Synthesis and characterization of magnetic organic montmorillonite: efficient adsorption of hexavalent chromium

Due to rapid industrialization, large amounts of toxic heavy metals have been released into the global groundwater ecosystem (Idrees et al. 2018). Chromium occurs in aquatic systems in different oxidation states with individual environmental, chemical, and biological characteristics (Hua et al. 2012). Cr(VI) is regarded as more toxic than Cr(III) because it contains a lethal dose of 80–160 mg/L (Ashour & Tony 2020). Moreover, the toxicity of Cr(VI) is easily absorbed by the body, causing damage to human liver organs and affecting the metabolic process of substances in the body (Rehman et al. 2021). Most methods are costly, infeasible and inefficient when heavy metal concentrations are 10–100 times the minimum (Zhang et al. 2021a, 2021b). Among the water treatment technologies, adsorption is one of the most successfully applied methods because it is easy to handle, simple, cost effective (Tian et al. 2022), and has strong industrial applicability compared with many other methods (Tao et al. 2019).

As a new type of clay, montmorillonite (Mt) is receiving more attention for metal removal due to its low cost, abundance, large surface area, good structural stability and high ion exchange capacity (Kang et al. 2018). Mt has the characteristics of being hydrophilic, with ultra-high swelling properties, is oleophobic and has little compatibility with polymers (Sun et al. 2016). Due to the negative charge on the surface of suspensions of Mt, original Mt usually have little or no affinity for anions and exhibit low adsorption for Cr₂O₇²⁻ which is electronegativity too. Thus, cationic surfactant molecules are considered to be the best component for electrostatic adsorption with clay minerals through ion exchange and electrostatic interactions. The use of quaternary ammonium salt cationic surfactant (STAC) intercalated montmorillonite not only significantly increases the organic matter content (Jemima et al. 2019), but also modifies the hydrophobic nature of the montmorillonite and improves the application of the clay in wastewater treatment. Simultaneously, STAC intercalates organic macromolecules into the clay mineral layers through exchange with inorganic cations, as well as hydrophobic bonding, increasing the montmorillonite layer spacing and changing the charge properties as well as the hydrophobic properties of the surface (Liu et al. 2019a, 2019b; Castro-Castro et al. 2020). As early as 2004, STAC-modified montmorillonite materials were synthesised by Soares et al. but the adsorption properties were not measured (Soares et al. 2004). Hong et al. used STAC-modified retortite for Cr(VI) adsorption, but had problems with low adsorption capacity and difficult separation (Hong et al. 2008). Liu et al. used STAC and ethylenediamine-modified montmorillonite to determine the adsorption properties on Cu²⁺ and obtained results of high adsorption, but relatively poor stability (Liu et al. 2019a, 2019b).

In recent years, magnetic (Fe₃O₄) particles, have received a lot of attention for pigments, magnetic resonance imaging, data storage media and magnetic drug delivery (Kalantari et al. 2014). Magnetic applications have been used in water treatment since 1873, but magnetic montmorillonite adsorbents are still a relatively new concept (Mehta et al. 2015). The introduction of magnetic particles (Fe₃O₄) into a sparse and porous adsorbent substrate with a large specific surface area can better improve the adsorption performance. The adsorption capacity of As(V) was enhanced by the introduction of magnetic Fe₃O₄ nanoparticles into wheat straw by Tian et al. (2011). The adsorption of methylene blue by Fe₃O₄-activated montmorillonite was investigated by Chang and colleagues who demonstrated that the composite had good stability and reproducibility (Chang et al. 2016). Jang and Lee successfully synthesised magnetite nanoparticles (MNP-OMMTs) of organically modified montmorillonite for the efficient removal of iodide (Jang & Lee 2018). These typical modified Mt materials mentioned above all exhibited efficient adsorption of contaminants and stability of the composite material. At the same time, magnetic separation technology is of great interest for its simplicity and greenness in the sorption and removal of pollutants.

We focused on the structure of the modified adsorbent. The chemical structure and porosity of montmorillonite can support the preparation of magnetic nanocomposites with excellent properties. However, the organic magnetic montmorillonite has been less studied, especially for its application for the adsorption of Cr(VI). Therefore, through STAC and inorganic iron salts modification of low-cost montmorillonite, nature-friendly, efficient, economical and easily synthesised quaternary ammonium salt-modified magnetic and non-magnetic adsorbent materials were synthesized in this study and their ability to bind Cr(VI) was evaluated. In particular, the key aspect of this study involved a simple method that was used to synthesize high particle size magnetic particles using alkali oxidation of ferrous sulfate. Finally, we also focused on the effects of wastewater pH, adsorbent dosage and contact time on the adsorption of Cr(VI) by magnetic materials.

The octadearyl dimethyl ammonium chloride (STAC, C₂₁H₄₆NCl) was purchased from the Tianjin Damao Chemical Reagent Plant, China. The montmorillonite (Mt) was purchased from the Zhejiang Fenghong New Material Co., China. The potassium dichromate (K₂Cr₂O₇), ferrous sulfate (FeSO₄), diphenylcarbonyl dihydrazide (C₁₃H₁₄N₄O), hydrochloric acid (HCl), and sodium hydroxide (NaOH) were supplied by Nanjing Chemical Reagents Ltd, China. All the chemicals were of analytical grade and used without further purification.

Briefly, 1.6 g of STAC was added to the Mt suspension (2% mass fraction). The suspension was stirred vigorously for 3 h using a magnetic stirrer. The precipitate was washed six times with deionised water and dried in an oven at 60 °C for 12 h. Finally, organically modified montmorillonite STAC-Mt was obtained. A certain amount of ferrous sulphate was dissolved in deionised water and this was stirred vigorously at 50 °C for 5 h with a magnetic stirrer to obtain a 0.3 M solution of ferrous sulphate. The ferrous sulphate solution was added to the STAC-Mt suspension and adjusted to pH of 11 using NaOH (0.1 M), the mixture was heated to a temperature of to 90°, stirred magnetically for 3 h, centrifuged, dried, ground and passed through a 300-mesh sieve to finally obtain FeSO₄-STAC-Mt.

A Hitachi SU8010 modelled scanning electron microscope with an energy-dispersive X-ray analyzer was adopted to examine the surface morphology and elemental analysis of the composite. The functionalities present in the synthesized materials were investigated using Fourier-transform infrared (FTIR, IR Affinity-ISWL) spectroscopy under the transmission mode, ranging from 400 to 4,000 cm⁻¹. The X-ray photoelectron spectroscopy (XPS) spectra of the composite before and after adsorption were studied using the Nexsa XPS System. The zeta potential of the samples was measured with a Zetasizer Nano ZS90 system (Malvern, UK). Transmission electron microscopy (TEM) was carried out at 100 kV using a Hitachi JEOL-JEM 2100F.

In order to measure the magnetic change before and after the sorbent process, a vibrating sample magnetometer (VSM) was used to measure the magnetic content of the sorbent. Thermogravimetric analysis (TG STA449F3) provided an insight into the weight loss of the adsorbent at different temperatures and facilitated the determination of the conversion of different structures and stability in the adsorbent. The test conditions were a flow rate of 50 ml/min under N₂ atmosphere and a heating rate of 10 °C/min, while ramping up to 800 °C.

In Figure S1 in Supporting Information, the FeSO₄-STAC-Mt diffraction peaks at 30.2°, 35.6°, 43.3°, 57.3 ° and 62.8° showed magnetite with a cubic spinel structure corresponding to the (220), (311), (400), (511) and (440) crystal planes respectively, which were all belonged to Fe₃O₄ (Siregar et al. 2018). Based on these data, the FeSO₄-STAC-Mt was assumed as a magnetite (Fe₃O₄) phase which has superparamagnetic properties, and hence will facilitate in easier magnetic separation (Irawan et al. 2019). With the introduction of STAC and Fe₃O₄, the characteristic diffraction peaks of Mt weaken or even disappear, and these changes indicate a strong bond between Mt (Si-O-) and magnetite, which eventually makes Mt magnetic. In addition, it is also possible that this is the result of the exfoliation of the Mt lamellar structure at higher alkaline environments (Cottet et al. 2014).

It can be observed in Figure 3(d) that the surface of FeSO₄-STAC-Mt was rougher than STAC-Mt and clearly covered with Fe₃O₄ particles, indicating that Fe₃O₄ was successfully loaded on the Mt (Chen et al. 2017). This is explained by the fact that Mt (negative charge) and Fe₃O₄ (positive charge) are more easily bonded together by electrostatic attraction.

TEM photography of the sample (In Figure S2 in Supporting Information) showing the presence of Fe₃O₄ particles, indicated that the nanoparticles contained in the synthesis were clustered together and showed a rhombic structure with a nanometer size of approximately 20 nm. This suggested that magnetite particles could be loaded onto the surface of montmorillonite by a single alkaline oxidation of ferrous sulphate, giving it magnetic properties.

The sample FeSO₄-STAC-Mt showed an increase in both the anti-symmetric peaks and symmetric stretching vibrational peaks in the methylene group at 2,922 cm⁻¹ and 2,852 cm⁻¹ compared with STAC-Mt. This was due to the modification of the inorganic iron salts, which enhanced the peaks at these two locations. In Figure 4(b), the attachment of Fe₃O₄ in the Mt structure was determined by the absorption band at 441 cm⁻¹, representing the Fe-O symmetric stretching vibration. Here, 621 cm⁻¹ represents Fe-O-Fe and the peak at 3,400 cm⁻¹ is assigned to Fe-OH, which mean that Fe₃O₄ is loaded onto the composite modified Mt (Irawan et al. 2019; Fatimah et al. 2021). In summary, the results of FT-IR analysis confirmed the successful modification of montmorillonite by quaternary ammonium salts and magnetite.

The key factor in determining the adsorption capacity is the surface area of the material, as the surface area can positively determine the number of effective collisions between reactants and functional sites (Muhammad et al. 2019). Table 1 shows that the specific surface area and total pore volume of the modified adsorbent were both smaller than those of the original Mt, which were caused by the addition of STAC to create hydrogen bonding (Shoukat et al. 2017). Conversely, the subsequent loading of Fe₃O₄ onto the Mt led to a decrease in pore diameter (Table 1). This is probably due to the insertion of ionic species within the Mt layer spacing and the blockage of interlayer channels, confirming the successful grafting of STAC and Fe₃O₄ in the clay matrix (Huang et al. 2017).

Figure 7(b) shows that the high resolution spectra of Cr2p had characteristic peaks at 588.97 eV (Cr2p_1/2), 576.47 eV (Cr2p_3/2), 585.57 eV (Cr2p_1/2), and 579.07 eV (Cr2p_3/2), corresponding to Cr(III) and Cr(VI), respectively. The fitting results showed that Cr(VI) and Cr(III) in FeSO₄-STAC-Mt accounted for 57.85% and 42.15% of the total Cr, respectively, indicating that some Cr(VI) reduction reactions occurred during the adsorption process. The high-resolution spectrum of Fe2p showed peaks at 710–726 eV corresponding to Fe₃O₄ particles loaded on montmorillonite (Zheng et al. 2017). Combined with the absence of new functional groups in the FTIR spectra, it is inferred that Fe(II) reacts with Cr(VI) as an electron donor to reduce it to Cr(III).

In the Langmuir isotherm equation, the dimensionless constant-separation factor R_L (Equation (4)) can be used as a measure of whether an adsorption process can occur thermodynamically. When 0 < R_L < 1, it means that the adsorption reaction can occur.

The Langmuir (Figure 13(a)), Freundlich (Figure 13(b)) adsorption isotherm model are used to explore the nature and mechanism of adsorption. Figure 13 compiles the results of the isothermal study as well as the results of the Langmuir and Freundlich linear fits for FeSO₄-STAC-Mt. The isothermal parameters derived from the linear plots are listed in Table 2.

Experimental data yielded maximum sorption values that were close to the q_m obtained from the Langmuir isotherm model. The q_m values obtained by linear fitting of the Langmuir model were 36.5065 mg/L. These results indicate that the adsorption was monomolecular layer adsorption and the values obtained were 1/n < 1, indicating that the adsorption was feasible (Irawan et al. 2019; Ain et al. 2020). This showed that the incorporation of cationic surfactants with multifunctional groups and the impregnation of magnetic nanoparticles were responsible for the excellent adsorption performance of the modified clay minerals, respectively. Conversely, the ultra-high swelling properties and extraordinary hydrophilicity of the modified Mt were also responsible for its enhanced adsorption capacity.

Adsorption kinetics is an important parameter for understanding the mechanism of Cr(VI) adsorption on adsorbent surfaces. In this study, the experimental data were further evaluated by using two kinetic models, pseudo-first-order and pseudo-second-order models, respectively (Haounati et al. 2021).

Response surface methodology (RSM) is considered to be an effective condition for analysing complex response conditions. The analysis was carried out using response surfaces in three dimensions to observe the correlation between the three factors mentioned above over the range of the experiment. Figure 15 shows the response surface 3D plots. From Figure 15 it can be seen that the curves exhibit an upward convex shape and that extreme value points can be found, indicating that the horizontal range was chosen reasonably well. All three factors can be seen to be significant in Table S1 of the supplementary material, indicating that all three factors have some influence on the adsorption of Cr(VI). In this paper, numerical optimisation was carried out by the direct search method and the optimum adsorption conditions were optimised for a time of 79 min, a pH of 5.3 and an adsorbent dosage of 0.22 g. The optimised best response value of 98.95% was obtained. Finally, three validation experiments were conducted to judge the optimisation results and the average removal rate obtained was 98.03%. The results showed a very good agreement between the predicted response experiments and the actual values.

In this work, the adsorbent was attached with magnetic properties, which can be easily recovered in regenerative use, further realizing the economic and efficient utilization of the adsorbent. The Cr(VI) adsorbed FeSO₄-STAC-Mt was weighed in a beaker, and NaOH (0.1 M) eluents were added into it, soaked for 5 h–8 h, washed, filtered and dried to make regenerated adsorbent. The reason for choosing NaOH is that when sodium hydroxide is used as the eluent, chromium ions will react with hydroxide ions to form a chromium hydroxide precipitate and, if the hydroxyl concentration is high, chromium ions will continue to react with hydroxide ions to form Cr(OH)₄⁻ and Cr(OH)₄⁻ can be dissolved in water and easily desorbed from the adsorbent surface, thus improving the cycle efficiency. NaCl (0.1 M) and HCl (0.1 M) were also used as eluents for comparison with NaOH.

This study demonstrates the feasibility of using STAC-Mt and FeSO₄-STAC-Mt as adsorbents for the removal of chromium from simulated waste water solutions. The following conclusions were drawn from this study.

• FeSO₄-STAC-Mt, a magnetic organic composite material synthesised using alkaline ferrous sulphate oxide, achieved up to 98.03% removal of Cr(VI) and performed better than STAC-Mt.

• XRD demonstrated that the composite modification increased the layer spacing of the modified montmorillonite from 1.26 nm to 2.03 nm. SEM and TEM clearly showed the loose structure of the modified montmorillonite and well dispersed diamond-shaped magnetic particles.

• The average pore size increased from 13.888 nm to 27.563 nm, it indicated that the modification expanded the pore size by introducing STAC between the Mt layers. Zeta potential indicated that two of the modified adsorbents exhibited positively charged nature and had significant adsorption effect on chromium anion.

• It was shown that the maximum adsorption of chromium by FeSO₄-STAC-Mt was 43.98 mg/g at pH = 6, and the amount of adsorbent was 0.2 g.

• The Langmuir isotherm is most consistent with the monomolecular layer adsorption of chromium by FeSO₄-STAC-Mt. Also, the adsorption of Cr on FeSO₄-STAC-Mt is consistent with quasi-secondary adsorption kinetics indicating that the process is chemisorption.

• Desorption experiments showed that the adsorption of Cr(VI) on FeSO₄-STAC-Mt using NaOH eluent (0.1 M) decreased to less than 72.68% after five adsorption–desorption cycles. This demonstrates that magnetic particle composites are easy to separate and are stable.

This work was supported by the key research and development program of Shaanxi, China (2018GY-067).

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.