Table of Contents
Fetching ...

Molecular Dynamics Simulations of $γ$-Belite(010)-Water Interfaces with High-Dimensional Neural Network Potentials

Bernadeta Prus, Jörg Behler

Abstract

Belite -- dicalcium silicate Ca$_2$SiO$_4$ -- is a main constituent of low-carbon cement. In this work, we study several terminations of the (010) surface of $γ$-belite, its most stable polymorph, by molecular dynamics simulations. The energies and forces are provided by a high-dimensional neural network potential trained to density functional theory data. Water can interact in molecular form as well as dissociatively with the investigated interfaces, and the degree of dissociation is determined primarily by the protonation of SiO$_4$ groups accessible at the surface. A major part of the simultaneously formed hydroxide ions is adsorbed at surface calcium atoms, whose octahedral coordination spheres are completed by additional water molecules. The T3 termination, which is most stable in vacuum, shows only little reactivity in water. For the only slightly less stable T2 termination, however, two distinct types of surface defects are observed. The type I defect is even stable in vacuum and leads to a reconstruction of the entire surface, while the type II defect is only found in the presence of water. Overall, our results suggest that a variety of structures may be formed at the Ca$_2$SiO$_4$(010) surface, which are stabilized in the presence of water.

Molecular Dynamics Simulations of $γ$-Belite(010)-Water Interfaces with High-Dimensional Neural Network Potentials

Abstract

Belite -- dicalcium silicate CaSiO -- is a main constituent of low-carbon cement. In this work, we study several terminations of the (010) surface of -belite, its most stable polymorph, by molecular dynamics simulations. The energies and forces are provided by a high-dimensional neural network potential trained to density functional theory data. Water can interact in molecular form as well as dissociatively with the investigated interfaces, and the degree of dissociation is determined primarily by the protonation of SiO groups accessible at the surface. A major part of the simultaneously formed hydroxide ions is adsorbed at surface calcium atoms, whose octahedral coordination spheres are completed by additional water molecules. The T3 termination, which is most stable in vacuum, shows only little reactivity in water. For the only slightly less stable T2 termination, however, two distinct types of surface defects are observed. The type I defect is even stable in vacuum and leads to a reconstruction of the entire surface, while the type II defect is only found in the presence of water. Overall, our results suggest that a variety of structures may be formed at the CaSiO(010) surface, which are stabilized in the presence of water.

Paper Structure

This paper contains 24 sections, 2 equations, 23 figures, 5 tables.

Figures (23)

  • Figure 1: DFT-optimized orthorhombic unit cell of $\gamma$-belite containing four formula units of Ca$_2$SiO$_4$. Atoms are represented as spheres in colors corresponding to the elements: Ca - green, Si - beige, O - red.
  • Figure 2: DFT-optimized (2$\times$2) supercells of the (010) surfaces of $\gamma$-belite with different terminations. The corresponding cleavage energies are given in Table \ref{['table:cleavage_energy']}. The first row presents top views, the second row presents side views, alternating between the [100] and [001] directions, while the last row shows bottom views. The top and bottom surfaces are identical for the non-polar T2 and T3 terminations, while they are different for the polar T1 and T4 terminated slabs.
  • Figure 3: Density profiles of different oxygen species (O*H$^-$ - adsorbed hydroxide ions, H$_2$O* - adsorbed water molecules, OH$^{-\dagger}$ non-adsorbed hydroxide ions, H$_2$O$^{\dagger}$ non-adsorbed water molecules, O$_\text{s}$H$^-$ - protonated surface oxygen atoms, O$_\text{s}^{2-}$ - non-protonated surface oxygen atoms) and calcium atoms along the $z$ ([010]) direction for the T2 and T3 terminations. $z=0$, indicated by a vertical grey line, corresponds to the average position of the outermost layer of silicon atoms in the MD simulations.
  • Figure 4: Top views of the relative probability distributions of the calcium atom positions during $NVT$ MD trajectories at 300 K. The red circles represent the atomic positions at surfaces optimized in vacuum, the (4$\times$4) simulation cell is shown as a red dotted box. Panel (a) shows the first layer calcium atom distribution for T2, (b) the first layer of calcium atoms for T3, and (c) the second layer of calcium atoms for T3. The yellow and blue ellipses highlight two Ca atom displacements at the T2 termination.
  • Figure 5: Lateral probability distributions of the oxygen atoms in the adsorbed hydroxide ions for the T2 (a,b) and T3 (c,d) terminations. Panels (a) and (c) show the first and panels (b) and (d) the second layers of oxygen atoms (cf. blue peaks in Fig. \ref{['fig:density_profiles']}). The positions of the surface atoms at the start of the $NVT$ simulations at 300 K are shown in grey, and the (4$\times$4) supercells used in the simulations are shown as red dotted boxes. The yellow and blue ellipses highlight two Ca atom displacements of type I and type II, respectively, observed for T2.
  • ...and 18 more figures