Multivariate design of sustainable and efficient ionic liquids

Abstract

Ionic Liquids (ILs) attracted extraordinary attention due to their versatility in different applications. Their low vapour pressure reduces the air pollution as compared to common volatile organic solvents, but this property is not sufficient to define them as green solvents. In fact, release of ILs from industrial processes into wastewaters may lead to pollute aquatic environments due to their high solubility and stability. Sustainability and efficiency of ILs are both fundamental requirements for their industrial applications which boost the knowledge of ILs toxicological and environmental properties to comply with the European Union regulation (REACH). Unfortunately toxicity and sustainability are wide terms being determined on different possible targets. Furthermore, reliable toxicity tests when available in the literature, are reported for a specific class of compounds. In this context a multivariate insight allowed to compact available toxicity data into 104 toxicity scores [1] representative of the most common aquatic living targets. The number of ILs resulting from the combination of cations and anions is estimated to be over a million, hence the ILs experimental space is so huge that it cannot be fully explored and requires a rational selection of highly informative experiments. The development of Quantitative Structure-Property Relationships (QSPRs) models implies the knowledge of structural descriptors for ILs. Unfortunately, experimentally determined physico-chemical properties in the case of ILs are very few, derived by different laboratories and scattered in the literature. To overcome this difficulty, we derived in silico cation and anion physico-chemical descriptors [2] by using the VolSurf+ approach. Such descriptors were tested in two QSPR models on ILs aquatic toxicity [2] and polarity [3], achieving good correlations and satisfactory predictions. Handling such a high number of descriptors may be difficult, especially for big ILs data sets. Consequently the descriptors were also compacted into few Principal Properties (PPs) [4] highly informative and suitable for multivariate experimental design. PPs can be used as descriptors in QSPR correlations to model both ILs biological activities [4,5] and physico-chemical properties [6,7] achieving reliable predictions and allowing an intelligent selection of further measurements in order to expand the explored experimental space. The work presented here moves a step forward for conjugating efficiency and sustainability of ILs, presently a major issue of the scientific community.

References: [1] Paternò, A.; D Anna, F.; Musumarra, G.; Noto, R.; Scirè, S., RSC Adv., 2014, 4, 23985. [2] Paternò, A.; Bocci, G.; Goracci, L.; Musumarra, G.; Scirè, S., SAR QSAR Environ. Res., 2016, 27, 1. [3] Paternò, A; D Anna, F.; Fortuna, C. G.; Musumarra, G., Tetrahedron, 2016, 72, 3282. [4] Paternò, A.; Bocci, G.; Cruciani, G.; Fortuna, C. G.; Goracci, L.; Musumarra, G.; Scirè, S., SAR QSAR Environ. Res., 2016, 27, 221. [5] Paternò, A.; Scirè, S.; Musumarra, G., Toxicol. Res., 2016, 5, 1090. [6] Paternò, A.; Fiorenza, R.; Marullo, S.; Musumarra, G.; Scirè, S., RSC Adv., 2016, 6, 36085. [7] Paternò, A.; Goracci, L.; Scirè, S.; Musumarra, G., ChemistryOpen, 2017, doi: 10.1002/open.201600119.