Evolution of Surface Tension and Hansen Parameters of a Homologous Series of Imidazolium-Based Ionic Liquids.

In this study, we systematically analyze the surface tension and Hansen solubility parameters (HSPs) of imidazolium-based ionic liquids (ILs) with different anions ([NTf2]-, [PF6]-, [I]-, and [Br]-). These anions are combined with the classical 1-alkyl-3-methyl-substituted imidazolium cations ([CnC1Im]+) and a group of oligoether-functionalized imidazolium cations ([(mPEGn)2Im]+) based on methylated polyethylene glycol (mPEGn). In detail, the influences of the length of the alkyl- and the mPEGn-chain, the anion size, and the water content are investigated experimentally. For [CnC1Im]+-based ILs, the surface tension decreases with increasing alkyl chain length in all cases, but the magnitude of this decrease depends on the size of the anion ([NTf2]- < [PF6]- < [Br]- ≤ [I]-). Molecular dynamics (MD) simulations on [CnC1Im]+-based ILs indicate that these differences are caused by the interplay of charged and uncharged domains, in particular in the different anions, which affects the ability of the alkyl chains of the cation to orient toward the liquid-gas interface. An increase in the mPEGn-chain length of the [(mPEGn)2Im][A] ILs does not significantly influence the surface tension. These changes upon variation of the cation/anion combination do not correlate with the evolution of the HSPs for the two sets of ILs. Finally, our data suggest that significant water contents up to water mole fractions of x(H2O) = 0.25 do not significantly affect the surface tension of the studied binary IL-water mixtures.

Reservoir computing decoupling memory-nonlinearity trade-off.

Reservoir computing (RC), a variant recurrent neural network, has very compact architecture and ability to efficiently reconstruct nonlinear dynamics by combining both memory capacity and nonlinear transformations. However, in the standard RC framework, there is a trade-off between memory capacity and nonlinear mapping, which limits its ability to handle complex tasks with long-term dependencies. To overcome this limitation, this paper proposes a new RC framework called neural delayed reservoir computing (ND-RC) with a chain structure reservoir that can decouple the memory capacity and nonlinearity, allowing for independent tuning of them, respectively. The proposed ND-RC model offers a promising solution to the memory-nonlinearity trade-off problem in RC and provides a more flexible and effective approach for modeling complex nonlinear systems with long-term dependencies. The proposed ND-RC framework is validated with typical benchmark nonlinear systems and is particularly successful in reconstructing and predicting the Mackey-Glass system with high time delays. The memory-nonlinearity decoupling ability is further confirmed by several standard tests.