Abstract

DFT calculations are reported on the affinity of An(III) (An=actinide) and Ln(III) (Ln=lanthanide) ions for ligands with nitrogen-donor and sulfur-donor groups in aqueous solution, aimed at evaluating such donor groups as the basis for separating Am(III) from Ln(III) ions in the processing of nuclear waste. DFT calculations of ΔG(DFT) at 298K for the gas-phase reactions: [M(H2O)6]3+(g)+L(g)→M(H2O)5L]3+(g)+H2O(g), where L is NH3 and M is a variety of Ln(III), An(III) ions, and UO22+ and NpO22+ are reported, as well as for the corresponding reactions for the nona-aqua ions for Ln(III) and An(III) ions. Also reported are ΔG(DFT) for the reactions for L=H2S, as representative of thioethers in aqueous solution, for M=Ln(III) and An(III) ions, as well as Al(III), Ga(III), In(III), Tl(III), Bi(III), Fe(III), and Cr(III). ΔG(DFT) for formation of NH3 complexes in the gas-phase correlates well with ΔG(aq) values for formation of the corresponding complexes in aqueous solution. The logK1(NH3) values were predicted by the equation: logK1=Ea·Eb+Ca·Cb (R.D. Hancock, A.E. Martell, Chem. Rev. 89 (1989) 1875.), where E and C are empirical parameters representing the tendencies of the acids (‘a’) and bases (‘b’) towards ionic and covalent bonding respectively. Correlations involving ΔG(DFT) for the reactions for L=H2S allow for prediction of logK1(R2S) for complexes of thioethers (R=CH2CH2OH). The importance of predicted logK1(NH3) values lies in the fact that logK1 for polydentate N-donor ligands correlates with logK1(NH3) for a series of metal ions, so that logK1(NH3) for any metal ion is an important consideration in ligand design. In the case of saturated N-donor ligands, this is a simple linear relationship, which suggests that saturated N-donors could lead to high Am(III)/Ln(III) selectivities. For ligands that contain pyridyl-type donors, such as 1,10-phenanthroline or terpyridyl, the correlation of logK1 with logK1(NH3) is separated into two linear relations for M(II) and M(III) ions, with M(III) ions having lower affinity for pyridyl donors than do M(II) ions of similar logK1(NH3). The low slope of, for example, logK1(terpyridyl) versus logK1(NH3) for M(III) ions suggests that ligands based only on pyridyl-type donor groups may not be able to produce large Am(III)/Ln(III) selectivities. The calculations on thioether complexes also suggest that such weakly basic S-donors would produce only weak Am(III)/Ln(III) selectivity, but that more strongly basic thiol-type donors might produce larger selectivities.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call