Removal of uranium(VI) ions from aqueous solutions using Schiff base functionalized SBA-15 mesoporous silica materials

• SBA-15 mesoporous silicas modified by Schiff bases ligands are prepared and characterized.
• These adsorbents efficiently, selectively and rapidly removed uranyl ions from water solutions.
• The procedure allows extraction of trace amounts of U(VI) with a preconcentration factor of 200.
• Adsorption-desorption studies show the multiple uses of the adsorbents.
• The adsorption data were suitably described by Langmuir isotherm.

Functionalized SBA-15 mesoporous silica particles, bearing N-propylsalicylaldimine and ethylenediaminepropylesalicylaldimine Schiff base ligands, abbreviated as SBA/SA and SBA/EnSA respectively, were prepared and characterized by FT-IR, elemental analysis, TGA, XRD, TEM and SEM techniques. The potentials of these adsorbents were examined by using them in solid phase extraction of U(VI) ions from water samples. It is shown that 20 mg of SBA/SA or SBA/EnSA can remove rapidly (∼15 min) and quantitatively uranium(VI) ions from 10 to 200 mL of water solutions (pH 4) containing 0.2 mg of the ions, at 25 °C. The adsorbed ions were stripped by 1 mL of dilute nitric acid solution (0.1 mol L−1). It means that the studied adsorbents are able to be used for removal and concentration of uranyl ions. This allowed achieving to a concentration factor of 200 for uranyl ions. The variation in the ionic strength in the range 0–1 mol L−1 did not affect the extraction efficiencies of the adsorbents. The adsorbents showed selective separation of uranyl ions from Cd2+, Co2+, Ni2+, Mn2+, Cr3+, Ba2+, Fe3+ and Eu3+ ions. Thermodynamic investigations revealed that the adsorption of uranyl ions by the adsorbents was spontaneous and endothermic. The Langmuir model described suitably the adsorption isotherms. This model determined the maximum adsorption capacity of the adsorbents SBA/SA and SBA/EnSA as 54 and 105.3 mg uranyl/g adsorbent, respectively. The kinetics of the processes was interpreted by using Pseudo-second-order model.