Versatile heavy metals removal via magnetic mesoporous nanocontainers

Abstract

Versatile heavy metal ions removal is performed via iron oxide core mesoporous silica shell nanocontainers functionalized with diethylene triamine pentaacetic acid. Magnetic separation from aqueous media is obtained for a dozen of the most toxic heavy metal ions with remarkable efficiencies. Furthermore, this study demonstrates enhancement of the adsorption capacities with the combination of porous and surface functionalization of the nanocontainers.

---

Heavy metal contamination from industrial processes and products has become a public health and environmental concern throughout the world.^1–6 Although there is no clear and scientific definition of heavy metals, it generally refers to metals having a specific density of more than 5 g cm⁻³.¹ In the year 2000, the European Union has legislated that “‘heavy metal’ means any compound of antimony, arsenic, cadmium, chromium(VI), copper, lead, mercury, nickel, selenium, tellurium, thallium and tin, as well as these materials in metallic form, as far as these are classified as dangerous substances.”⁷ The toxicity to human health of heavy metal ions, among which lead, cadmium, mercury and arsenic are considered to be the most dangerous,¹ can damage mental and central nervous functions. Besides, the essentiality of metal ions transport (e.g. copper, zinc, lead) via proteins for cellular biochemical processes in the body, can also become the Trojan horse for toxic metal ions accumulation. As a result, various damages can be caused by heavy metal ions to vital organs such as the lungs, the kidneys, and the liver.^8–11 Therefore, many devices have been elaborated to remove such ions, mainly ion-exchange materials,^12–14 membrane filtration and adsorption,^5,15–17 and colloidal adsorbents.^18–22

Superior efficiencies are obtained via adsorbents,^18,23–25 and porous colloids would be most appropriate due to their high surface area, and the variability of surface modifications.¹⁹ One challenge that colloidal adsorbents face is to separate the particles after ion adsorption. An efficient method utilizes the magnetic properties of iron oxide nanocrystals (Fe₃O₄NCs).^22,26,27 Note that, for such an application Fe₃O₄NCs must be coated in order to avoid iron ions leaching in the waste solutions. Besides, to enhance the pollutant removal capacities of Fe₃O₄NCs, porous coatings could be tailored. Hence, micron-size Fe₃O₄NCs core mesoporous silica shell (Fe₃O₄@MSN) have been designed,^18,28 Fe₃O₄@MSN@chitosan microspheres,²⁹ Fe₃O₄NCs@MSN foams.³⁰ and porous magnetite-carbon nanocapsules.³¹ To our knowledge micro-sized Fe₃O₄@MSN materials were only tested with Pb²⁺, Cu²⁺, Cd²⁺, Hg²⁺, and Ag⁺.^18,28,30

In the present study, we designed Fe₃O₄@MSN nanocontainers functionalized with a tridentate chelate ligand for heavy metal ions removal. The functionalization was performed with a mono-alkoxysilylated derivative of diethylene triamine pentaacetic acid (DTPA). Thus excellent colloidal stability was obtained for an effective magnetic separation in aqueous media at neutral pH. The nanocontainers were found to be effective for twelve of the most toxic heavy metal ions, namely Pb²⁺, Cu²⁺, Zn²⁺, Cd²⁺, Tl⁺, Pd²⁺, Ni²⁺, Hg²⁺,Cr³⁺, Co²⁺, Mn²⁺, and Sn⁴⁺. Furthermore, the adsorption capacity was enhanced by taking advantage of the porosity of the mesoporous silica framework (Scheme 1).

Scheme 1 Heavy metal removal strategies: surface functionalized Fe₃O₄@MSN NPs (A), the porous functionalized Fe₃O₄@MSN NPs (B), and surface/porous functionalized Fe₃O₄@MSN NPs (C).

First of all, DTPA functionalized magnetic mesoporous nanocontainers were designed with different strategies (Scheme 2). Spherical Fe₃O₄NCs of 21 ± 2 nm were synthesized via thermal decomposition at 340 °C of hydrated iron oxide with oleic acid in docosane. Then, the mesoporous silica layer was grown in a sol–gel process at 80 °C with a sodium hydroxide catalyst in a cetyltrimethylammonium bromide–water mixture for 2 h. Three types of nanocontainers were obtained (Scheme 2), surface functionalized NPs (Fe₃O₄@MSN-L), porous functionalized NPs (Fe₃O₄@MSL), and the combined surface/porous functionalized NPs (Fe₃O₄@MSL-L). The Fe₃O₄@MSN-L NPs were obtained through post-functionalization of Fe₃O₄@MSN by the alkoxysilylated DTPA. IR spectra demonstrated the ligand functionalization (ESI Fig. S1 and S2†), with the apparition of ν_Si–C at 1204 cm⁻¹, ν_N–C at 1398 cm⁻¹, ν_C=O at 1644 cm⁻¹, and ν_C–C from at 1888 to 1942 cm⁻¹ vibration modes corresponding to the DTPA. Solid state nuclear magnetic resonance (NMR) ¹³C and ²⁹Si CPMAS spectra on a control of ligand functionalized MSN NPs further confirmed the surface functionalization (Fig. S3†). Elemental analysis and energy dispersive spectroscopy quantitative composition determinations measured 37.8 and 31.0 wt% of ligand functionalization respectively (ESI Fig. S4†). The porous functionalization of Fe₃O₄@MSL was constructed in two steps. Initially, a mixture of the tetraethoxysilane silica precursor (TEOS) and (3-aminopropyl)trimethoxysilane moiety was condensed on Fe₃O₄NCs, leading to iron oxide core mesoporous amino-silica shells (Fe₃O₄@MSNH₂, see Scheme 2). Then, the DTPA precursor was reacted in the pores to obtain Fe₃O₄@MSL NPs. This grafting is also supported by the IR spectra of Fe₃O₄@MSNH₂ and Fe₃O₄@MSL, with the disappearance of the sharp ν_N–C vibration band at 1385 cm⁻¹, and the appearance of the ν_C=O at 1631 cm⁻¹, and the ν_C–C at 1850–1961 cm⁻¹ (ESI Fig. S5 and S6†). Finally, Fe₃O₄@MSL-L was designed via the grafting of the alkoxysilylated DTPA on Fe₃O₄@MSL nanocontainers.

Scheme 2 Synthetic pathways to design ligand functionalized Fe₃O₄@MSN NPs.

The Fe₃O₄@MSN nanocontainers were characterized via transmission and scanning electron microscopies (Fig. 1A and B). Spherical radial mesostructures were observed, and most nanocontainers were composed of a 20 nm iron oxide core, leading to an overall size of 130 nm. The size monodispersity of the NPs was confirmed by dynamic light scattering (DLS) measurements (Fig. 1D). The mesoporous framework was analyzed via nitrogen-adsorption–desorption technique, and the typical type IV isotherm profile was acquired (Fig. 1E). A surface area of 530 m² g⁻¹ was calculated by the BET theory, and a pore diameter of 3 nm was determined. Similarly, the Fe₃O₄@MSNH₂ nanocontainers were fully characterized and found to be 50 nm monodisperse and non-aggregated mesoporous NPs (ESI Fig. S7†).

Fig. 1 Transmission and scanning electron microscopy of Fe₃O₄@MSN NPs (A, B and C respectively), the DLS size distribution (D), and N₂-adsorption–desorption analysis (E).

The heavy metal ions removal was then magnetically carried out with Fe₃O₄@MSN-L NPs (Fig. 2). All the experiments were performed in aqueous media at neutral pH. The nanocontainers were mixed in deionized water with various metal salts excess to continually saturate the solution, and the solution was stirred overnight at room temperature. Afterwards, a magnet was placed on the side of the reaction tube for 10 minutes, the supernatant and the metal salts were removed. Finally, the NPs were washed thrice in water, twice in ethanol, and once in acetone, to remove the non-adsorbed metal residues (see ESI Fig. S8†).

Fig. 2 Nanocontainers, before (A) and after magnetic separation of Fe₃O₄@MSN-L∩Cu²⁺ NPs (B).

The metal adsorption capacities (Fig. 3) were determined by statistical energy dispersive spectroscopy measurements (ESI, Fig. S9, Table S1–S12†). Interestingly, the ligand functionalized nanocontainers were found to be a versatile platform for many different kinds of metal. Moreover, such a separation was very efficient for certain metal ions, such as palladium(II), copper(II), chromium(III), tin(IV), nickel(II), and zinc(II), with adsorption capacities ranging from 1000 to 2400 μmol g⁻¹. Note that, the absence of chloride on the EDS spectra for most of the NPs using metal chloride precursors validates the adsorption-washing procedure that was used to remove precipitated salts. Not surprisingly, the exceptions are the magnetic chromium chloride as well as the very soluble palladium chloride, probably partly accumulated in the pores. Thus, comparing solely the surface adsorption capacities of coated iron oxide nanocontainers (to avoid iron ions pollution), our system is very competitive with the current literature. Comparing surface functionalized amino-MSN with Fe₃O₄@MSN-L nanocontainers, the adsorptions of Ni²⁺, Cd²⁺, and Pb²⁺ were nearly doubled.²¹ Consequently, the DTPA ligand is most appropriated to pollutant removal green materials and nanomaterials. Moreover, the selectivity of the ion removals was studied and found to be preferential towards palladium (58%), copper (21%), and lead (7%) in a mixture of all types of metals (Fig. S10†) which further illustrates the applicability of these nanocontainers.

Fig. 3 Heavy metal removal capacities of t Fe₃O₄@MSN-L NPs.

The metal ions could then be removed via acidic treatment, as reported in the literature.²⁸ At pH 4.5 the efficiency of the nanocontainers regeneration was 90 and 99% for copper and zinc ions respectively after one desorption process, according to inductively-coupled plasma analysis (Fig. S11†). Note that, the iron oxide weight percent remained unchanged after the acidic extraction, thus showing the stability of the core thanks to the silica protective layer. Besides, the removal was found to be reversible with 99% efficiency after 2 cycles (Fig. S12†).

The enhancement of the adsorption capacities was considered for copper and zinc ions. The three types of ligand functionalized nanocontainers, Fe₃O₄@MSN-L, Fe₃O₄@MSL, and Fe₃O₄@MSL-L were compared (Fig. 4). Surprisingly, the adsorption capacities of Fe₃O₄@MSL were smaller than that of Fe₃O₄@MSN-L, with 503 and 570 μmol g⁻¹ of zinc and copper respectively in the former, and 797 and 1122 μmol g⁻¹ in the later. This may be due to steric pore hindrance of amine groups which lowered the porous surface, as well as the steric effect occurring when partial metal adsorption has occurred. However, the combination of the surface and porous ligand grafting in Fe₃O₄@MSL-L led to proportional enhancement of the adsorption capacities, reaching 1363 and 1904 μmol g⁻¹ of zinc and copper respectively. In both cases, an additional 40 percent of metal ions was adsorbed. Note that, the correlation between the amount of ligand and the adsorbed quantity of heavy metals was confirmed by varying the amount of DTPA in the pores and on the surface (ESI Fig. S13†). Furthermore, the metal ions were not merely physically entrapped in the MSN framework, as suggested by the 5 to 9 cm⁻¹ shifts in the carbonyl vibration modes (ESI Fig. S14–S17†). Besides, the observed increase of the intensity of the ν_C=O bands is in accord with conformational change associated with the DTPA-Mⁿ⁺ complex formations. That is why various colors were observed for each type of metal complexes in the NPs framework.

Fig. 4 Heavy metal removal capacities optimization of the various nanocontainers strategies.

Conclusions

In summary, we have developed a novel diethylene triamine pentaacetic acid ligand functionalized with monodisperse Fe₃O₄@MSN and Fe₃O₄@MSNH₂ nanocontainers. Such NPs were found to be capable of a versatile and efficient magnetic separation on a dozen of heavy metal ions. The adsorption capacities tended to be higher for the metal having high valence and electronic affinities, such as Pd²⁺ (2341 μmol g⁻¹), Co²⁺ (1452 μmol g⁻¹), Cr³⁺ (1362 μmol g⁻¹), Sn⁴⁺ (1198 μmol g⁻¹), and Ni²⁺ (1128 μmol g⁻¹). Palladium(II), copper(II), and lead(II) ions were selectively removed in a mixture of all types of ions, which makes the nanocontainers selective platforms for heavy metal removal. Besides, the nanocontainers could be regenerated and reused with high efficiencies. Additionally, the adsorption capacities have been enhanced of 40 percent by the porous and surface functionalization of the nanocontainers. We are currently investigating the optimization of the adsorption capacity with larger pores and higher surface areas, in order to avoid the steric hindrance inside the mesopores. It is envisioned that such Fe₃O₄@MSL-L NPs could be powerful tools to efficiently clear heavy metal liquid waste.

Acknowledgements

Financial support from ANR Mechanano and the Université des Sciences et de la Technologie Houari Boumediene are gratefully acknowledged.

Notes and references

1. L. Järup, Br. Med. Bull., 2003, 68, 167–182.
2. X. Liu, Q. Song, Y. Tang, W. Li, J. Xu, J. Wu, F. Wang and P. C. Brookes, Sci. Total Environ., 2013, 463–464, 530–540.
3. M. Pizzol, P. Christensen, J. Schmidt and M. Thomsen, J. Cleaner Prod., 2011, 19, 646–656.
4. N. Malik, A. K. Biswas, T. A. Qureshi, K. Borana and R. Virha, Environ. Monit. Assess., 2010, 160, 267–276.
5. F. Fu and Q. Wang, J. Environ. Manage., 2011, 92, 407–418.
6. J. S. Barbosa, T. M. Cabral, D. N. Ferreira, L. F. Agnez-Lima and S. R. Batistuzzo de Medeiros, Ecotoxicol. Environ. Saf., 2010, 73, 320–325.
7. T. C. O. T. E. Communities, Off. J. Eur. Comm., 2000, 226, 3.
8. S. H. Laura, D. K. Thomas, M. Nieuwenhuijsen and L. Jarup, Environ. Health Perspect., 2009, 117, 181–184.
9. N. Johri, G. Jacquillet and R. Unwin, BioMetals, 2010, 23, 783–792.
10. Q. Gu and R.-L. Lin, J. Appl. Physiol., 2010, 108, 891–897.
11. K. Gulati, B. Banerjee, S. B. Lall and A. Ray, Indian J. Exp. Biol., 2010, 48, 710–721.
12. M. Y. Vilensky, B. Berkowitz and A. Warshawsky, Environ. Sci. Technol., 2002, 36, 1851–1855.
13. W. Plazinski and W. Rudzinski, Environ. Sci. Technol., 2009, 43, 7465–7471.
14. G. Mahajan and D. Sud, J. Chem. Environ. Eng., 2013, 1, 1020–1027.
15. M. A. Hasan, Y. T. Selim and K. M. Mohamed, J. Hazard. Mater., 2009, 168, 1537–1541.
16. B. Yu, Y. Zhang, A. Shukla, S. S. Shukla and K. L. Dorris, J. Haz. Mater., 2000, 80, 33–42.
17. W. S. Wan Ngah, L. C. Teong, R. H. Toh and M. A. K. M. Hanafiah, Chem. Eng. J., 2013, 223, 231–238.
18. G. Li, Z. Zhao, J. Liu and G. Jiang, J. Haz. Mater., 2011, 192, 277–283.
19. M. Hua, S. Zhang, B. Pan, W. Zhang, L. Lv and Q. Zhang, J. Haz. Mater., 2012, 211–212, 317–331.
20. J. Wang, Y. Shao, J. Liu, Z. Xu and D. Zhu, J. Colloid Interface Sci., 2010, 349, 293–299.
21. M. Najafi, Y. Yousefi and A. A. Rafati, Sep. Purif. Technol., 2012, 85, 193–205.
22. M. Chen, Y. Liu and Y. Hao, Chem. Eng. J., 2013, 218, 48–54.
23. S. Dib and M. Boufatit, Desalination, 2009, 5, 106–110.
24. F. Mohammed-Azizi, S. Dib and M. Boufatit, Desalination, 2013, 51, 4447–4458.
25. H. Assameur and M. Boufatit, Desalination, 2012, 45, 315–323.
26. L. Wang, J. Li, Q. Jiang and L. Zhao, Dalton Trans., 2012, 41, 4544–4551.
27. P. Xu, G. M. Zeng, D. L. Huang, C. L. Feng, S. Hu, M. H. Zhao, C. Lai, Z. Wei, C. Huang, G. X. Xie and Z. F. Liu, Sci. Total Environ., 2012, 424, 1–10.
28. Q. Yuan, N. Li, Y. Chi, W. Geng, W. Yan, Y. Zhao, X. Li and B. Dong, J. Haz. Mater., 2013, 254–255, 157–165.
29. Y. Ren, H. A. Abbood, F. He, H. Peng and K. Huang, Chem. Eng. J., 2013, 226, 300–311.
30. J. Chung, J. Chun, J. Lee, S. H. Lee, Y. J. Lee and S. W. Hong, J. Haz. Mater., 2012, 239–240, 183–191.
31. K. Cheng, Y.-M. Zhou, Z.-Y. Sun, H.-B. Hu, H. Zhong, X.-K. Kong and Q.-W. Chen, Dalton Trans., 2012, 41, 5854–5861.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra01323a

---