Dominique C.
Stepinski
,
Mark P.
Jensen
,
Julie A.
Dzielawa
and
Mark L.
Dietz
*
Chemistry Division, Argonne National Laboratory, Argonne, IL 60439. E-mail: mdietz@anl.gov; Fax: +1 (630) 252-7501; Tel: +1 (630) 252-3647
First published on 10th January 2005
Addition of tri-n-butyl phosphate (TBP) is shown to markedly increase the extraction of strontium from acidic nitrate media into certain 1-alkyl-3-methylimidazolium bis[(trifluoromethyl)sulfonyl]imides by dicyclohexano-18-crown-6 (DCH18C6), the apparent result of the formation of a synergistic adduct between the strontium-DCH18C6 complex and TBP. The magnitude of the synergistic enhancement is shown to depend on the alkyl chain length of the ionic liquid (IL) cation, with the effect diminishing as the cation hydrophobicity increases. The effect also diminishes at high (>50% v/v) TBP concentrations, the likely result of changes in solvent polarity unfavorable to the extraction of metal-crown ether complexes.
Recent work in this laboratory in the area of metal ion separations has examined the possibilities afforded by ionic liquids (ILs),5–11 low-melting organic salts whose unique physicochemical properties (e.g., low volatility, high ionicity) have garnered them intense interest as potential “green” alternatives to conventional organic solvents in a wide range of applications.12–18 Of particular note in our studies is the observation that in contrast to the extraction of strontium-crown ether (CE) complexes into molecular organic solvents, strontium extraction by dicyclohexano-18-crown-6 (DCH18C6) from acidic nitrate media into various 1-alkyl-3-methylimidazolium-based ILs takes place predominantly via a mechanism in which the cationic 1 ∶ 1 metal-CE complex is exchanged for the cationic constituent of the ionic liquid. This is clearly an undesirable pathway if the objective is the development of environmentally benign extraction systems.5 Although this problem can be addressed by employing a sufficiently hydrophobic IL cation (e.g., 1-decyl-3-methylimidazolium) to induce a change in the extraction mechanism,11 increased cation hydrophobicity is accompanied by substantially decreased metal ion extraction efficiency, thus diminishing an important advantage of ILs as extraction solvents over ordinary molecular diluents. The exploitation of synergistic effects offers one potential route to overcoming this difficulty. That is, by judicious combination of extractants, it may be possible to boost metal ion extraction efficiencies into ionic liquids incorporating even relatively hydrophobic cations. To date, the possibility of synergistic interactions between extractants in ionic liquids has not been explored, however. In this report, we examine the use of neutral organophosphorus reagents as synergists in the extraction of alkali and alkaline earth cations by crown ethers into 1-alkyl-3-methylimidazolium-based ionic liquids and consider the implications of the results for the design of improved IL-based systems for metal ion separations.
The strontium coordination environment of representative organic phases was probed with X-ray absorption spectroscopy. Liquid samples containing approximately 0.03 M Sr and varying IL/TBP ratios were placed in 6 mm I.D. polyethylene tubes and mounted at 45° to the incident X-ray beam at beamline 12-BM of the Advanced Photon Source.20 The monochromator energy was calibrated against the first inflection point of the K absorption edge of a Zr foil (17.998 keV) before and after the data collection. No drift in the monochromator energy was detected. Data at the Sr K-edge were collected in the fluorescence mode with a 13-element Ge detector, averaging four to seven scans. The EXAFS signal was extracted and analyzed by standard procedures21 using IFEFFIT22 with the threshold energy (E0) set to the measured first inflection point of the Sr K-edge of the samples, 16.114 keV.
The resulting k3-weighted EXAFS (k = 2.5–11.0 Å−1) were fit in R-space (R = 1.22–4.22 Å) to the theoretical phase and amplitude functions generated with FEFF8.0023 from the atomic positions of Sr(NO3)2(18-crown-6),24 Sr(H2O)2(18-crown-6)2+, or Sr(TBP)2(18-crown-6)2+. In the fitting procedure, the values of the amplitude reduction factor (S02 = 1.0) and the Debye–Waller factors of the Sr coordinated oxygen atoms (σO2 = 0.0136) and the carbon atoms of the crown ether ring (σC2 = 0.0108) were fixed at the values previously determined from the EXAFS of Sr-crown ether complexes.6 The number of coordinated crown ether molecules and the number of coordinated nitrate, water, or TBP molecules; the Debye–Waller factors of all other scattering paths; the average scattering pathlengths; and a single threshold energy shift (ΔE0) were allowed to vary during each fit. The number of parameters varied in each fit was between 7 and 10, depending on the model used (maximum number of floating parameters allowed by the Nyquist criterion = 16).
Previous work concerning the mechanism of strontium ion transfer between acidic nitrate media and [Cnmim][Tf2N] (n = 2–8) ionic liquids in the presence of the crown ether DCH18C6 has shown that strontium partitioning proceeds predominantly via exchange of the cationic 1 ∶ 1 Sr[DCH18C6]2+ complex for the cationic constituent of the IL.5 Subsequent EXAFS investigations have shown that the extracted species has a structure in which water molecules occupy the two axial positions in the complex.6 As is well known, synergistic effects frequently have their origins in the replacement of coordinated water with molecules of the second extractant.25–28 It thus seems reasonable to anticipate that the extraction of strontium by DCH18C6 from acidic nitrate media into these ILs might be more susceptible to synergistic enhancement than is its extraction into a conventional organic solvent (e.g., 1-octanol), in which the extracted complex bears no inner-sphere water molecules.6 In fact, significantly improved strontium extraction efficiency might well be obtained simply by pairing the crown ether with a second neutral extractant.
Among neutral extractants, few have received more attention, particularly in the context of the development of processes for metal ion separation, than tri-n-butyl phosphate (TBP), which has been extensively investigated both as a metal ion extractant and as a solvent or co-solvent.29,30 Earlier work by Visser et al.9 has shown that various hydrophobic ionic liquids (e.g., [C8mim][Tf2N]) can dissolve high (ca. 1 M) concentrations of TBP. As might be anticipated from this result, [C5mim][Tf2N] and TBP are miscible in all proportions. Fig. 1A depicts the effect of varying the proportions of these two reagents on the extraction of strontium by DCH18C6-A from aqueous nitric acid. As can be seen, the distribution ratios for strontium obtained with the TBP–IL mixtures exceed the sum of the values obtained for the individual reagents over a wide range, sometimes by a substantial margin. A 1 ∶ 1 (v/v) mixture of [C5mim][Tf2N] and TBP, for example, yields a DSr of 91.5, more than an order of magnitude greater than the value expected. That the effect is not a peculiarity of this particular combination of ionic liquid and organophosphorus reagent is demonstrated by the results presented in Figs. 1B and C, which depict the results of analogous measurements for [C5mim][PF6]–TBP and for [C2mim][Tf2N]–diamyl amylphosphonate (DAAP), a neutral phosphonate ester that has been employed as a substitute for TBP in various metal ion separations.31,32 In each of these two systems, the extraction of strontium into the mixtures also exceeds that expected from its extraction by the individual reagents.
![]() | ||
Fig. 1 Effect of co-extractant concentration (expressed as % v/v) on the partitioning of Sr-85 between 1 M HNO3 and [C5mim][Tf2N] (panel A), [C5mim][PF6] (panel B), or [C2mim][Tf2N] (panel C) containing 0.1 M DCH18C6-A. |
If the influence of TBP addition upon molecular solvents is taken as a guide, this improvement in strontium extraction may have its origins in any of several effects. When used as a co-solvent in a conventional system (e.g., for the modification of paraffinic hydrocarbons to yield the TRUEX process solvent33), TBP serves primarily to raise the solvent polarity,34 thereby improving the compatibility of the extractant and the metal-extractant complex(es) with the organic phase and thus, increasing the extraction of the metal ion of interest. In the present system, however, addition of TBP, with its dielectric constant of 8.14,29 to [C5mim][Tf2N], whose polarity approximates that of methanol (ε = 32.66) or acetonitrile (ε = 35.94),35 would be expected to have the opposite effect, rendering the mixture less polar and presumably, less suitable as a medium for extraction of hydrated complexes. This is, in fact, a possible explanation for the decline in DSr values observed at high (>50% v/v) concentrations of TBP (Fig. 1, panels A and B).
A change in extraction mechanism (to strontium-nitrato-crown ether complex extraction) induced by the addition of high concentrations of TBP obviously represents another possible explanation for this decline. As shown in Fig. 2, however, even for an organic phase that is 80% (v/v) TBP, an increase in aqueous acidity leads to a decrease in DSr, behavior which, as has been noted previously,5 is inconsistent with the extraction of a strontium-nitrato-crown ether complex, such as is observed for TBP alone. This observation, while perhaps unexpected, is consistent with the results of ion chromatographic measurements of nitrate ion extraction into various TBP–[C5mim][Tf2N] mixtures containing DCH18C6. That is, over the range 10–90% (v/v) TBP, no measurable co-extraction of nitrate ion occurs upon extraction of strontium. That nitrato complex extraction remains an insignificant route for strontium partitioning into [C5mim][Tf2N] regardless of TBP concentration is further demonstrated by the results of extended X-ray absorption fine structure (EXAFS) measurements on the organic phases obtained upon extraction of Sr(NO3)2 by DCH18C6-A in [C5mim][Tf2N] containing various amounts of TBP, summarized in Table 1. As reported previously,6 the nearest neighbor coordination environment of strontium in the Sr(DCH18C6)(H2O)22+ cation extracted into [C5mim][Tf2N] comprises 12 carbon atoms and 8 oxygen atoms, 6 from the crown ether and two from water molecules occupying axial coordination sites. The Sr K-edge EXAFS thus exhibits two major peaks, the first, at 2.7 Å, arising from coordinated oxygen atoms, the second, at 3.5 Å, from crown ether carbon atoms. Addition of TBP has no appreciable effect on the EXAFS observed for the extracted complex. In all cases, the number of coordinated oxygen atoms remains as 8, two of which occupy axial positions. In addition, in no instance is there evidence of the presence of distal (i.e., uncoordinated) oxygen atoms, which are associated with coordinated nitrate ions and which give rise to a strong, multiple-scattering peak at 4.3 Å. Two distal oxygens are, however, observed in the EXAFS results for extraction into TBP itself, as expected for partitioning of the strontium-nitrato crown ether complex depicted in Fig. 3A. Taken together, these results strongly indicate that a change in the mechanism of extraction does not underlie the decrease in strontium partitioning observed at high concentrations of TBP.
![]() | ||
Fig. 2 D Sr nitric acid dependencies for strontium extraction by DCH18C6-A into TBP, [C5mim][Tf2N], and an 80 ∶ 20 (% v/v) mixture of TBP and [C5mim][Tf2N]. |
![]() | ||
Fig. 3 Coordination environments of Sr(NO3)2(DCH18C6) in a conventional organic solvent (e.g., 1-octanol) (panel a) and the Sr(DCH18C6)(H2O)22+ cation in [C5mim][Tf2N] (panel b). |
Organic phase | No. DCH18C6 | No. coordinated O | No. axial O | No. distal O |
---|---|---|---|---|
[C5mim][Tf2N] (IL) | 1.1 ± 0.2 | 8.7 ± 0.8 | 2.0 | 0 |
40% IL + 60% TBP | 0.9 ± 0.2 | 7.3 ± 0.8 | 1.9 | 0 |
10% IL + 90% TBP | 1.0 ± 0.2 | 7.9 ± 0.8 | 2.0 | 0 |
TBP | 0.9 ± 0.2 | 8.6 ± 0.8 | 3.4 | 1.7 ± 0.8 |
Increased organic phase water content arising from the addition of TBP represents a second possible origin of the increase in strontium extraction observed for many of the TBP–IL mixtures. That is, addition of TBP has previously been demonstrated to increase the water content of various conventional organic solvents,36 and strontium extraction by crown ethers from acidic media has been shown to be favored by high organic phase water content.36,37 Both the solubility of water in alkanes and the extraction of strontium from nitric acid into alkanes by crown ethers, for example, increase in direct proportion to the concentration of TBP added.36 Recent work, however, has shown that increasing the water content of an ionic liquid does not necessarily lead to an increase in metal ion extraction efficiency. In fact, a decrease in DSr with increasing solvent water content has been observed for extraction by DCH18C6 into [C2mim][Tf2N] from acidic nitrate media under certain conditions.7 Moreover, Dai et al. have reported that the presence of dissolved water is not an important factor in determining the extent of strontium partitioning into RTILs containing crown ethers.38 These studies suggest that increased solubilization of water in the ionic liquid is unlikely to account for the increase in strontium extraction upon TBP addition.
The formation of a synergistic adduct between the strontium-crown ether complex and TBP represents another possible explanation for the observed enhancement of extraction. Takeda39–41 and Hasegawa et al.42 have shown that under appropriate conditions, any of a number of metal-CE complexes, among them the rubidium and caesium complexes of 12-crown-4 (12C4), 15-crown-5 (15C5), or benzo-15-crown-5 (B15C5),39,40 the copper(II) and zinc(II) complexes of 12C4 or 15C5,42 and the thallium complex of 15C5,41 can form 1 ∶ 1 adducts with TBP, and that these adducts can be more extractable than the metal-CE complexes themselves. To investigate the possibility of adduct formation in our system, a continuous variation study, in which the relative proportion of DCH18C6 and TBP in [C5mim][Tf2N] was varied while maintaining a constant total ([DCH18C6] + [TBP]) concentration, was carried out. Such studies are well established as a means of determining the stoichiometry of extracted metal complexes and establishing the existence of synergistic interactions between extractants.34 As can be seen from Fig. 4, the values of DSr observed at a given mole fraction of TBP or DCH18C6 are slightly higher than those calculated on the basis of the sum of the distribution ratios for the individual extractants, assuming first-power dependencies. In addition, a maximum (albeit a not especially well-defined one) is observed in the continuous variation plot at a 1 ∶ 1 TBP ∶ DCH18C6 mole ratio. These results point to the formation of a synergistic adduct between the Sr-DCH18C6 complex and TBP. Further evidence of adduct formation is provided by the results of measurements of the dependence of DSr on the concentration of TBP at fixed crown ether concentration in [C5mim][Tf2N]. If TBP addition is accompanied by 1 ∶ 1 adduct formation (eqn. 1), a log–log plot of DSrvs. [TBP] would be expected to yield a line of unit slope, and in fact, such a relationship (slope = 0.85 ± 0.06) is observed (Fig. 5).
Sr(DCH18C6)2+ + TBP ↔ Sr(DCH18C6)(TBP)2+ | (1) |
![]() | ||
Fig. 4 D Sr vs. mole fraction of DCH18C6 for extraction of strontium by DCH18C6–TBP mixtures in [C5mim][Tf2N]. ([HNO3] = 1 M; [DCH18C6] + [TBP] = 0.1 M). |
![]() | ||
Fig. 5 D Sr extractant dependency for the extraction of strontium by tri-n-butylphosphate (TBP) into [C5mim][Tf2N] containing 0.1 M DCH18C6. (Aqueous phase = 1 M HNO3). |
Thus, as we anticipated, extraction of strontium by DCH18C6 into ILs is susceptible to appreciable synergistic enhancement by addition of a second neutral extractant. This behavior stands in contrast to that of conventional solvents, for which synergistic effects are small or non-existent for systems employing either a pair of neutral extractants (e.g., CE + TBP39–42 or CE + tri-n-octylphosphine oxide43) or a polar organic solvent and for those not incorporating an organophilic anion (e.g., picrate), A−, to maintain electroneutrality:44,45
Sr2+ + DCH18C6org + TBPorg + 2A− ↔ Sr(DCH18C6)(TBP)A2 org | (2) |
To determine if the synergistic effects in these systems are sufficiently large to outweigh the decline in metal ion extraction efficiency that accompanies increased IL cation hydrophobicity,11 the extraction of strontium by DCH18C6–TBP mixtures into a series of [Cnmim][Tf2N] ILs of increasing alkyl chain length, n, was measured. As shown previously for Sr2+-DCH18C6 in the absence of TBP, an increase in the length of the alkyl chain in 1-alkyl-3-methylimidazolium bis[(trifluoromethyl)sulfonyl]imides from n-pentyl to n-hexyl, n-octyl, and finally, to n-decyl results in a gradual shift in the mechanism of ion transfer from aqueous nitrate media into ionic liquids containing DCH18C6 from cation exchange to nitrato complex partitioning.11 That is, a decreasing percentage of the ion transfer occurs via ion exchange as the hydrophobicity of the IL cation increases. As shown in Fig. 6, for Sr2+-DCH18C6–TBP systems, the magnitude of the synergistic enhancement observed decreases as the fraction of strontium partitioning occurring via ion exchange falls, until for [C10mim][Tf2N], addition of TBP yields little or no enhancement (Fig. 7). Thus, the synergistic effect is apparently dependent on the presence of the cationic (and water-bearing) Sr(DCH18C6)2+ complex. Clearly then, synergistic effects arising from combinations of neutral extractants are unlikely to be useful as a means of overcoming the decrease in metal ion extraction efficiency that accompanies increased IL cation hydrophobicity. These effects are, unfortunately, large only for systems in which the extraction efficiency is already high.
![]() | ||
Fig. 6 Variation in the magnitude of the synergistic enhancement of strontium extraction by TBP-DCH18C6 mixtures into [Cnmim][Tf2N] ILs with the fraction of partitioning occurring via nitrato complex extraction.11 (S ≡ synergistic factor = DSr(TBP+CE)/(DSr(TBP) + DSr(CE)). Organic phase: 0.1 M DCH18C6 in a 50 ∶ 50 v/v mixture of TBP and the indicated ionic liquid. Aqueous phase: 1 M HNO3). |
![]() | ||
Fig. 7 Effect of co-extractant concentration (expressed as % v/v) on the partitioning of Sr-85 between 1 M HNO3 and [C10mim][Tf2N] containing 0.1 M DCH18C6-A. |
Despite this, synergistic interactions between extractants in ILs may still have practical value, particularly if they are accompanied by a significant improvement in the extraction selectivity for the ion of interest over common interferents. With this in mind, the extractant dependency of the distribution of Sr2+, Ca2+ and Na+ between nitric acid and solutions of DCH18C6 in [C5mim][Tf2N] was determined both in the presence and absence of TBP (Fig. 8). In [C5mim][Tf2N], the extraction selectivity (Sr2+ > Na+ > Ca2+) is that expected on the basis of the known complexation behavior of DCH18C6.46 In a 1 ∶ 1 (v ∶ v) mixture of [C5mim][Tf2N] and TBP, however, appreciable enhancement of the extraction of both Sr2+ and Ca2+ is observed, while only a modest enhancement is observed for Na+, consistent with the expected weaker interaction of the latter with TBP. The result is a significant increase in the extraction selectivity for Sr2+ over Na+ (but not Ca2+) upon addition of TBP. Additional work is required to establish the practical utility of this observation.
![]() | ||
Fig. 8 D M extractant dependency for the extraction of strontium, calcium, and sodium ions by DCH18C6-A/B in [C5mim][Tf2N] in the absence (panel A) or presence (panel B) of TBP (50% v/v). (Aqueous phase = 0.1 M HNO3). |
Although our focus in these initial studies has been on combinations of neutral extractants, results obtained in conventional molecular solvents raise the possibility that much greater synergistic effects may be observed in ionic liquids for other extractant combinations (e.g., neutral–acidic). Such combinations may thus offer substantial opportunities for the design of improved IL-based metal ion separation systems. Work addressing these opportunities is now underway in this laboratory.
Footnotes |
† Work performed under the auspices of the Office of Basic Energy Sciences, Division of Chemical Sciences, US Department of Energy under contract number W-31-109-ENG-38. |
‡ Electronic supplementary information (ESI) available: k3-weighted Sr K-edge EXAFS of Sr-crown ether complexes in RTIL–TBP mixtures, scattering path lengths, energy threshold shifts and Debye–Waller factors. See http://www.rsc.org/suppdata/gc/b4/b414756a/ |
This journal is © The Royal Society of Chemistry 2005 |