Extraction of S- and N-compounds from gasoline and diesel oil by ionic liquids (ILs) indicates that such a process could be an alternative to common hydrodesulfurization (HDS) for deep desulfurization down to values of 10 ppm S or even lower. The results show the selective extraction properties of ILs, especially with regard to those S-compounds which are hard to remove by HDS, e.g. dibenzothiophene derivatives present in middle distillates like diesel oil. The application of mild process conditions (ambient pressure and temperature) and the fact that no hydrogen is needed, are additional advantages compared to HDS. Very promising ILs are [BMIM][OcSO₄] and [EMIM][EtSO₄], as they are halogen-free and available from relatively cheap starting materials. Extraction with ILs is not limited to diesel oil, but probably even more attractive for FCC-gasoline. Although HDS of S-species present in this gasoline constituent – mainly thiophenes – is relatively straightforward, a major drawback is the loss in octane number by olefin saturation, which favours extraction with ILs.

The thermal stability of an ionic liquid (IL) is an important parameter and limits the maximum operation temperature. However, the definition of stability and of the maximum operation temperature, respectively, is still an open question. Typically, non-isothermal thermogravimetrical analysis (TGA) is used to determine the stability, which is then mostly defined by the onset temperature, i.e. by the temperature where a certain mass loss of e.g. 1% is reached. Unfortunately, the rate of mass loss depends on the apparatus and conditions (e.g. heating rate), and may be governed by evaporation or by thermal decomposition or by a combination of both. In this work, isothermal as well as non-isothermal TG/DTG measurements at different heating rates were used as basis to model the combined kinetics of evaporation and decomposition, thereby taking 1-butyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide [BMIM][BTA] as an example. The measured and predicted mass losses are in good agreement, and the simulation of TG/DTG experiments by the methods outlined in this work leads to a reliable estimation of the evaporation (as shown by comparison with literature data) as well as of the decomposition rate. For a closed system, where the mass loss by evaporation is negligible, a novel criterion (1% mass loss by thermal decomposition within one year) is presented to estimate the maximum operation temperature of ILs.

Ionic liquids (ILs) are widely discussed as alternative green solvents not only because of their unique chemical properties, but also because of their extremely low vapour pressure and – at least in some cases – relatively high thermal stability. Two complementary methods are analyzed and compared to determine both the rate constant of decomposition and the vapour pressure of four ILs: (1) thermogravimetrical analysis at ambient pressure (TG_ap) with an overflow of inert gases, and (2) high vacuum (HV) experiments with a magnetic suspension balance (MSB). At ambient pressure, [EMIM][MeSO₃] and [EMIM][CF₃SO₃] decompose without a significant contribution of evaporation, which leads to the rate constant of thermal degradation. For both ILs, the vapour pressure can only be determined at HV by the MSB, because the evaporation rate is then higher than the decomposition rate. For the relatively volatile ILs [EMIM][NTf₂] and [BMIM][NTf₂] the vapour pressure can be derived both by the MSB at HV as well as by TG_ap. General strategies to determine the volatility and stability of ILs and criteria for the maximum operation temperature with regard to decomposition and evaporation are presented.