The quantitative and selective extraction of Li + from aqueous solutions is a difficult subject because of a strong hydration of Li + . Up to now, several kinds of Li + -selective ionophores have been synthesized, such as spherands 1 and 14-crown-4 derivatives, 2 most of which have been studied as sensing materials for Li + -selective electrodes. Recently, Severin and coworkers synthesized a new class of ionophores, macrocyclic trinuclear organometallic complexes bridged by 2,3dioxopyridine.

more »

... These complexes are electrically neutral, and some of them function as ligands or extractants for Li + and Na + . One of the features of the trinuclear complexes is that they are self-assembly formed and can be more easily synthesized than the conventional synthetic ionophores. Additionally, it has been shown that the extraction selectivity of the trinuclear complexes of (p-cymene)ruthenium(II) (1) and (pentamethylcyclopentadienyl)rhodium(III) (2) for Li + over Na + is comparable to, or higher than, that of a commercially available 14-crown-4 derivative. 7 However, their extraction ability and selectivity are not sufficient for the quantitative extraction and separation of Li + from Na + by a single extraction process. In this communication, we describe a newly synthesized trinuclear complex of (1,3,5-trimethylbenzene)ruthenium(II) bridged by 2,3-dioxopyridine (3), which has a much higher selectivity for Li + in solvent extraction. The structural formulas of 1, 2, and 3 are shown in Fig. 1 . The trinuclear complex 3 was obtained by a room-temperature reaction of [(1,3,5-trimethylbenzene)RuCl2]2 with 2,3dihydroxypyridine. 8 Here, [(1,3,5-trimethylbenzene)RuCl2]2 was prepared by refluxing a suspension of [(p-cymene)RuCl2]2 in degassed 1,3,5-trimethylbenzene. 9 The solid complex was relatively stable, even in air, and could be kept for several months under a nitrogen atmosphere. The stability of 3 in a dichloromethane/water system was examined without degassing and nitrogen purging treatments. Complex 3 had an absorption maximum at 333 nm (e = 2.43 ¥ 10 4 dm 3 mol -1 cm -1 ) in dichloromethane at 25˚C. After shaking the dichloromethane solution of 3 (3 ¥ 10 -3 mol dm -3 ) with deionized water, a small amount of precipitate formed at the liquid-liquid interface; however, the light absorption of 3 in the organic phase was practically the same as that before shaking, and remained unchanged within a shaking period of 40 h. It was also found that the partition of 3 from the organic phase to the aqueous one was negligibly small. The extraction of lithium picrate (1 ¥ 10 -3 mol dm -3 ) or sodium one (1 ¥ 10 -2 mol dm -3 ) in an aqueous solution with a dichloromethane solution of 3 (6 ¥ 10 -4 -6 ¥ 10 -3 mol dm -3 ) was conducted at 25˚C in a similar manner to that described previously. 7 The distribution ratio (D) was calculated from the alkali metal concentrations in both the organic and aqueous phases, determined by atomic absorption spectrophotometry. The recovery of the alkali metal from the two phases was always quantitative. The shaking time required to attain equilibrium was about 10 h; this time is longer than that with 1 (ca. 1 h), 7 but shorter than that with 2 (ca. 60 h) 7 under similar experimental conditions. Therefore, the order of the extraction rate is 1 > 3 > 2. Rapid Communications A macrocyclic trinuclear complex of (1,3,5-trimethylbenzene)ruthenium(II) bridged by 2,3-dioxopyridine was synthesized, and the extraction properties for lithium and sodium picrates were investigated in a dichloromethane/water system at 25˚C. The complex was found to have extremely high extractability and selectivity for lithium picrate; the logarithmic values of the extraction constants are 5.86 and 2.63 for Li + and Na + , respectively. By using this complex as an extractant, nearly quantitative extraction and separation of Li + from Na + could be achieved by a single extraction. Fig. 1 Structural formulas of macrocyclic trinuclear complexes.