Grand canonical simulations of ions between charged conducting surfaces using exact 3D Ewald summations.

We present a useful methodology to simulate ionic fluids confined by two charged and perfectly conducting surfaces. Electrostatic interactions are treated using a modified 3D Ewald sum, which accounts for all image charges across the conductors, as well as the 2D periodicity, parallel to the surfaces. The energy expression is exact, and the method is trivial to implement in existing Ewald codes. We furthermore invoke a grand canonical scheme that utilizes a bias potential, that regulates the surface charge density. The applied bias potential also enables us to calculate individual chemical potentials of the ions. Finally, we argue that our approach leads to a pedagogically appealing description of the Donnan potential, and what it measures in these systems.

Aggregation of Calcium Silicate Hydrate Nanoplatelets.

We study the aggregation of calcium silicate hydrate nanoplatelets on a surface by means of Monte Carlo and molecular dynamics simulations at thermodynamic equilibrium. Calcium silicate hydrate (C-S-H) is the main component formed in cement and is responsible for the strength of the material. The hydrate is formed in early cement paste and grows to form platelets on the nanoscale, which aggregate either on dissolving cement particles or on auxiliary particles. The general result is that the experimentally observed variations in these dynamic processes generically called growth can be rationalized from interaction free energies, that is, from pure thermodynamic arguments. We further show that the surface charge density of the particles determines the aggregate structures formed by C-S-H and thus their growth modes.

The growth of charged platelets.

Growth models of charged nanoplatelets are investigated with Monte Carlo simulations and simple theory. In a first model, 2-dimensional simulations in the canonical ensemble are used to demonstrate that the growth of a single weakly charged platelet could be limited by its own internal repulsion. The short range attractive interaction in the crystal is modeled with a square well potential while the electrostatic interactions are described with a screened Coulomb potential. The qualitative behavior of this case can also be described by simply balancing the attractive crystal energy with the screened Coulomb repulsion between the crystal sites. This repulsion is a free energy term dominated by counterion entropy and of course reduced by added salt. For a strongly coupled system, that is with high charge density and divalent counterions as in calcium silicate hydrate, the main product of cement hydration, the screened Coulomb approximation becomes inadequate and the growth behavior has to be described with the full primitive model. In this case, the energetic interactions become relatively more important and the entropy of the system plays a minor role. As a consequence, the electrostatic interactions gradually become less of a hindrance for aggregation and in extreme cases electrostatics actually promote the growth. This is manifested as an increased aggregation with, for example, increasing surface charge density. In the presence of divalent calcium ions and at the high negative surface charge density typical for calcium silicate hydrate, electrostatic interactions are not a hindrance for an infinite growth of the particles. By combining experimental and simulated data we can show that the limited sized platelets found in cement paste is due to a very fast nucleation rate compared to the growth rate.