Dynamics in Clays - Combining Neutron Scattering and Microscopic Simulation

Abstract

Mobility of ions and water in clays is at the heart of their remarkable properties of water retention and ion-exchange. It has been addressed here using two microscopic techniques: neutron scattering and molecular dynamics simulations. Neutron scattering gives access exclusively to water dynamics in clays, due to the exceptional sensitivity of neutrons to H atoms. The data interpretation can be challenging, especially for natural clays such as montmorillonite, with inhomogeneous swelling characteristics. A great improvement is achieved with the use of synthetic materials, as demonstrated here on the case of synthetic (fluoro)hectorite. The standard analytical models for long-range diffusive motion, isotropic translation and its derivative, powder averaged two dimensional translation, have been used to interpret the neutron scattering data. They both agree on the order of magnitude for the diffusion coefficient of water in monohydrated and bihydrated clays, 10⁻¹⁰ m²s⁻¹ and 10⁻⁹ m²s⁻¹ respectively. While the two-dimensional nature of water diffusion in clays is seen clearly from molecular dynamics simulations, its signature in neutron scattering data is obscured by the powder-averaging of the signal. A novel method, based on a multi-resolution analysis of scattering functions from powder samples, allows never-the-less a clear determination of the dimensionality of water motion in the system. Extracting information on local water motion is difficult on the basis of neutron scattering data only. Various models for localised motion, rotation on a sphere or jump diffusion, have been proposed and used to interpret the observed neutron data, however their applicability is questionable in light of information from molecular dynamics simulations. Aside from aiding the interpretation of neutron scattering data, MD simulations are most valuable in providing information on the behaviour of ions in clays. MD estimates the interlayer ion coefficients as of the some order of magnitude as water, even if the details of ionic motion are strikingly different between the two ions considered here, Na⁺ and Cs⁺. Further, MD has also allowed to address the topic of ion exchange between clay interlayers and bulk aqueous solution. The microscopic picture of water and ion motion in clays, emerging from neutron scattering and MD simulations, should be treated as a building block of the overall modelling of macroscopic transport in clays, the ultimate property of interest for many clay applications.