CHARGE DENSITY IN NON-ISOTROPIC ELECTROLYTES CONDUCTING CURRENT Glyn F. Kennell* and Richard W. Evitts Department of Chemical and Biological Engineering, University of Saskatchewan, 57 Campus Drive, Saskatoon, SK Canada S7N 5A9 This article presents applications of an electrochemical model that can predict concentrations and electric current distributions assuming neither electroneutrality nor negligible concentration gradients. This numerical model is used to analyse two cases: a liquid-junction and a lithium-ion cell. For both cases, it is shown how the inclusion of charge density effects on the electric field is beneficial. For the case of a liquid-junction the predicted potential gradient is compared with values experimentally and numerically determined by other researchers. For the case of a charging lithium-ion cell, a phenomenon seen experimentally (but not previously reported by other models) is predicted. Keywords: charge density, electrolytic transport, lithium-ion cells, liquid-junction, electric field, numerical model, electrochemical phenomena INTRODUCTION A theory commonly used as a foundation for electrochem- ical models is dilute solution theory (Newman and Thomas-Alyea, 2004). Dilute solution theory assumes bulk electroneutrality and often further assumptions, such as uniform electrolyte concentration, are required to model multi- dimensional current distributions and mass transfer. Instead, the model presented in this article uses some aspects of dilute solu- tion theory but does not make assumptions of electroneutrality nor uniform electrolyte concentrations. Predictions made by this model demonstrate the benefits of not making these assumptions. Simulations presented in this article show how the inclusion of the effects of charge density allows for the simulation of a liquid- junction using a single transport property for each dissolved species: the diffusion coefficient. Simulations also demonstrate how the absence of these assumptions allow for the modelling of phenomenon experimentally observed in lithium-ion cells but not previously successfully modelled with dilute solution theory. West et al. (1982) presented a model accounting for the cou- pled transport in the electrolyte and electrode phases of a cell with porous insertion electrodes and a liquid electrolyte. The system was modelled one-dimensionally using infinitely dilute solution theory. Some assumptions made included: electroneutral- ity, very high conductivity in the electrode phase, a mono-valent electrolyte salt and negligible flow due to concentration gradients or solid matrix expansion. Simulations demonstrated how elec- trolyte depletion was the principal factor limiting the discharge capacity of the system. Doyle et al. (1993) presented a model for a lithium-ion cell based on concentrated solution theory that was implemented considering one-dimensional transport for a gal- vanostatic current. The cell was comprised of a lithium foil anode, a polymer electrolyte and composite cathode. The model was used to simulate concentration gradients across the system. It was found that the decreased lithium concentration in the composite cathode (below one molar) illustrated the need for higher lithium concentrations to maintain adequate conductivity. This model was expanded by Fuller et al. (1994) who considered a porous insertion anode instead of a lithium foil anode. This composite electrode consisted of an inert conducting material, the electrolyte and the solid active insertion particles. Transport was considered one-dimensionally through the solid electrode via diffusion and through the electrolyte via concentrated solution theory. Arora et al. (1999) used the macroscopic model of Fuller et al. (1994) for one-dimensional lithium-ion battery predictions. The influence of lithium deposition on the charging and overcharging of inter- calation electrodes was examined. It was observed that lithium deposition was predicted for cells with lower excess negative electrode capacity and no deposition was predicted for cells with higher excess negative capacity. It can be noted that models based ∗ Author to whom correspondence may be addressed. E-mail address: glyn.kennell@usask.ca Can. J. Chem. Eng. 90:377–384, 2012 © 2011 Canadian Society for Chemical Engineering DOI 10.1002/cjce.20641 Published online 31 August 2011 in Wiley Online Library (wileyonlinelibrary.com). | VOLUME 90, APRIL 2012 | | THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING | 377 |