A comparative study of the efficiency of ion exchange and extraction chromatography for the separation of milligram amounts of scandium from gram amounts of calcium

Abstract

The growing demand for scandium in various applications such as aerospace, special alloys, electronics and the nuclear industry, has lead to the need to finf more effiecient and cost-effective methods for its separation. The radioactive isotopes of scandium, in particular Scandium-44 and Scandium-47, have generated interest for use in radiopharmaceuticals, for both medical imaging and radiotherapy. It is commonly produced using a cyclotron in a calcium or sometimes a titanium based irradiation target. As the radiopharmaceutical use of scandium radionuclides commonly requires chelation, scandium needs to be separated from the target matrix. This is most often carried out either via extraction chromatography using a suitable solid phase or ion exchange chromatography. The main focus of this study was to compare two methods, namely ion exchange chromatography and extraction chromatography, for the separation of milligram amounts of scandium from gram amounts of calcium. For extraction chromatography (Solvent Impregnated Resins (SIR), unimpregnated XAD-4 resin (pure Amberlite XAD-4 resin) was used as a preliminary sorbent for the sorption of scandium and calcium from aqueous solutions. Owing to a low% sorption (about 70%) obtained, the extractant, Di-2-Ethylhexyl phosphoric acid (D2EHPA), was impregnated in the Amberlite XAD-4 resin for the sorption and separation of scandium and calcium from aqueous solutions. The Di-2-Ethylhexyl phosphoric acid - Amberlite XAD-4 impregnated resin (D2EHPA – XAD-4 impregnated resin) was prepared using the dry method of impregnation. The optimal conditions for the impregnation of the extractant (D2EHPA) were investigated using different volumes of the extractant (D2EHPA), ranging from 1 mL to 25 mL. The maximum value for the distribution coefficient was observed at a volume of 12mL of the extractant, which is approximately 11,500 g D2EHPA/g XAD-4. X-Ray Diffraction (XRD), Fourier Transform Infrared Spectroscopy (FTIR), Brunauer-Emmett-Teller (BET), Scanning Electron Microscopy (SEM), and Transmission Electron Microscopes (TEM) techniques were used to characterize the physical properties of the unimpregnated and impregnated resins. Parameters such as sorbent dosage, pH, contact time, and initial metal ion concentration were investigated for the batch sorption studies of scandium and calcium using the D2EHPA – XAD-4 impregnated resin. The sorption studies were quantified using Inductively Coupled Plasma-Optical Emission Spectroscopy (ICP-OES). Experimental equilibrium data were subjected to four isotherm models (Langmuir, Freundlich, Temkin and Dubinin-Radushkevich). The kinetic reactions were assessed using pseudo-first order, pseudo-second order reactions, and the Weber Morris diffusion model.

The XRD spectra obtained for the unimpregnated and impregnated resin before adsorption, and impregnated resin after the adsorption showed no significant differences. The FTIR spectra showed minor changes in the P=O bond (1225 cm-1) and the P-O-C stretching (1024 cm-1) for the D2EHPA - XAD-4 impregnated resin compared to unimpregnated XAD-4, indicating the interaction between D2EHPA and the Amberlite XAD-4 resin. A decrease in surface area and an increase in pore size of the impregnated XAD-4 resin compared to the unimpregnated resin was observed, further supporting the impregnation of the extractant (D2EHPA) into the pores of the XAD-4 resin. The aggregated morphology of the unimpregnated resin appeared more agglomerated after impregnation with D2EHPA in both the SEM and TEM results, confirming the uptake of D2EHPA within the resin structure. Batch adsorption studies demonstrated that the impregnated resin has a stronger affinity for scandium at a lower pH value (pH=2) compared to calcium (pH=4). The obtained sorption capacity for scandium, at a concentration of 2.0 mg/L, was in agreement with the maximum sorption capacity in the pH studies. The acquired sorption capacity for calcium (0.2428 mg/g) at a concentration of 10.0 mg/L was closely compared with the maximum sorption capacity in the pH studies. A scandium sorption efficiency of 100%, with a sorption capacity of 0.1 mg/g, was observed within the first 6 minutes. The maximum sorption efficiency (100%) for calcium was achieved within 15 minutes. The Langmuir isotherm provided the best fit for the scandium sorption data, and the Dubinin-Radushkevich isotherm fitted the calcium sorption data best. The sorption energy calculated from the Dubinin-Radushkevich plot for both Sc and Ca metals could be the confirmation of a chemical mechanism process. The adsorption data for both Sc and Ca followed pseudo-second-order reaction kinetics, and theoretically, the second-order kinetics is customarily associated with chemisorption. Theoretical data had a good correlation with experimental data. The multi-linearity shown by the Weber-Morris diffusion model for both metals suggests both pore and film diffusion mechanisms are present and can thus control the sorption process. For ion exchange chromatography, the sorption behaviour of scandium and calcium was assessed using Dowex 50W-X8 cation exchange resin in various HBr-methanol mixtures. Based on the experimental distribution coefficients and separation factors, a two-component separation of scandium and calcium was performed in a 2M HBr– 60% methanol mixture using a column method. The results showed that 100% of the scandium was absorbed by the resin and all of the calcium was eluted after 30 min (14 fractions) using 2M HBr-60% methanol mixture. Scandium was completely eluted after 14 min (7 fractions) using 5M HCl.