium was displaced to the reduced tautomeric form (Figure 4c,d) when the materials were synthesized by Methods 2 and 3. The main explanation could be the presence of electronic interactions between the π-π system of GO and the aromatic system of the functional groups, favoring the tautomers with the most extended electronic system. This effect is stronger when the DPTH and PSTH are close to the GO sheet, meaning the functionalization through the -COOH groups of the GO sheet (Methods 2 and 3). In Method 1, the functional groups are attached only to the MNPs, resulting in a greater distance between the electronic systems.  Figure 8a. These spectra confirm the double coupling of MNPs-GO in Method 3 (chemical and physical interactions). After the first step of the synthesis method (MNPs-GO coupling) two types of nitrogen are shown (Figure 8b), being assigned to amine groups (398.7 eV) and amide (400.8 eV) bonds [34]. Amide bonds resulted from the condensation reaction between aminopropyl groups of MNPs and -COOH functional groups on GO (chemical interaction) and the amine groups from the non-condensed aminopropyl groups of immobilized MNPs by physical interactions. In this way, the double coupling of MNPs-GO was demonstrated. As can be seen, the number of MNPs coupled by physical interactions was lower than MNPs coupled by chemical interactions (36% in front to 64% area, respectively). Figure 8c shows the sorbent after EDA-MNPs-GO coupling reaction and before functionalization of the material with the organic ligand. In this case, the peak assigned to amine bond is much higher than the peak assigned to amide bond (80% in front to 20% area, respectively), which confirms the presence of EDA in the structure.

CNHS-Fe Elemental Analysis
The MNPs-GO coupling and the functionalization grade of the new synthesized materials were studied by the determination of Fe concentration, using SS-HR CS GFAAS and CNHSO elemental analysis (Table 2). Method 3 showed a higher percentage of S and N, which confirms a better yield of functionalization than Methods 1 and 2. Moreover, 1-PSTH, 2-PSTH and 3-PSTH contain a higher percentage of S than 1-DPTH, 2-DPTH and 3-DPTH, respectively, due to the fact that PSTH presents an additional S atom in the structure (Figure 4b). Moreover, from the percentage of C, it could be concluded that Method 3 could bind a higher amount of GO to the material with 45% C, approximately; nevertheless, in Methods 1 and 2, the C percentages were lower (approximately, 8% and 30%, respectively). Furthermore, the Fe percentage was lower when C percentage increased in the structure, with both parameters being inversely proportional.

Mass Spectrometry
From MS spectra, the characteristic peaks were assigned to different fragments ( Table 3). Some of them were found for every material, such as m/z = 129 and m/z = 207 peaks. Both were related to the presence of the thiocarbonohydrazide and pyridyl-thiocarbohydrazide, a reagent used in DPTH and PSTH synthesis. Other peaks were just found in the case of DPTH materials (m/z = 105 and 184) and PSTH materials (m/z = 64), agreeing with the structural differences between the functional groups.

Capacities of Adsorption of Heavy and Noble Metals from Aqueous Solution
In order to know which of these new nanomaterials offer better benefits to extract and preconcentrate metals ions, the capacities of adsorption toward 12 elements were studied (noble metals and transition metals). The capacities of adsorption of the materials were determined in batch, using the procedure described above (Section 2.6), by ICP-MS. From this study, it could be established that, at acid pH (Table 4), the highest values of adsorption were reached by 3-DPTH and 3-PSTH (Method 3) for each element. However, Method 3 materials presented lower surface areas than Method 1 and Method 2 materials (Table 1). This fact seems to explain that the main mechanism for the adsorption of metals is the chemisorption, not physisorption. Moreover, both materials reach the maximum value of the experiment (0.5 mg/g).

Capacities of Adsorption of Heavy and Noble Metals from Aqueous Solution
In order to know which of these new nanomaterials offer better benefits to extract and preconcentrate metals ions, the capacities of adsorption toward 12 elements were studied (noble metals and transition metals). The capacities of adsorption of the materials were determined in batch, using the procedure described above (Section 2.6), by ICP-MS. From this study, it could be established that, at acid pH (Table 4), the highest values of adsorption were reached by 3-DPTH and 3-PSTH (Method 3) for each element. However, Method 3 materials presented lower surface areas than Method 1 and Method 2 materials (Table 1). This fact seems to explain that the main mechanism for the adsorption of metals is the chemisorption, not physisorption. Moreover, both materials reach the maximum value of the experiment (0.5 mg/g).

Capacities of Adsorption of Heavy and Noble Metals from Aqueous Solution
In order to know which of these new nanomaterials offer better benefits to extract and preconcentrate metals ions, the capacities of adsorption toward 12 elements were studied (noble metals and transition metals). The capacities of adsorption of the materials were determined in batch, using the procedure described above (Section 2.6), by ICP-MS. From this study, it could be established that, at acid pH (Table 4), the highest values of adsorption were reached by 3-DPTH and 3-PSTH (Method 3) for each element. However, Method 3 materials presented lower surface areas than Method 1 and Method 2 materials (Table 1). This fact seems to explain that the main mechanism for the adsorption of metals is the chemisorption, not physisorption. Moreover, both materials reach the maximum value of the experiment (0.5 mg/g).

Capacities of Adsorption of Heavy and Noble Metals from Aqueous Solution
In order to know which of these new nanomaterials offer better benefits to extract and preconcentrate metals ions, the capacities of adsorption toward 12 elements were studied (noble metals and transition metals). The capacities of adsorption of the materials were determined in batch, using the procedure described above (Section 2.6), by ICP-MS. From this study, it could be established that, at acid pH (Table 4), the highest values of adsorption were reached by 3-DPTH and 3-PSTH (Method 3) for each element. However, Method 3 materials presented lower surface areas than Method 1 and Method 2 materials (Table 1). This fact seems to explain that the main mechanism for the adsorption of metals is the chemisorption, not physisorption. Moreover, both materials reach the maximum value of the experiment (0.5 mg/g).

Capacities of Adsorption of Heavy and Noble Metals from Aqueous Solution
In order to know which of these new nanomaterials offer better benefits to extract and preconcentrate metals ions, the capacities of adsorption toward 12 elements were studied (noble metals and transition metals). The capacities of adsorption of the materials were determined in batch, using the procedure described above (Section 2.6), by ICP-MS. From this study, it could be established that, at acid pH (Table 4), the highest values of adsorption were reached by 3-DPTH and 3-PSTH (Method 3) for each element. However, Method 3 materials presented lower surface areas than Method 1 and Method 2 materials (Table 1). This fact seems to explain that the main mechanism for the adsorption of metals is the chemisorption, not physisorption. Moreover, both materials reach the maximum value of the experiment (0.5 mg/g). Table 4. Capacity of adsorption of the synthesized materials at acid pH (mg/g). The highest capacity of adsorption that could be obtained was 0.5 mg/g.
The capacities of adsorption of both materials were studied at higher concentration levels, and a second experiment was conducted (maximum level of 20 mg/g). In this case, the adsorption capacities of the materials were determined in batch by ICP-OES, and the data obtained are shown in Table 5. A comparison between non-functionalized GO and the materials synthesized by Method 3 was conducted to verify the effect of the functionalization and MNPs-GO coupling. As can be seen in Table 5, the prepared materials by Method 3 (3-DPTH and 3-PSTH) offered better benefits to be used as SPE adsorbent than non-functionalized GO at acid pH.
These benefits were obtained due to the double functionalization achieved in 3-DPTH and 3-PSTH synthesis; both MNPs and GO sheets were functionalized with the organic ligand. The ligands were anchored through the amine groups of: the aminopropyl silica coating of MNPs and the EDA-GO. This double functionalization was possible due to the double coupling. MNPs were coupled to the GO through two mechanisms: covalent binding by condensation with the carboxylic acid groups, and physical interactions with the GO adsorbent. The highest capacity of adsorption that could be obtained was 20 mg/g.

Conclusions
In this work, three different mechanisms for the coupling and functionalization of the MNPs and GO materials were addressed. For the synthesis of 1-DPTH and 1-PSTH, silica-covered MNPs were first functionalized, and after, they were dispersed on GO. For 2-DPTH and 2-PSTH synthesis, MNPs were coprecipitated into GO-EDA sheets before the functionalization of the GO layer. In 3-DPTH and 3-PSTH synthesis, both MNPs and GO sheets were functionalized with the organic ligand, and this process was called double functionalization. This double functionalization was possible due to the double coupling of MNPs and the GO layer (a fraction of MNPs coupled by Van der Waals interactions and a second fraction attached to the surface by covalent bond to the GO sheets), being the first time in which this double functionalization was addressed.
From the studies of the capacities of adsorption could be concluded that 3-DPTH and 3-PSTH showed higher capacity of adsorption toward different elements (transition and noble metals), being ideal adsorbents for decontamination purposes. Method 3 materials presented lower surface areas than Method 1 and Method 2 materials (Table 1). This fact seems to explain that the main mechanism for the adsorption of metals is the chemisorption, not physisorption. Therefore, the capacities of adsorption of the materials rely mostly on the superficial concentration of the functional groups. From the characterization studies was concluded that the functional groups were attached to the surface successfully, and the third procedure offered better yields of functionalization. Therefore, Method 3 was chosen as the more appropriate method to synthesize functionalized MNPs-GO coupling's different functional groups, to improve the selectivity toward any chemical species.