Ionic liquids (ILs) are a new class of alternative solvents that make ideal non-volatile media for a variety of industrial processes such as organic synthesis and biocatalysis, as alternative electrolytes, as phases and phase modifications in separation techniques, and as alternative lubricants. Once the large-scale implementation of ILs begins, the industrial application will follow. In view of their great stability, they could slip through classical treatment systems to become persistent components of the environment, where the long-term consequences of their presence are still unknown. Sorption on soils has a critical effect on the transport, reactivity, and bioavailability of organic compounds in the environment. So far, the IL sorption mechanism was investigated solely on the basis of batch experiments, which precluded any assessment of the dynamics of the process. An understanding of the mobility of ILs in soil columns is crucial for an accurate prediction of their fate in the soil. The aim of this study therefore was to investigate in detail the mobility of selected imidazolium ILs on three soil types. Moreover, it was decided to study these processes in soils from the coastal region (GdaA"sk, Poland), which usually constitute a very important geochemical compartment, participating in the transport of contaminants on their way to the sea. The mobility of alkylimidazolium ILs was investigated in columns containing soils from the coastal area. In addition, the sorption processes in all the soil systems studied were described isothermally and the equilibrium sorption coefficient was evaluated. The sorption capacities were determined according to OECD guidelines. Sorption dynamics was studied with use of polypropylene columns (diameter-10 mm, height-100 mm) packed with 10 g of soil. The ionic liquid solution was then injected into the soil column and left for 24 h to equilibrate. After this, a solution of 0.01 mM CaCl2 was pumped through the column at a rate of 0.3 ml min(-1). Effluents were collected from the bottom of the column and analyzed by HPLC. Sorption was strongest on the Miocene silt and the alluvial agricultural soil and weakest on the podsolic soil and Warthanian glacial till. The K (d) value of long-chain ILs was far higher than that of the short-chain ones. Among the substances tested, hydroxylated ILs were usually more weakly sorbed. Desorption of ILs is inversely correlated with sorption intensity. The experimental results of the column tests correlate well with those from batch experiments. In the cases of weakly binding soils, ILs were detected almost immediately in the eluent. The elution profiles of long-chain ILs indicate that these compounds are very strongly sorbed onto most soils, although certain amounts were transported through the soil. ILs exhibit a certain mobility in soils: in particular, salts with short and/or hydroxylated side chains are extremely mobile. The results indicate a stronger binding of ILs in the first sorption layer; once the first layer is saturated, there are no more active sites on the soil surface (no free charged groups); hence, there are no more strong electrostatic binding sites, and dispersive interaction becomes the dominant interaction potential. The influence of the structure of the ILs, especially the side-chain length was also confirmed: The K (d) value of long-chain ILs was far higher than that of the short-chain ones. The long alkyl side chain facilitates dispersive interactions with soil organic matter and intermolecular binding, and the build-up of a second layer becomes possible. Among the substances tested, hydroxylated ILs were usually more weakly sorbed. The hydroxyl group in the side chain can alter the polarity of the compound so strongly that interaction with organic matter hardly occurs; these salts then remain in the aqueous phase. The experimental results from the column tests correlate well with those from batch tests. In the weakly binding soils (with low organic matter), the only binding to the soil surface must be via electrostatic interactions, although intermolecular van der Waals (ionic liquid-ionic liquid) interactions could also be taking place. The elution profile maxima for organic rich soils are far smaller than for the other soils. In the former, hydrogen bonding, dispersive and pi aEuro broken vertical bar pi interactions play a more important part than electrostatic interactions. The rapidly "disappearing" maxima of the elution peaks may indicate that, after elution of ILs from the second layer, it is difficult to extract further sorbed ILs. In the first layer, the ILs are bound by much stronger electrostatic interactions. To break these bonds, a greater energy is required than that sufficient to extract ILs from double sorption layers. Results indicate, moreover, that hydrophobic ILs will be sorbed in the first few centimeters of the soil; migration into the soil will therefore be almost negligible. Sorption of ILs was the strongest in soils with the highest cation exchange capacities and a high organic content. ILs were also more strongly bound to the first sorption layer. The sorption coefficients of long-chain ILs were far higher than those of short-chain ones; usually, hydroxylated derivatives were the least strongly sorbed. Results of soil column experiments to investigate the mobility of ILs in soils correlated well with those from batch tests, and the elution profiles were also well correlated with organic matter content. The observed rapidly disappearing elution peak maxima probably indicate that, after elution of the ILs from the second layer, it is difficult to extract further sorbed compounds. Obtained results gave an interesting insight into mobility of ionic liquids in soil columns. However, several questions are now opened. It is therefore necessary to undertake further studies focused on total cycle of ionic liquids in the soil environment. This should include their evapotrasporation (lysimeter test), bioaccumulation by plants as well as degradation and transformation processes (chemical, biological, and physical) typically occurring in soils. Moreover, a further risk assessment of ILs is desirable since this study has indicated that these compounds, especially those with low lipophilicities, are generally mobile in the soil matrix. It is already known that short-chain ILs are characterized by low toxicities; should they enter the environment, they will probably migrate within the soil and pose a risk of contamination of surface and ground waters. This topic is relevant to the audience. Environmental threat of short-chain ionic liquids is currently unknown. From the predictive point of view, judging on known low acute toxic effects or high polarities of these compounds seems to be not enough to confirm their "environmental friendliness". If we are to fully understand the potential environmental effects, one should also have an insight into long-term biological consequences of these ionic liquids, including chronic toxicity tests, bioaccumulation, and biotransformation rates as well as stability against natural elimination mechanisms.
