An unusual common ion effect promotes dissolution of metal salts in room-temperature ionic liquids: a strategy to obtain ionic liquids having organic–inorganic mixed cations

Abstract

A simple strategy has been reported to prepare new ionic liquids with binary systems of organic–inorganic cations exploiting the common ion effect, i.e. dissolving metal salts with organic or inorganic anions (bistriflylimide or nitrate) in ionic liquids bearing the same anions. The resulting concentrated solutions of metal cations in ionic environments, which may have great potential for electrochemical processes, have been characterized by X-Ray photoelectron spectroscopy (XPS) and electrospray ionization mass spectrometry (ESI-MS)

---

Introduction

Transition metals are among the most important elements for chemical and technological industry, with an extremely widespread utility range. Among the possible ways to process these elements, electrodeposition from concentrated solutions of metal salts is used for metal plating and finishing of a wide range of products, improving their environmental resistance and lifetime.

Ionic liquids have received much attention in recent years as promising conductive media for electrodeposition, due to their negligible volatility and large thermal and electrochemical stability.¹ Despite this, the solubility of transition metal salts in ionic liquids is quite limited: Afonso et al.² evaluated the solubility of LiCl, HgCl₂ and LaCl₃ in [C_nmim][BF₄] and [C_nmim][PF₆], obtaining solubilities scarcely exceeding 10⁻⁴ wt%. Very limited solubilities were also reported for titanium chlorides in [bmim][Tf₂N]³ and for various other metals.⁴

Concentrated solutions are required for the electrodeposition process,⁵ but these are not easy to obtain. Concentrations of about 50% mol were prepared in some cases;⁵ however, these solutions were been obtained by mixing ILs and metal salts in molecular solvents and then removing the solvent (or water), probably obtaining supersaturated solutions or solutions with an appreciable amount of water. Recently, it has been reported that concentrations up to 2.5 mol L⁻¹ have been achieved by dissolving AlCl₃ in [EMIM][TFSA], where a biphasic melt is obtained, going further on adding the aluminium complex.⁶ Similarly, a good solubility is reported for transition metal chlorides in dicyanamide-based ILs;⁷ nevertheless, for some metals, solubility does not exceed 0.01 M. In the above examples, high solubilities are generally obtained only by means of an ionic liquid based on a strongly complexing anion, probably leading to highly coordinated metal–anion complexes which affect the electrochemical properties of the metal and thus limit the electrodeposition process, as previously noted.⁷

In this work, we report on the possibility of obtaining high concentrations of metal salts in hydrophilic and hydrophobic ILs by simply dissolving the salts in ILs which contain the same anion. Ionic liquids with different cations (based on imidazolium, pyridinium and pyrrolidinium salts) and anions (triflate, bistriflylimide, nitrate, acetate) were tested, in order to ensure that this result can be obtained independently of the chemical nature of the ionic liquid.

Experimental

The metal salts Ag(Tf₂N), Ni(Tf₂N)₂, Al(Tf₂N)₃, Co(Tf₂N)₂, Y(Tf₂N)₃ (Solvionic, 99% pure) and Ag(NO₃), Ni(NO₃)₂, Al(NO₃)₃, Cr(NO₃)₃, AgBF₄, AgPF₆, AgTfO, Cu(TfO)₂ (Aldrich, 99% pure) were always dried (70 °C under vacuum, 2 × 10⁻³ Torr, for 10 h) before use. The ionic liquids 1-butyl-3-methylimidazolium bistriflylimide [bmim][Tf₂N], 1-ethyl-3-methylimidazolium bistriflylimide and 1-butyl-3-methylimidazoium triflate [bmim][TfO] were synthesized as previously reported.⁸ 1-Butyl-3-methylimidazolium nitrate was prepared from 1-butyl-3-methylimidazolium bromide by reaction with AgNO₃ in water. After filtration of the precipitate (AgBr), water was removed by distillation at reduced pressure. The colorless liquid was dissolved in dried acetone and stored at −40 °C for a week. Residual traces of AgBr were removed by filtration of the solution on celite.

The water content of the ionic mixtures was determined by the Karl Fisher technique using an apparatus composed of a standard titrator and a coloumeter.

ESI-MS analyses were performed on a Finningan LCQ Advantage (Thermo Finningan, San Jose, CA, USA) ion trap instrument equipped with Excalibur Software.

XPS measurements

The X-ray photoelectron spectroscopy (XPS) measurements reported in this paper were performed in a system using a non-monochromated Al Kα source (1486.6 eV) and a VSW HAC 5000 hemispherical electron energy analyser.

The sample holder consisted of planar gold foil (dimension 2 × 2 cm), ground with SiC paper down to 1200 grit, and than carefully degreased using HNO₃ (69.5% for analysis, Carlo Erba) and acetone (ultrapure for spectroscopy, Merck). It was introduced into the UHV system via a loadlock under inert gas (N₂) flux, in order to minimize the exposure to air contaminants. The sample thin films were prepared by deposition of one or two drops of IL solution, under nitrogen flux, directly onto the Au foil; the surface tension is enough to cover the surface with a film of liquid, as demonstrated by the absence of peaks attributable to gold. The overall time required for the loading procedure was less than one minute. Before the measurements, the sample was kept in the introduction chamber for at least 12 h, allowing the removal of volatile substances such as water and traces of organic solvents, as confirmed by the pressure value achieved (2 × 10⁻⁹ Torr), just above the instrument base pressure.

Photoelectron spectra were acquired in the constant-pass-energy mode at E_pas = 44 eV, and the overall energy resolution was 1.2 eV, measured as a full-width at half maximum (FWHM) of the Ag 3d_5/2 line of a pure silver reference. To vary the measurements probing depth, the spectra were collected at θ = 0° (normal emission), θ = 45° and θ = 60° (grazing angle). The probing depth varies mainly with cos θ, and since the energy of the collected photoelectrons ranges between 800 and 1300 eV and their inelastic mean free path (imfp) in such compounds is about 2 nm, we can consider that measurements probe 6–7, 4–5 and 2–3 nm at 0, 45 and 60°, respectively. Since the electron distribution within the imidazolium ring is little affected by the counter ion, we correct the energy scale using the N 1s peak (401.6 eV) of the imidazolium ring. Peak positions in different samples were reproduced within 0.2 eV. The recorded spectra were fitted using XPSPeak 4.1 software employing Gauss–Lorentz curves to fit the data after subtraction of a Shirley-type background.

Solubility measurements

Solutions were prepared by adding aliquots of the previously dried metal salt into a calibrated flask containing the weighed ionic liquid (5 mL), previously dried at 70 °C under vacuum (2 × 10⁻³ Torr) for 10 h. Additions were made at ambient temperature. For slowly solubilizing salts (Cr and Al salts) the flask has been gently heated until complete dissolution was obtained, and then allowed to cool before successive addition. Comparable results have also been obtained by performing the experiments at room temperature using longer dissolution times.

Results and discussion

The results for the solubilities are reported in Tables 1, 2, 3 and 4, where the molar concentration of metal salts solutions that we were able to obtain are reported.

Table 1 Solubilities (M) of nitrate salts in [bmim][NO₃]

Ag(NO3)	Ni(NO3)2	Al(NO3)3	Cr(NO3)3
0.670	0.520	0.523	0.609

---

Table 2 Solubilities (M) of bistriflylimide salts in [bmim][Tf₂N]

Ag(Tf2N)	Ni(Tf2N)2	Al(Tf2N)3
0.416	1.143	0.404

---

Table 3 Solubilities (M) of bistriflylimide salts in [emim][Tf₂N]

Al(Tf2N)3	Co(Tf2N)2	Y(Tf2N)3
0.66	1.28	2.12

---

Table 4 Solubilities (M) of various salts in different ionic liquids. The asterisk (*) indicates that the experiment stopped before the maximum solubility was attained

IL	Salt	Solubility/M
HePyTf2N	Y[Tf2N]3	0.847
HePyTf2N	Al[Tf2N]3	0.638
[bmim][BF4]	Ag[BF4]	0.036
[bmim][PF6]	Ag[PF6]	0.060
[bmim][TfO]	Cu[TfO]2	1.108
[bmim][TfO]	Ag[TfO]	0.846
[BMPyrr][Tf2N]	Co[Tf2N]	1.000(*)
[bmim][Ac]	Ag[Ac]	0.5(*)

---

Initially, the solubility of transition metal salts with the nitrate ([NO₃]⁻) and bistriflylimide ([Tf₂N]⁻) anions was measured in three ionic liquids having the same anion; specifically, in the hydrophilic [bmim][NO₃] and in the hydrophobic [bmim][Tf₂N] and [emim][Tf₂N].

Data reported in Tables 1–3 show that all the investigated ILs were able to dissolve the metal salts to a high extent, with solubilities greater than 0.4 M. We remark that the concentrations reported do not correspond to the solubility limit concentrations; in fact, many experiments were stopped to prevent the large volume increase of the solution exceeding the vial capacity. In other cases, we found a concentration of about 2 M satisfactory and found no purpose in going further.

As can be seen in Fig. 1, the colors of the nitrate solutions vary—bright green (Ni), yellow–pale green (Al), dark green (Cr), no colour (Ag, Y)—and all the obtained solutions were clear. The solution of Ag(Tf₂N) in [bmim][Tf₂N] is initially light pink, showing a deposition of metallic residues with time. Conversely, solutions of Ni and Al complexes are clear and transparent, with a bright green color (Ni) or no colour (Al). Washing the Ni(Tf₂N)₂ ionic liquid solution with water, a sudden decoloration of the ionic liquid phase is observed, with a subsequent movement of the green color to the aqueous phase, indicating that the salt is more soluble in water than in the IL.

Fig. 1 From left to right: pure [HePy][Tf₂N] and its solution with [Al][Tf₂N], [Ni][Tf₂N]₂ and Y[Tf₂N]₃.

All the solutions reported in Tables 1–3 were stable; no crystallization was observed, even after prolonged storage at room temperature (three months).

At the same time, we performed dissolution tests on Al(NO₃)₃ in [bmim][Tf₂N] and of Al(Tf₂N)₃ in [bmim][NO₃]; both these tests showed no detectable solubilization of the solid metal complex added, even after heating at the same conditions for times much longer (1 h) than reported for the solutions above.

Water content was determined for some solutions in order to understand if the addition of hygroscopic salts could have affected the dissolution process. We found a moderate water content, ranging from 1.23% for [bmim][Tf₂N] + Al[Tf₂N] to 0.23% for [bmim][NO₃] + Al[NO₃]. We can conclude from this that water intake is not a key point in the dissolution process.

To verify the general character of the dissolution phenomena, further analogous experiments with ILs bearing different cations and/or anions were performed. The solubilities obtained, when the anion of the metal salt is shared with the ionic liquid, are still very high, with the only exceptions being the salts based on BF₄⁻ and PF₆⁻ ions.

Even if high solubilities were obtained for mostly all homoanionic solutions, the metals do not share the same solvation structure in all these solutions. This was evident from preliminary X-ray photoelectron spectroscopy analyses, with which we studied the near-surface chemical composition. By performing angle-resolved XPS measurements at three different angles (0, 45 and 60°) on solutions of Al³⁺, Ni²⁺, Cr³⁺ and Ag⁺ we found strong differences between the metal ion distribution at the surface.

For Al³⁺ and Ag⁺ solutions (see Fig. 2), the metal peak is clearly detectable at all angles, providing evidence of the presence of metal ions up to a depth of no more than 2 nm from the physical surface. However, for the Cr³⁺ solution (see Fig. 3), the metal peak fades away with increasing take-off angle, showing that metal ions are not present at the physical surface of the liquid. This is also the case for the Ni²⁺ solution. We interpreted this behavior as probably being due to the different coordination spheres of the ions: while Ag⁺ and Al³⁺ stand in these solutions as single ions without a strong coordination sphere, independent of the nature of the anion, Cr³⁺ and Ni²⁺ have a strong coordination sphere which does not allow the metal ion to reach the physical surface of the liquid. This creates a strong barrier that shields the photoelectrons generated by the Cr³⁺ and Ni²⁺ ions, preventing them from leaving the surface and being detected. The coordination sphere affects not only the surface properties of these ionic systems, but also the electrochemical properties of the metal cations.

Fig. 2 XPS spectra at varying take off-angle for Al(2p) and Ag(3d) in [bmim][NO₃] (top) and [bmim][Tf₂N] (bottom).

Fig. 3 XPS spectra at varying take-off angles for the Cr(2p) core peak in [bmim][NO₃].

The electrospray ionization mass spectrometry (ESI-MS) analysis of the homoanionic solutions of Ag(Tf₂N), Al(Tf₂N)₃ and Ni(Tf₂N)₂ in [bmim][Tf₂N], registered under identical conditions (injection solvent, concentration, ionization energy, temperature, etc.) confirmed the different ability of the ionic mixtures to give clusters involving the inorganic cations. While the peak of the charged complex Ni[Tf₂N]₃⁻ (with mass 897.4 amu) was clearly detectable in the negative ion ESI mass spectrum of the Ni solution, significantly smaller peaks were detected for Ag[Tf₂N]₂⁻ (668.15 amu) and Al[Tf₂N]₄⁻ (1147.5 amu) in the corresponding spectra.

All these data led us to infer that the dissolution of a metal salt in an IL having the same anion gives a sort of liquid organic–inorganic mixed salt with structural features not dissimilar to the bulk ionic liquid. This is not an absolute novelty; actually, inorganic (alkali) mixed salts with lower melting points and wider electrochemical windows than the pure components have recently been described in the literature.^8,9 Nevertheless, the melting points of these inorganic salts are still well above the usual room temperature (about 340 K for LiFSI/KFSI). By solving the inorganic salt in an organic ionic liquid which is already liquid at room temperature, we are still able to obtain a room-temperature liquid phase with a high metal content (about 0.3 in molar fraction for Ni(Tf₂N)).

If we consider the IL as a solvent, a noticeable aspect of the phenomenon is that it appears to go apparently against the so-called “common ion” effect: the fact that, in aqueous solvents, the solubility of insoluble or sparingly soluble substances will be decreased by the presence of a common ion, leading to the conclusion that a salt will be less soluble if one of its constituent ions is already present in the solution.

Actually, in our experiments, the presence of a common ion leads to a significantly higher solubility of metal cations in ILs, while on the basis of the principle mentioned above, such common-ion salts would have been expected to be scarcely (or not at all) soluble in corresponding-ion liquids, particularly if we consider that these metal salts are sparingly soluble in other ILs. This once again indicates that care must be taken when applying concepts typical of molecular liquids to explain phenomena in the field of ionic liquids.

Conclusions

In conclusion, we have demonstrated that high concentrations of metal salts in both hydrophilic and hydrophobic ionic liquids can be obtained by directly dissolving the salts in ILs which have the same anion. From these results, we may hypothesize that this rather unexpected behavior may be reproduced by many other ionic liquids regardless of their particular chemical nature. This strategy provides the opportunity to prepare new ionic systems with properties, limitations and application possibilities that are currently under investigation.

Acknowledgements

This work has been supported by EC, Project FP6(STREP), Contract No. 517002 (IOLISURF).

References

1. T. Welton, P. Wasserscheid, Ionic Liquids in Synthesis. Wiley-VCH, Weinheim, 2007.
2. L. C. Branco, J. N. Rosa, J. J. M. Ramos and C. A. M. Afonso, Chem.–Eur. J., 2002, 8, 3671–3677.
3. F. Endres, S. Z. El Abedin, A. Y. Saad, E. M. Moustafa, N. Borissenko, W. E. Price, G. G. Wallace, D. R. MacFarlane, P. J. Newman and A. Bund, Phys. Chem. Chem. Phys., 2008, 10(16), 2189–2199.
4. A. E. Visser, R. P. Swatloski, W. M. Reichert, S. T. Griffin and R. D. Rogers, Ind. Eng. Chem. Res., 2000, 39, 3596–3604.
5. F. Endres, A. P. Abbot, D. R. MacFarlane, Electrodeposition from Ionic Liquids. Wiley-VCH, Weinheim, 2008.
6. S. Zein el Abedin, E. Moustafa, R. Hempelmann, H. Natter and F. Endres, ChemPhysChem, 2006, 7, 1535.
7. M.-J. Deng, P.-Y. Chen, T.-I. Leong, I.-W. Sun, J. Chang and W.-T. Tsai, Electrochem. Commun., 10(2), 213–216.
8. R. Bini, C. Chiappe, V. Llopsis Mestre, C. S. Pomelli and T. Welton, Org. Biomol. Chem., 2008, 6, 2522.
9. K. Kubota, T. Nohira, T. Goto and R. Hagiwara, J. Chem. Eng. Data, 2008, 53, 2144–2147; K. Kubota, T. Nohira, T. Goto and R. Hagiwara, Electrochem. Commun., 2008, 10, 1886–1888.

---