Paper Number
PO19 

Session
Poster Session

Title
Can we predict viscosity of electrolytes? Simple theories versus molecular dynamics simulations with LiTFSI in acetonitrile

Presentation Date and Time
October 23, 2019 (Wednesday) 6:30

Track / Room
Poster Session / Ballroom C on 4th floor

Authors (Click on name to view author profile)

  1. Wang, Yilin (University of Illinois at Urbana Champaign, Mechanical Science and Engineering)
  2. Farag, Hossam (University of Illinois at Urbana Champaign)
  3. Zhang, Yang (University of Illinois at Urbana Champaign)
  4. Ewoldt, Randy H. (University of Illinois at Urbana-Champaign, Department of Mechanical Science and Engineering)

Author and Affiliation Lines (in printed abstract book)
Yilin Wang1, Hossam Farag2, Yang Zhang2, and Randy H. Ewoldt1
1Department of Mechanical Science and Engineering, University of Illinois at Urbana-Champaign, Urbana, IL 61801; 2University of Illinois at Urbana Champaign, Urbana, IL

Speaker / Presenter 
Wang, Yilin

Text of Abstract
We collect experimental viscosity data for LiTFSI salt in acetonitrile solvent using a microfluidic channel flow viscometer (Rheosense, mVROC) over a range of concentrations (0 to 0.7 M) and temperatures (293 to 333 K). The experimental data is used to calibrate and test various methods of predicting the viscosity of this electrolyte system, motivated by the design and engineering of liquids for redox flow battery systems. We assess the trade-off between model complexity and model accuracy with a simple theory and molecular dynamics simulation. The simple theory is based on the Jones-Dole equation and Eyring's theory, a semi-predictive model that requires calibration of model parameters to some experimental viscosity data points. For comparison, molecular dynamic (MD) simulations were conducted to make a priori prediction of the viscosity, though the potential field model also requires calibration to some (non-viscosity) experimental data. Here the forcefield used is the optimized potentials for liquid simulations (OPLS) – all atom force field, modified based on our mp2 quantum chemistry calculations for atomic partial charge and geometry. The partial atomic charges for coulombic interactions are rescaled by a calibrated fraction (80%) to mimic the average charge screening due to polarization and charge transfer effects. In order to predict zero-shear viscosity from atomic trajectories, we use the Green-Kubo relations, integrating the time auto correlation function of the off-diagonal stress tensor elements. Predictions from both models are compared to the experimental data which shows the relative trade-off between model complexity and model accuracy in predicting transport properties in these small molecule complex fluids.