Separation of Some Lanthanide (III) Ions by using 18-Crowns-6 Derivatives from Acidic Solution

MATERIALS AND METHODS

This research project was conducted from February 2008 to April 2009 in Fule Cycle Research School in Esfahan and was sponsored by Nuclear Science and Technology Research Institute.

Reagent grade nitrate of lanthanides (all from Merck) were of the highest purity available and used without any further purification. Macrocycles 18C6, DB18C6, DCH18C6, DBP18C6; (Fig. 1a-d), were purchased from Merck. Reagent grade Dimethylsulfoxide (DMSO) and Trichloroacetic Acid (TCA) from Merck were also used. Test solutions for the extraction experiments were prepared by diluting the stock solution, which was a solution containing 1000 ppm of each lanthanide ions in 1 M HNO₃. The stock solution was diluted with ultra pure deionized water and trichloroacetic acid added until the concentration of each metal ions became 100 ppm. Trichloroacetic acid with concentrations of 0.5, 1, 1.5, 2, 2.5 and 3 mol L^-1 were used. The macrocycles concentration used in the extraction solutions was 0.05 mol L^-1; the solvent used was DMSO.

Fig. 1:	Chemical structure of derivative 18 membered crown ether that used in this study; (a) 18-Crowns-6 (18C6), (b) dibenzo-18-crown-6 (DB18C6), (c) dicyclohexano-18-crown-6 (DCH18C6) and (d) dibenzopyridino-18-crown-6 (DBP18C6)

After the extraction and phase separation by 20 mL chloroform, back extraction with ultra pure deionized water was carried out.

Adsorption experiments were performed in 50 mL glass bottles containing 10 mL of different concentrations of each lanthanide ions and 10 mL of crowns solution to evaluate ion separation factor. To investigate the effect of trichloroacetic acid concentration different concentration of this acid were added to each bottle. The trichloroacetic acid concentration was determined by titration of the back extraction solutions with 0.1 M of NaHCO₃ solution. All batch experiments were carried out at ambient temperature (25±2°C) and all the solutions were shaken on a magnetic stirrer for 5 min. The quantities of adsorbed ions were calculated by the difference of the initial and residual amounts of ions in solution divided by ion analysis.

The measure of the extraction efficiency of macrocycles represented by the distribution coefficient factor D in the whole extraction of 5 lanthanide ions. The degree of metal extraction and selectivity criterion showed by the separation factor D₁/D₂, where D₁ and D₂ are the distribution coefficients of cation 1 and cation 2, respectively.

The determination of Lanthanides was performed utilizing an ion chromatograph system (Dionex, DX-100 USA). This instrument was equipped with a gradient pump module, a conductivity and UV-Vis detector and with the use of Dionex Ion Pac CS3 + CG3 column (sulfonic acid type), α-hydroxyisobutyric acid (a-HIBA) as an eluent and Arsenazo III as color forming reagents, was elaborated. The injection volume was 50 μL and three replicate injections were made for each sample.

RESULTS

Interesting results were obtained from the extraction experiments in the presence of trichloroacetic acid. It was found from new data that 18C6 and DCH18C6 extracted the lanthanide ions under these conditions; the extraction by DB18C6 and DBP18C6 was insignificant. Distribution factors of the total of 5 lanthanide ions as a function of trichloroacetic acid concentration in the extraction by 0.05 M 18C6 showed in Fig. 2. The curves for each of 5 cations shows a rising trend with increasing trichloroacetic acid concentrations. Figure 3 represent extraction of 5 lanthanides ions with DCH18C6. In Fig. 3, D factor in comparison shows the same trend as Fig. 2 except that maximums for all ions were diminished. Analogous curves were obtained for DB18C6 and DBP18C6. In Fig. 4 extraction of 5 cations with DB18C6 are displayed. Also in Fig. 4 as we can see decreasing trend of D factor by change of ligand.

Fig. 2:	Distribution factor of La3+, Ce3+, Pr3+, Eu3+ and Er3+ as a function of Trichloroacetic Acid (TCA) concentration in the extraction of them by 0.05 M 18C6

Fig. 3:	Distribution factor of La3+, Ce3+, Pr3+, Eu3+ and Er3+ as a function of TCA concentration in the extraction of them by 0.05 M DCH18C6.

Figure 5 shows extraction of 5 lanthanides ions with DBP18C6. we can see least D factor for Fig. 5. It is interesting that the lanthanides distribution coefficients values decreased with increasing atomic number. Figure 6 which is a plot of Log D factor versus Log Crown ethers concentrations, showed that the metal: crown ratio in complex for all cases is 1:1.

Fig. 4:	Distribution factor of La3+, Ce3+, Pr3+, Eu3+ and Er3+ as a function of TCA concentration for the extraction of them by 0.05 M DB18C6

Fig. 5:	Distribution factor of La3+, Ce3+, Pr3+, Eu3+ and Er3+ as a function of TCA concentration for the extraction of them by 0.05 M DBP18C6

Table 1 showed the numerical values of lanthanide ions separation factor in DMSO/water system and in the presence of trichloroacetic acid and nitric acid solution.

Table 1:	Numerical values of lanthanide ions

Fig. 6:	Distribution factor of La3+, Ce3+, Pr3+, Eu3+ and Er3+ as a function of 18C6 concentration in the extraction of them in the presence of 2 M TCA

Also the trichloroacetic acid distribution in the lanthanides extraction were studied. We know that the extraction of the lanthanide ions in pure nitric or perchloric acid did not occur (Demin et al., 2006). The trichloroacetic acid distribution between the two phases were not affected by the type and concentration of the macrocycles and the lanthanide metals concentration in the organic phase in the extraction process. In all cases, the trichloroacetic acid concentration in the organic phase was greater than the total lanthanide ions concentration.

DISCUSSION

As a general reaction for extraction of lanthanide ions, the extraction process represented as:

(1)

So, e.g., for La³⁺:

(2)

reaction constant of extraction process is:

(3)

and distribution constant is:

(4)

In this manner, the extraction constant became:

(5)

Then, equation of a straight line showed by:

(6)

where, L_org = L_total-m c_org, m = 3 and c_org is the metal concentration in the organic phase. Here, mc_org«L_total.

By a plot of (D/[A¯]^m_aq) versus L_org for the extraction of lanthanide ions mixture with four crown ethers (Fig. 6), the n value for the extracted complex, from the slope of the linear plot were obtained. Each plot has one slope, which indicates that a 1 : 1 complexes is formed. Thus the metal: crown ether ratio in the extracted complex in all cases is 1:1 which in this regard supports previous work result. In addition to the slope method, also the stoichiometry of the complexes of these elements were examined in our previous work (Zolgharnein et al., 2003) by the mole ratio method. As we reported there, the stability of the La³⁺, Ce³⁺, Pr³⁺, Eu³⁺ and Er³⁺ complexes with the various 18-crowns-6 in DMSO increases in the order La³⁺ > Ce³⁺ > Pr³⁺ > Eu³⁺ > Er³⁺. This order in stability is fairly small when compared with the specificity displayed by the groups IIA ions. Since the best fit of metal ion to ligand cavity is a prime necessary for complex stability. La³⁺ ion with an ionic radius of 1.061 Å should fit nicely inside the cavities of 18-crowns with radii of 1.3-1.6 Å (Ohyoshi, 1985; Jain and Gupta-Bahaya, 1992). The other cations with smaller radii are too loose for the cavities of the 18 membered ring. Therefore weaker complexes were expected. The fact that the trivalent lanthanide ions have radii comparable to Na⁺ and Ca²⁺ leads to the expectation that lanthanide ions would also form stable complexes with 18-crowns. There are at least two factors that may account for the lack of high stability constants in lanthanide complex formation. First, the radii decreases by only ca, 0.2 Å from La³⁺ to Er³⁺. This is much less than the deferences in radii among the ions of groups IA and IIA. Second, using dimethylsulfoxide as the solvent removes difficulties associated with metal ion hydrolysis, but the solvent may specifically interact with metal complexes (Pizer and Selzer, 1983). Extension of these ideas to lanthanide ions suggest that because of their higher coordination number (Bunzli et al., 1982; Habeschuss and Spedding, 1980), lanthanide complexes may interact significantly with the solvent. Comparison of the evaluated distribution coefficient for different 18-crown-6 with lanthanide ions, follow the order 18C6 > DCH18C6 > DB18C6 > DBP18C6.

This sequence of order except for DBP18C6 has been reported before only for potassium ion (Tawarah and Ababneh, 1997), transition metal ions (Alizadeh and Shamsipur, 1993), thallium ion (Parham and Shamsipur, 1993) and alkali metal ions (Frensdorff, 1971; Izatt et al., 1976). Information about the complexation of DBP18C6 is rare (Tawarah and Ababneh, 1997; Shamsipur and Zolgharnein, 2001). The presence of three aromatic moieties in this macrocycle substantially reduced the affinity of the donor atoms toward the lanthanide ions. Introducing the pyridyl unit in to the ring decreases both the flexibility and interaction of the macrocycle with lanthanide ions. This may be due to the pyridino nitrogen not being able to donate to the inner d orbital of lanthanide ions; on the other hand, these cations are so strongly solvated that considerably more energy must be expended in the desolvation step. The contribution of the solvent-complex and even solvent-ligand interactions on the stability of the resulting complexes cannot be ignored. Since, the two main parameters for specificity and affinity of complexation of metal ions with these macrocycles are size and solvating power. Literature review (Izatt et al., 1976, 1985, 1991; Zhu and Izatt, 1990; Alizadeh and Shamsipur, 1993; Frensdorff, 1971; Kashanian and Shamsipur, 1989), clearly shows both two trends for complexation of 18-membered crowns with various metal ions group. These order are 18C6 > DCH18C6 > DB18C6 > DBP18C6 for macrocycles and group IA > group IIA > lanthanide (III) ions. So this trend for similar ionic size cations is Na⁺ (Frensdorff, 1971; Izatt et al., 1976), > Ca²⁺ (Demin et al., 2006), > La³⁺ (Pedersen, 1967; Zolgharnein et al., 2003), for all 18-membered crowns.

The conclusion of this research shows that the 18-crowns-6 and DCH18C6 in presence of trichloroacetic acid extract the lanthanide ions and separate them from each other. These macrocyclic ligands also promote the isolation and recovery efficiency of extraction of the lanthanide ions, in particular when phosphorus and binary extractants are used (Nishihama et al., 2002; Egorova et al., 2005; Zhu et al., 1996).

ACKNOWLEDGMENTS

The authors gratefully acknowledges the NSTRI providing facilities for this study. The authors also thanks Dr. A. Mostofizadeh for useful comments in preparation of this research and M.R. Goodarzi for her helps in data analyses.