We have developed a simple and quantitative explanation for the relatively low melting temperatures of ionic liquids (ILs). The basic concept was to assess the Gibbs free energy of fusion (Delta(fus)G) for the process IL(s) --> IL(l), which relates to the melting point of the IL. This was done using a suitable Born-Fajans-Haber cycle that was closed by the lattice (i.e., IL(s) --> IL(g)) Gibbs energy and the solvation (i.e., IL(g) --> IL(l)) Gibbs energies of the constituent ions in the molten salt. As part of this project we synthesized and determined accurate melting points (by DSC) and dielectric constants (by dielectric spectroscopy) for 14 ionic liquids based on four common anions and nine common cations. Lattice free energies (Delta(latt)G) were estimated using a combination of Volume Based Thermodynamics (VBT) and quantum chemical calculations. Free energies of solvation (Delta(solv)G) of each ion in the bulk molten salt were calculated using the COSMO solvation model and the experimental dielectric constants. Under standard ambient conditions (298.15 K and 10(5) Pa) Delta(fus)G degrees was found to be negative for all the ILs studied, as expected for liquid samples. Thus, these ILs are liquid under standard ambient conditions because the liquid state is thermodynamically favorable, due to the large size and conformational flexibility of the ions involved, which leads to small lattice enthalpies and large entropy changes that favor melting. This model can be used to predict the melting temperatures and dielectric constants of ILs with good accuracy. A comparison of the predicted vs experimental melting points for nine of the ILs (excluding those where no melting transition was observed and two outliers that were not well described by the model) gave a standard error of the estimate (s(est)) of 8 degrees C. A similar comparison for dielectric constant predictions gave s(est) as 2.5 units. Thus, from very little experimental and computational data it is possible to predict fundamental properties such as melting points and dielectric constants of ionic liquids.

We have used microwave dielectric relaxation spectroscopy to study the picosecond dynamics of five low-viscosity, highly conductive room temperature ionic liquids based on 1-alkyl-3-methylimidazolium cations paired with the bis((trifluoromethyl)sulfonyl)imide anion. Up to 20 GHz the dielectric response is bimodal. The longest relaxation component at the time scale of several 100 ps reveals strongly nonexponential dynamics and correlates with the viscosity in a manner consistent with hydrodynamic predictions for the diffusive reorientation of dipolar ions. Methyl substitution at the C2 position destroys this correlation. The time constants of the weak second process at the 20 ps time scale are practically the same for each salt. This intermediate process seems to correlate with similar modes in optical Kerr effect spectra, but its physical origin is unclear. The missing high-frequency portion of the spectra indicates relaxation beyond the upper cutoff frequency of 20 GHz, presumably due to subpicosecond translational and librational displacements of ions in the cage of their counterions. There is no evidence for orientational relaxation of long-lived ion pairs.

No abstract

Ionic liquids (ILs) have captured the imagination of a large and steadily growing community of scientists due to their applications as reaction media, [1,2] in batteries and supercapacitors, [3] in solar and fuel cells, [4] for electrochemical deposition of metals and semiconductors, [5] for protein extraction and crystallization, [6] in nanoscience, [7] in physical chemistry, [8] and many others. By choosing different combinations of ions, or by modifying the chemical structures of the constituent ions, the physical properties of an IL can be significantly altered. However, the number of possible modifications is huge, and one can envisage an enormous number of salts that have the potential to form ILs-some say as many as 10 12 to 10 18 . [9, 10] Since the vast majority of these have yet to be synthesized, it is imperative to develop methods to predict the physical properties of unknown ILs in order to facilitate the design of new materials and reduce the need for time-consuming trial-and-error syntheses. The "Holy Grail" is the full characterization of an unknown IL prior to its laboratory synthesis.Previous attempts to make quantitative predictions of the physical properties of ILs by using quantitative structureproperty relationships (QSAR), molecular mechanics (MM) simulations, as well as modifying older ideas, such as the concept of "hole theory" or the "Parachor", have had some success. [10][11][12] However, these methods all have significant drawbacks which limit their application for predicting the properties of unknown salts. These include the need for large experimental datasets to derive correlations, time-consuming computational methods, or the need for at least some experimental data from the IL under study.We recently showed that the relatively low melting points of ILs can be understood by a simple thermodynamic cycle based on lattice and solvation energies. [13,14] This model also allowed the prediction of the melting points (and dielectric constants) of ILs with good accuracy. Subsequently, we noticed a strong relationship between the molecular volumes V m of ILs and their fundamental physical properties: viscosity, conductivity, and density. Herein we describe these simple relationships and show that it is possible for nonspecialists to predict the physical properties of even unknown ILs with very good accuracy from only their molecular volumes and an anion-dependent correlation.The molecular volume V m (or formula-unit volume) of a salt is a physical observable and is defined as the sum of the ionic volumes V ion of the constituent ions. For example, for a binary IL V m is given by Equation (1).The ionic volume is a measure of the size of an ion, similar to the traditional ionic radius.[15] However, ionic radii are poorly defined and arguably not physically meaningful for nonsymmetrical ions such as those found in many ILs. In contrast, ionic volumes are well defined and equally valid for symmetrical and nonsymmetrical ions. The ionic volume can be derived from crystal structures (e.g., the...

In continuation of recent work on the dielectric response of imidazolium-based ionic liquids (ILs) (J. Phys. Chem. B, 2006, 110, 12682), we report on the effect of cation variation on the frequency-dependent dielectric permittivity up to 20 GHz of ionic liquids. The salts are comprised of pyrrolidinium, pyridinium, tetraalkylammonium, and triethylsulfonium cations combined with the bis-((trifluoromethyl)sulfonyl)imide anion. The dielectric spectra resemble those observed for imidazolium salts with the same anion. In all cases, the major contribution results from a diffusive low-frequency response on the time scale of several 100 ps, which shows a broadly distributed kinetics similar to that of spatially heterogeneous states in supercooled and glassy systems rather than that observed in fluid systems. There is evidence for a weak secondary process near 10-20 ps. Perhaps the most interesting difference to imidazolium salts is founded in the missing portions of the spectra due to processes beyond the upper cutoff frequency of 20 GHz. These are lower than that observed for imidazolium-based salts and seem to vanish for tetraalkylammonium and triethylsulfonium salts. As for imidazolium salts, the extrapolated static dielectric constants are on the order of epsilon(S) congruent with 10-13, classifying these ILs as solvents of moderate polarity.

A review of the solvent effects that control the productivity of cross-coupling reactions, and suggested safer alternative solvents.

The liquid-vacuum interfaces of a series of room-temperature ionic liquids (RTILs) containing the 1-alkyl-3-methylimidazolium cation ([C n mim]) were investigated by reactive-atom scattering (RAS). The length of the alkyl chain (n = 4, 6, 8, and 12) and the anion type (bis(trifluoromethylsulfonyl)imide ([Tf 2 N]), trifluoromethanesulfonate ([OTf]), and tetrafluoroborate ([BF 4 ])) were varied systematically to determine their effects on the preferential occupancy of the surface by alkyl chains. The experiments employed collisions with gas-phase, ground-state oxygen atoms, O( 3 P), generating OH and H 2 O products that revealed the abundance of abstractable H atoms at the liquid surface. Two complementary approaches with different Oatom energies and detection methods were employed: we denote these RAS-laser induced fluorescence (RAS-LIF) and RAS-mass spectrometry (RAS-MS). [C n mim][BF 4 ] RTILs were studied by both methods, giving consistent trends of strongly increasing alkyl coverage with chain length. Even for the longest alkyl chain, n = 12, the surface is not saturated with alkyl chains, with some fraction still occupied by other groups. RAS-LIF results for RTILs with the three different anions, over the range of alkyl chain lengths, showed that their surfaces can be distinguished clearly. Alkyl surface coverage depends sensitively on the anionic volume, indicating that the packing of ions at the surface is driven largely by steric effects. Molecular dynamics simulations of the liquid surfaces support all the experimental findings, including the rationalization of expected quantitative differences between the RAS-LIF and RAS-MS results.

Ionic-liquid (IL) mixtures hold great promise, as they allow liquids with a wide range of properties to be formed by mixing two common components rather than by synthesizing a large array of pure ILs with different chemical structures. In addition, these mixtures can exhibit a range of properties and structural organization that depend on their composition, which opens up new possibilities for the composition-dependent control of IL properties for particular applications. However, the fundamental properties, structure, and dynamics of IL mixtures are currently poorly understood, which limits their more widespread application. This article presents the first comprehensive investigation into the bulk and surface properties of IL mixtures formed from two commonly encountered ILs: 1-ethyl-3-methylimidazolium and 1-dodecyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide ([Cmim][TfN] and [Cmim][TfN]). Physical property measurements (viscosity, conductivity, and density) reveal that these IL mixtures are not well described by simple mixing laws, implying that their structure and dynamics are strongly composition dependent. Small-angle X-ray and neutron scattering measurements, alongside molecular dynamics (MD) simulations, show that at low mole fractions of [Cmim][TfN], the bulk of the IL is composed of small aggregates of [Cmim] ions in a [Cmim][TfN] matrix, which is driven by nanosegregation of the long alkyl chains and the polar parts of the IL. As the proportion of [Cmim][TfN] in the mixtures increases, the size and number of aggregates increases until the C12 alkyl chains percolate through the system and a bicontinuous network of polar and nonpolar domains is formed. Reactive atom scattering-laser-induced fluorescence experiments, also supported by MD simulations, have been used to probe the surface structure of these mixtures. It is found that the vacuum-IL interface is enriched significantly in C12 alkyl chains, even in mixtures low in the long-chain component. These data show, in contrast to previous suggestions, that the [Cmim] ion is surface active in this binary IL mixture. However, the surface does not become saturated in C12 chains as its proportion in the mixtures increases and remains unsaturated in pure [Cmim][TfN].