Abstract

A novel hybrid, water-based, sol-gel method was developed for the synthesis of functionalized silica with advanced adsorptive capacities towards heavy metal ions. The study employs the hyperbranched poly(ethylene imine) (HBPEI) as a reactive template for the synthesis of silica. The reaction was followed by FTIR, whereas the material was evaluated in terms of its microstructure and adsorptive properties. The results revealed a very potent adsorbent that can remove heavy metal ions (namely, Pb, Cu, and Zn) from water in a fast and efficient way. Greater selectivity was observed for lead, whereas the maximum adsorption capacities calculated from the Langmuir model were 833.3, 502.5, and 193.4 mg/g for lead, copper, and zinc, respectively.

1. Introduction

Heavy metal pollution in water is an issue of great concern, constituting a severe threat for the human health and ecological systems. Copper, lead, zinc, cadmium, mercury, and nickel are considered the most dangerous elements and are included in the US Environmental Protection Agency’s (EPA) list of priority pollutants. The reason is their high toxicity in combination with their growing discharge in the environment and bioaccumulation effects. In fact, heavy metals, in contrast to the more toxic organic compounds, are not degradable, being accumulated in the environment and eventually ending up in the humans through the food chain [13].

Various techniques have been developed for the removal of metal ions from wastewater such as reverse osmosis, nanofiltration [4], ion exchange [5], adsorption [6], precipitation [7], and coagulation [8]. Among them, adsorption is one of the most effective methods for the remediation of heavy metal pollution [9] as the process offers flexibility in both design and operation, providing also in most of the cases the advantages of a reusable effluent and of an economic regeneration procedure due to its reversible nature [10].

Several adsorbents have been proposed to remove heavy metal ions from wastewater, such as plenteous and inexpensive cellulose and its modified forms [11], multifarious agricultural waste material, and potentially low cost sorbent clays [12], iron oxides, and hydroxides [13, 14] as well as carbon nanotubes [15]. Recently, dendritic polymers containing amine groups have shown a high binding capacity for metal ions dissolved in water. The properties of these materials are related to the degree of branching and the number of amino groups. Among them, the hyperbranched poly(ethylene imine) (HBPEI) is a low cost polymer, well known for its high activity and therefore high functionalization potential [16, 17].

HBPEI consists of a spheroidal, asymmetric structure with a high branching degree (between 65 and 75%) and relatively narrow polydispersity. It is a cationic polymer demonstrating the maximum density of positive charge when it is protonated in an aqueous solution. These properties make HBPEI useful for a number of industrial applications such as in the paper industry for the flocculation of negatively charged fibers or as additive in paper production printers, in drug delivery systems. Its chemical properties and especially its proved high efficiency in entrapping heavy metals render it also very promising for applications in wastewater treatment and metal ion removal [1821]. For these reasons, it is often used to increase the adsorption capacity of an adsorbent by modifying its surface. Such an adsorbent is usually silica that is a low-cost ceramic with excellent thermal and mechanical stability, unique large surface area, and easily modified surface characteristics [22]. Due to these excellent qualifications, silica is very often the material of choice for the immobilization of poly(ethylene imine) (PEI) to develop a hybrid for the effective removal of heavy metals from water [23].

Silica-based adsorbents have been developed by various ways over the years. Porous structures have been synthesized by sol-gel mainly in the presence of anionic and neutral substances or by utilizing hyperbranched polyesters and dendritic polymers as precursors or pore-forming agents [2426]. The design of hybrid nanostructured materials has also drawn immense attention recently as these materials offer great potential for applications in many industrial fields including adsorption processes [27]. In this context, the surfactant method has been employed to synthesize mesoporous silica materials with controlled porosity and structure by changing the type or length of the chain of the surfactant matrix. The as-synthesized silica powders were subsequently functionalized with organoalkoxysilanes and tested for their adsorptive capacity to remove heavy metals from drinking water. Hybrid organic-inorganic nanospheres synthesized through the interaction of HBPEI with silicic acid in water have also been used successfully for the sorption of toxic metal ions and polycyclic aromatic hydrocarbons from water [16]. Further, novel adsorbents resulting from the functionalization of chitosan–silica hybrids with EDTA (ethylenediaminetetraacetic acid) ligands were also synthesized and tested for their adsorptive capacities with respect to Cu(II), Ni(II), Cd(II), and Pb(II) [28].

In this work, a functionalized silica network, synthesized through a water-based sol-gel approach, is evaluated as a potential adsorbent of heavy metals (Pb, Cu, and Zn) from contaminated water demonstrating excellent results. The synthesis approach involves the reaction of HBPEI with 3-(triethoxysilyl)propyl isocyanate (TESPI) followed by the subsequent hydrolysis and cross-linking sol-gel reactions. The whole procedure is based on the method discussed in our previous work [26] which can afford nanostructured, dendritic-based networks which upon calcination lead to nanoporous silica with tailored pore structure and high surface area (~600 to 1100 m2/g depending on the conditions employed). However, in the present work, the organic solvent for HBPEI (N,N-dimethylformamide or chloroform) is replaced by water with the aim to further simplify the procedure and reduce both the cost of the material and the environmental drawbacks.

2. Experimental

2.1. Materials and Methods

Hyperbranched poly(ethylene imine) (HBPEI) (Mn = 25,000) and 3-(triethoxysilyl)propyl isocyanate (TESPI, 95%) were provided by Sigma-Aldrich. Lead(II) nitrate (Pb(NO3)2, ≥99.99%) was provided by Aldrich, and zinc nitrate hexahydrate (Zn(NO3)2·6H2O, ≥99%) and copper nitrate hemi(pentahydrate) (Cu(NO3)2·2.5H2O, ≥98%) were purchased by Sigma-Aldrich.

2.2. Synthesis of Hybrid Network

The organic-inorganic network was prepared via a novel hybrid sol-gel templating route (Figure 1) which has been described in detail in our previous work [26]. However, in this case, the solvent employed was water. In short, appropriate amount of poly(ethylene imine) was dissolved in ultrapure water so that the HBPEI : water molar ratio with respect to the sum of primary and secondary amino groups of HBPEI reaches the value 1 : 8. The addition of HBPEI was made under continuous stirring and after dissolution; the resulting solution was allowed to stand for about 30 min. Afterward, to this solution, which was kept at 0°C, 3-(triethoxysilyl)propyl isocyanate (TESPI) was added dropwise, under continuous stirring, to reach molar ratio SiO2/HBPEI = 0.16, 0.35, or 0.8 and the reaction mixture was held at this temperature for one more hour. Then, it was allowed to reach room temperature where it remained for several minutes, again under continuous stirring, so that interaction between the terminal amino groups of hyperbranched poly(ethylene imine) and the isocyanate group of silica’s precursor takes place. Finally, the reaction product was left for curing at room temperature for 24 hours, in order to attain hydrolysis of triethoxy groups and silanol formation, and then, it was dried at 90°C for 100 h to complete the formation of the silica network.

2.3. Characterization Techniques

The presence of the hyperbranched poly(ethylene imine) in the silica particles and the formation of the siloxane bridges were established by FTIR spectroscopy employing a Nicolet Magna-IR 550 Spectrometer. Thermogravimetry experiments were also performed up to 1000°C under air flow at a heating rate of 10°C/min, employing a TGA 2050 analyzer (TA Instruments).

The morphology and the microstructure of dispersed powder were examined with a high-resolution transmission electron microscope (HRTEM, JEOL 2100 HR). Bright field images were recorded over selected areas for this purpose. The sample was washed several times with deionized water and acetone, in order to remove the excess of the polymer and to have a better observation of it. Portions of samples were crushed in an agate mortar and suspended in ethanol.

2.4. Heavy Metal Adsorption Experiments

For the investigation of the adsorption of heavy metals on nanohybrid surfaces, adsorption experiments were made by varying the concentration of metals into pure water as well as the contact time. For the realization of adsorption as a function of concentration of metals, 2.5 mg of the hybrid material (SiO2-HBPEI-0.16) was dispersed in 10 mL ultrapure water. The pH value was adjusted with 0.1 M HCl and 0.1 M NaOH and kept at 6.2. Then, the appropriate volume of standard solution was added to a suspension of metal so that the final concentration of the metal is 200 ppm, 150 ppm, 100 ppm, 80 ppm, 60 ppm, 40 ppm, 20 ppm, and 10 ppm, and the suspension was allowed under stirring at room temperature for 24 hours in order to ensure that the adsorption equilibrium is reached. The pH values even after the adsorption process were remained at the range of 6.1–6.2. The concentration of the metal ions in the solution was measured using an atomic absorption spectrophotometer (AS-PerkinElmer 4100). The adsorption capacity of the adsorbent at equilibrium was calculated as follows: where is the equilibrium adsorption capacity of the adsorbent in mg/g, is the initial concentration in mg/L, is the concentration at equilibrium of metal ions in mg/L, is the volume in L of metal ion solution, and is the weight in g of the adsorbent [29].

The kinetics of the adsorption process was investigated to study the effect of the initial concentration of metal ions on with respect to time and the time required to achieve equilibrium between aqueous and solid phases. The kinetics study was carried out using 2.5 mg material dispersed in 10 mL of 1 ppm Pb(II), Cu(II), and Zn(II) metal ion solutions. The suspensions were left under stirring for 5 min, 10 min, 15 min, 30 min, 2 h, 6 h, 12 h, and 1 day. The concentration of the pollutant in the supernatant was determined spectrophotometrically as described above. The removal efficiency, , of the heavy metal ions was calculated by the following equation: where is the initial concentration of the metal ions (mg/L) and is the concentration of the metal ions (mg/L) at time [30].

3. Results and Discussion

3.1. FTIR Analysis

FTIR analysis was employed to follow the reactions occurring for the formation of the hybrid silica network functionalized with HBPEI (Figure 2).

It is well known in literature the reaction of isocyanates with water. The kinetics of this reaction to form urea as a final product is studied in detail in the work of Hai et al. [31]. The isocyanate group reacts initially with water to form an intermediate, unstable, carbamic acid which spontaneously decarboxylates into a primary amine and carbon dioxide. The amine, in turn, reacts again with another isocyanate group producing a urea cross-link. Since, in our case, the reaction of TESPI with HBPEI was performed in water, we also examined by the FTIR characterization technique a sample that was developed following the same procedure but in the absence of HBPEI (named as SiO2-free of HBPEI). The reason was to examine if HBPEI is finally involved in the formation of the silicate network, since, in this case, there coexists the reaction of TESPI with water. For this reason, materials received employing different TESPI to HBPEI molar ratios (see Experimental) were also analyzed by FTIR. The patterns of the different materials that were synthesized as well as the pattern received from the HBPEI itself are depicted and compared in Figure 2.

Examining the spectrum of the SiO2 free of the HBPEI sample (the sample prepared in the absence of HBPEI during reaction), it becomes obvious that a different material has been produced compared to the one synthesized employing the highest amount (ratio of TESPI to HBPEI, 0.16) of HBPEI whereas the spectra of the other materials, with intermediate HBPEI content during reaction, present a progressive evolution from the pattern of the HBPEI-free sample to the one with the highest polymer content. In fact, in the spectrum of the SiO2 free of HBPEI sample, strong bands are revealed, characteristic of the urea bonds’ vibrations. These are the bands at 3335 cm−1, 1695 cm−1, 1630 cm−1, 1479 cm−1, and 1016 cm−1, which are assigned to the N-H stretching, C=O stretching, N-H deformation, C-N asymmetric stretching, and C-N symmetric stretching vibrations in urea, respectively [32, 33]. Therefore, in agreement with the literature data, in the case of the HBPEI-free sample, the TESPI reacts with water leading to the formation of urea links.

In contrast, all these characteristic urea bands are totally absent from the SiO2-HBPEI-0.16 sample. This permits the hypothesis that in this sample the reactions followed a different path involving the HBPEI. The spectrum of this sample reveals the characteristic vibration of the amide I band vibration at 1654 cm−1 suggesting the formation of secondary amide groups possibly through the reaction of isocyanate groups of TESPI with the primary amino groups of HBPEI. The reaction of HBPEI amino groups with the precursor of silica (TESPI) is evidenced by the absence of the N–H asymmetric stretching vibration in HBPEI at 3350 cm−1, while the symmetric stretching is hardly observed at 3276 cm−1 [26]. In addition, the absorption band attributed to the bending of the NH2 groups in the HBPEI spectrum at 1583 cm−1 is absent from the spectrum of the SiO2-HBPEI supporting the reaction of HBPEI [16].

Further, the absence of antisymmetric (2974 cm−1) and symmetric (2885 cm−1) stretching vibrations of the methyl groups from the spectrum of the SiO2-HBPEI-0.16 sample suggests the hydrolysis of ethoxysilane groups to silanols. The weak band at 921 cm−1 corresponding to the stretching vibrations of the Si-OH groups further supports the assumption of hydrolysis of ethoxysilane groups [26]. Again, the SiO2 free of the HBPEI sample shows a different spectrum regarding the hydrolysis degree, as it reveals the bands at 2975 cm−1 and 2883 cm−1 as well as a band at 953 cm−1 corresponding to the vibration zone of H3C-C groups, whereas the band at 921 cm−1 is absent. Therefore, it can be derived that, in this case where the HBPEI is not present in the reaction, the hydrolysis of ethoxysilanes is incomplete [26].

The formation of the silica network as a result of the condensation of silanols to form Si-O-Si bonds in the SiO2-HBPEI-0.16 sample can be evidenced by the presence in the spectrum of the characteristic vibrations at 1103 cm−1 and at 1040 cm−1 assigned to the transverse optical stretching modes of Si-O-Si bonds which are present in large cyclic structures of the dense network of silica [3436] and in smaller siloxane rings encountered in the sol [26], respectively. A band associated with Si-O rocking can be also observed at 455 cm−1 [35].

Finally, the hybrid nature of the silica network in the SiO2-HBPEI-0.16 sample is evidenced by the presence of the asymmetric and symmetric stretching bands of CH2 that appear at 2931 cm−1 and 2809 cm−1 [26]. Further, the bands at 1454 cm−1 and 1363 cm−1 correspond to asymmetric and symmetric vibrations of CH2, respectively [36], while this at 753 cm−1 is assigned to the zone oscillation of CH2 [16].

3.2. XRD Analysis

To confirm the formation of the silica network, the SiO2-HBPEI-0.16 material (prepared with the highest HBPEI content during synthesis) was analyzed by XRD in the as-received state and after calcination at different temperatures in the range 1000 to 1400°C. The as-received material is amorphous showing peaks which correspond to the polymer that is present in the structure. Upon calcination, these peaks disappear and the pattern corresponds to pure amorphous silica that is crystallized to cristobalite after calcination at 1400°C (Figure 3).

3.3. Thermogravimetric Analysis

Thermogravimetry analysis (TGA) of the SiO2-HBPEI-0.16 material was performed under flowing air, up to 1000°C to investigate the presence of organic matter into the material. As it can be seen in Figure 4, a large mass loss reaching ∼35% of the initial weight is revealed in the temperature range 100–800°C. The initial weight loss up to about 150°C, which is estimated at around 10%, can be attributed to the evaporation of moisture. However, further weight losses taking place at higher temperatures can be associated with the decomposition and burning of the hyperbranched polymer (an exothermal peak appears in the DTA curve at 350°C revealing the burning of polymer), with the dissociation of structural water and possibly with additional condensation reactions of silanol groups. These entire phenomena stop above 800°C as no further mass loss can be noticed, suggesting that up to this temperature HBPEI and amines have been fully decomposed affording silica free of any organic material.

3.4. TEM Analysis

A characteristic TEM image of the SiO2-HBPEI-0.16 material is shown in Figure 5. As it can be seen, the material is consisted of particles with an almost spherical shape with diameters ranging between 3 and 10 nm.

3.5. Adsorption Isotherms for Heavy Metals

Finally, the material that was prepared with the highest HBPEI content during synthesis, namely, the SiO2-HBPEI-0.16 material, was selected to be further evaluated for its adsorptive capacities towards heavy metals (Pb2+, Cu2+, and Zn2+). For this, sorption experiments were performed by varying the contact time and the initial concentration of the metal ion into the water solution. The sorption isotherms received are depicted in Figure 6.

The material shows similar behavior in all three cases under investigation. Load increases with increasing concentration of the metal in the solution. However, there is different selectivity for each metal following the order Pb > Cu > Zn. In fact the maximum adsorption load measured in these experiments was 605.4, 462.2, and 188 mg/g for Pb(II), Cu (II), and Zn (II), respectively.

Heavy metal sorption capacity of silica has been correlated to the cation ionic radius as metals of larger radius possess greater tendency to polarize and thus to form complexes with the deprotonated negatively charged silanol groups of silica surface. Similarly, the interaction strength between the metal ions and the amino groups in the hybrid material is also proportional to the metal ion polarization [37]. As a consequence, a greater bonding strength with both the amino and the silanol groups is expected for Pb2+ which exhibits the highest molecular weight and the largest electronegativity among the three ions (Pb(II), Zn(II), and Cu(II)) investigated [38], a fact that can explain the higher selectivity shown for Pb sorption.

Table 1 compares the results of the adsorption capacity for the adsorbent prepared in the present work to those obtained by various adsorbents reported by other authors [16, 23, 3943]. It can be easily seen that the as-synthesized novel material is a very effective adsorbent of metal ions showing capacities much higher than the previously reported values.

The obtained experimental results were analyzed by the Langmuir adsorption model, which is based on the hypothesis that the maximum adsorption corresponds to a saturated monolayer of adsorbate molecules on the adsorbent surface. Consequently, adsorption appears uniformly on the adsorbent’s active sites, and once an adsorbate occupies a site, no further adsorption will take place at this site. The Langmuir model is expressed by the following: where is the equilibrium adsorption capacity of the adsorbent in mg/g, is the concentration of residual solute in mg/L at equilibrium, is the maximum uptake of the adsorbent (mg/g), and (L/mg) is the Langmuir constant that expresses the binding energy of adsorption (affinity) [44].

Therefore, with the proper processing of experimental results by Langmuir equation (3), the related constant value and the value were calculated (Figure 7). These values along with the nonlinear regression correlation coefficients () for Langmuir isotherms are given in Table 2.

As it can be seen in Table 2, the correlation coefficients () were very high in all three cases examined, suggesting that the adsorption follows the Langmuir model. Further, the analysis of the results confirmed that the SiO2/HBPEI hybrid nanopowder exhibits very high adsorption capacities following the order Pb(II) > Cu(II) > Zn(II) in accordance with the experimental results discussed above.

The type of the Langmuir isotherm can be also used to predict whether the adsorption is favorable or unfavorable. This can be realized by either evaluating the equilibrium parameter () or a dimensionless constant separation factor expressed as follows: where (mg/L) is the initial concentration of metal ion. In fact, for values of the separation factor constant  < 1, the adsorption is considered favorable, whereas the adsorption is considered unfavorable for  > 1 [45]. In this respect, the values were calculated from the experimental data and their range for each system is also given in Table 2. The results show values smaller than unity suggesting a favorable adsorption in all cases.

3.6. Adsorption Kinetics

The results of kinetic experiments are presented in Figure 8. The loading of all three metal ions on the material increases rapidly in the first five minutes followed by a very slow increase up to one day. In complete agreement with the data obtained from the analysis of the adsorption isotherms, the greatest load during the period of the first five minutes is measured for lead, confirming the strong chelating ability of the hybrid SiO2-HBPEI-0.16 material to Pb2+ [16].

Adsorption kinetic parameters, i.e., the rate constants and equilibrium adsorption capacities, which can provide valuable insights into wastewater treatment process design, are of great importance for the application of adsorbents. In order to evaluate the kinetic adsorption mechanism, the pseudo-first-order (5) and pseudo-second-order (6) models were investigated: where is the amount of metal adsorbed at equilibrium (mg g−1), is the amount of metal ions adsorbed at time (mg g−1), and and are the rate constants of pseudo-first-order and pseudo-second-order (g mg−1 min−1) models, respectively [46].

The calculated parameters of the two models are summarized in Table 3. Based on the correlation coefficients , the results clearly indicate that the adsorption kinetics follow the pseudo-second-order kinetic model (Figure 9). Furthermore, the equilibrium adsorption capacities calculated () based on the pseudo-second-order model are in perfect agreement with the experimental data (). The fitting of this model suggests that chemisorption is the controlling mechanism with the sorption capacity being proportional to the number of active sites on the hybrid silica [47].

4. Conclusions

In this research, an environmentally friendly method for the preparation of hybrid materials with improved adsorptive properties was developed and implemented. The synthesis concerns a cost effective process, performed in water, which employs hyperbranched PEI as a reactive template for the synthesis and simultaneous functionalization of silica nanopowder.

The results of FTIR confirmed that the proposed synthesis method is effective leading to the successful substitution of the amino groups of the silane and the successful formation of the silica network. The hybrid material has a spherical shape with a mean diameter at about 7 nm.

The as-synthesized hybrid (SiO2-HBPEI-0.16) material was evaluated for its adsorption potential to remove heavy metal ions (namely, Zn2+, Pb2+, and Cu2+) from water. The adsorption experiments demonstrated that the material is an excellent adsorbent. Greater selectivity was shown for lead followed by that for copper. The kinetics experiments revealed that the process of adsorption is carried out in a relatively short time (5 min) as the metal loading on the material increases strongly during this time.

Data Availability

The data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest

The authors declare that there is no conflict of interest regarding the publication of this paper.