Integrated sorption and diffusion model for bentonite. Part 2: porewater chemistry, sorption and diffusion modeling in compacted systems

Electrokinetics of the silica and aqueous electrolyte solution interface: Viscoelectric effects

The manipulation of biomolecules, fluid and ionic current in a new breed of integrated nanofluidic devices requires a quantitative understanding of electrokinetics at the silica/water interface. The conventional capacitor-based electrokinetic Electric Double Layer (EDL) models for this interface have some known shortcomings, as evidenced by a lack of consistency within the literature for the (i) equilibrium constants of surface silanol groups, (ii) Stern layer capacitance, (iii) zeta (ζ) potential measured by various electrokinetic methods, and (iv) surface conductivity. In this study, we consider how the experimentally observable viscoelectric effect - that is, the increase of the local viscosity due to the polarisation of polar solvents - affects electrokinetcs at the silica/water interface. Specifically we consider how a model that considers viscoelectric effects (the VE model) performs against two conventional electrokinetic models, namely the Gouy-Chapman (GC) and Basic Stern capacitance (BS) models, in predicting four fundamental electrokinetic phenomena: electrophoresis, electroosmosis, streaming current and streaming potential. It is found that at moderate to high salt concentrations (>5×10(-3)M) predictions from the VE model are in quantitative agreement with experimental electrokinetic measurements when the sole additional adjustable parameter, the viscoelectric coefficient, is set equal to a value given by a previous independent measurement. In contrast neither the GS nor BS models is able to reproduce all experimental data over the same concentration range using a single, robust set of parameters. Significantly, we also show that the streaming current and potential in the moderate to high surface charge range are insensitive to surface charge behaviour (including capacitances) when viscoelectric effects are considered, in difference to models that do not consider these effects. This strongly questions the validity of using pressure based electrokinetic experiments to measure surface charge characteristics within this experimentally relevant high pH and moderate to high salt concentration range. At low salt concentrations (<5×10(-3)M) we find that there is a lack of consistency in previously measured channel conductivities conducted under similar solution conditions (pH, salt concentration), preventing a conclusive assessment of any model suitability in this regime.