Protic ionic liquids (PILs) are cost efficient “designer” solvents which can be tailored to have properties suitable for a broad range of applications. The vast number of possible cation and anion combinations of PILs, along with additions of co-solvents and solutes, makes it difficult and time-consuming to optimise their properties. Here we have used high throughput methodologies which enable rapid synthesis, characterisation and analysis to explore an extensive range of PIL containing samples. A selection of eight alkylammonium nitrate, formate and acetate PILs were used. These had alkyl chain lengths of 2 to 5 carbons, with some containing a hydroxyl group. Non-stoichiometric PILs were prepared through additions of the corresponding precursor acid or base. Aqueous solutions of the non-stoichiometric and stoichiometric PILs were also included from neat water to neat PIL. A latin hypercube sampling method was used to select 25 compositions for each PIL, covering the composition space. This led to a total of 200 samples for the 8 non-stoichiometric and stoichiometric aqueous PIL solutions. The liquid nanostructure, surface tension and density was obtained for each sample, along with the pH where there was considerable water present. Machine learning approaches were used to determine the effect of different cation and anion structural features, non-stoichiometry and water for these systems. As an example, Figure 1 shows the coefficients for structural and solvent changes on the surface tension. This work enabled structure-property relationships to be developed for a multivariate sample space for PIL containing solutions. This methodology is useable for developing solvent systems containing multiple species, such as one or more ILs with co-solvents, and for developing deep eutectics.

High throughput and machine learning approaches to characterising stoichiometric and non-stoichiometric protic ionic liquid-water solutions

255th National Meeting (2018)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.