Disentangling cation effects on ion mobility and structure in ionic liquid electrolytes

Ionic liquids (ILs) are low-temperature molten salts, and therefore the transport of ions within ILs is dominated by ion-ion interactions. However, the influence of organic IL cations on key electrolyte properties, such as ion dissociation and overall transport behavior, in lithium-salt-doped ILs remains poorly understood. Moreover, despite their critical role in designing IL-based electrolytes for energy storage applications, ion-ion interactions and ion-specific transport under an applied electrical potential are seldom quantified, largely due to the unique experimental and computational challenges involved. Herein, we compare transport properties obtained using 1H, 7Li, and 19F pulsed-field gradient nuclear magnetic resonance (PFG NMR) and electrophoretic NMR (eNMR) with those measured by electrochemical impedance spectroscopy (EIS). Non-equilibrium molecular dynamics (MD) simulations and eNMR confirm the presence of negatively charged [Li(TFSI)n](1-n) aggregates that migrate towards the positive electrode, resulting in negative lithium transference numbers. Equilibrium MD simulations reveal a vehicular Li ion transport mechanism facilitated by long-lived aggregates with Li+ cations strongly bound to multiple TFSI– anions. Finally, we observe an inverse relationship between the apparent charge of the TFSI– anion in the neat IL, which is dictated by the IL cation, and Li+ transport in the salt-doped systems. This highlights the opportunity to tune electrolyte performance by tailoring cation chemistry.

Methods
PFG NMR:

All NMR measurements were performed on a Bruker Avance III 300 spectrometer with a 7.05 T super-wide bore (150 mm) superconducting magnet equipped with a Bruker Diff50 probe (maximum gradient strength of 28.98 T m–1), operating at 300.15 MHz for 1H, 116.65 MHz for 7Li, and 282.40 MHz for 19F. All measurements were performed at 313.15 K following at least 30 minutes of equilibration with the sample temperature controlled using dry N2 at a flow rate of 800 L h–1. 313.15 K was chosen as an application-relevant temperature that would allow for faster ionic mobility through a decreased viscosity and, therefore, more accurate measurements. Self-diffusion coefficients of the IL (C4C1im, C4C1pyrr, and C4C1pip) and Li+ cations, and TFSI– anion, were obtained from 1H, 7Li, and 19F PFG NMR experiments, respectively. Diffusivities were determined using a variable gradient strength stimulated echo pulse sequence designed to minimize signal loss from short transverse relaxation times (T2). The self-diffusion coefficient of species i was determined by fitting the observed signal intensity as a function of the variable gradient strength using the Stejskal−Tanner equation.

eNMR:

eNMR measurements were performed using a double stimulated echo convection-compensating

pulse sequence, in which the polarity of the DC potential pulse was reversed halfway through the pulse sequence. Applied potentials and gradient strengths were varied from 0 to 90 V, and from 0.2 to 3 T m–1, respectively.

Equilibrium MD simulations:

Equilibrium atomistic molecular dynamics (MD) simulations were performed for C4C1im, C4C1pyrr, and C4C1pip ionic liquids at 313 K, each containing LiTFSI at a molar fraction of 0.1 (defined as the ratio of the number of Li+ ions to the total number of cations). Bonded (bond, angle, dihedral) and nonbonded interactions were described using the CL&P force field for C4C1im and C4C1pyrr, and the OPLS-AA force field for C4C1pip.

Full methods can be found in the Supplementary Information of the published paper.

Non-equilibrium MD simulations:

Non-equilibrium molecular dynamics (NEMD) simulations were performed to evaluate ionic drift

and transport under an applied external electric field. External electric fields ranging from 0.02 to 0.10 V nm–1 were applied along a fixed direction, with increments of 0.02 V nm–1. All parameters, including force fields, cutoffs, and thermostat/barostat settings, were identical to those used in the equilibrium simulations.