Abstract:It has been increasingly common to investigate dynamic and thermodynamic properties of green solvents at atomistic scales with molecular simulation. These designed solvents such as ionic liquids are often highly charged species, which pose a problem for molecular modelling with classical fixed-charge force fields. Simulation outcomes with atomic charges derived from ab initio calculations often show significant deviations from the experimental reference, and charge scaling is widely applied as a simple yet efficient solution to achieve satisfactory experiment-simulation accordance. Although the massive emphasis on the reproduction of bulk properties such as density of the liquids as charge scaling criteria, a more crucial thermodynamic observable is the solvation behavior of external agents in these green solvents. Astonishingly, our recent large-scale benchmark calculation on solvation thermodynamics suggests that the solvation-free-energy-derived scaling factor is generally slightly larger (~0.1) than the bulk-property-derived estimate. This phenomenon is rather not unexpected, as the density-matching estimate only considers the solvent-solvent interaction (overfitting), while accurate calculations of solvation free energies require balanced descriptions of solute-solvent and solvent-solvent interactions. A more interesting observation is that different solute-solvent pairs exhibit different responses to the variation of the scaling parameter, which arises from the competing electrostatic and vdW contributions. Another perspective provided in our previous extensive benchmark is about the suitability of general-purpose force fields for bonded and vdW interactions in the modelling of ionic liquids. The bond stretching and angle bending terms in pre-fitted GAFF derivatives are often problematic, while the torsional potential shows satisfactory reproduction of ab initio results. In the current work, we expect to accumulate more experiences from large-scale fast-growth solvation free energy simulations in ionic liquids with compositions different from our previous benchmark. The obtained new results are combined with our previous dataset to form a large solvation set (and also partition or water-ionic-liquids transfer set), from which a universally applicable charge scaling factor with at least half-optimal performance is derived. Aside from the force-field issue, another extremely important modelling detail considered in the current work is the finite-size effect. It is observed that the finite-size artifacts in solvation thermodynamics are much more severe than mass density, which emphasizes the use of a sufficiently large ionic solvent box in molecular simulations of ionic liquids derivatives. Finally, combining the extensive computational perspectives accumulated in our series works, general guidelines for molecular modelling of ionic liquids with fixed-charge force fields are summarized.

Building Water Models Compatible with Charge Scaling Molecular Dynamics

Polarizable Water Model with Ab Initio Neural Network Dynamic Charges and Spontaneous Charge Transfer

Building Water Models, A Different Approach

A Transferrable Range-Separated Force Field for Water: Combining the Power of Both Physically-Motivated Models and Machine Learning Techniques

Electrokinetic properties of NaCl solution via molecular dynamics simulations with scaled-charge electrolytes

Electronic coarse graining enhances the predictive power of molecular simulation allowing challenges in water physics to be addressed

Electrostatic interactions in water: nonlocal electrostatic approach

Charge equilibration model with shielded long-range Coulomb for reactive molecular dynamics simulations

Polarizable Mean-Field Model of Water for Biological Simulations with Amber and Charmm force fields

Molecular Modelling of Ionic Liquids: General Guidelines on Fixed-Charge Force Fields for Balanced Descriptions

Molecular Theories and Simulation of Ions and Polar Molecules in Water

Scalable molecular simulation of electrolyte solutions with quantum chemical accuracy

Non-Aqueous Ion Pairing Exemplifies the Case for Including Electronic Polarization in Molecular Dynamics Simulations

A force field of Li$^+$ , Na$^+$ , K$^+$, Mg$^{2+}$, Ca$^{2+}$, Cl$^-$, and SO$_4^{2-}$ in aqueous solution based on the TIP4P/2005 water model and scaled charges for the ions

Simulating Monovalent and Divalent Ions in Aqueous Solution Using a Drude Polarizable Force Field.

Bottom-up Derived Flexible Water Model with Dipole and Quadrupole Moments for Coarse-Grained Molecular Simulations.

Completely Multipolar Model for Many-Body Water-Ion and Ion-Ion Interactions

Completely Multipolar Model as a General Framework for Many-Body Interactions as Illustrated for Water

Accounting for electronic polarization in nonpolarizable force fields

Coupling Molecular Density Functional Theory with Converged Selected Configuration Interaction Methods to Study Excited states in Aqueous Solution

Balancing Group I Monoatomic Ion-Polar Compound Interactions for Condensed Phase Simulation in the Polarizable Drude Force Field