Molecular Dynamics Simulation of the Precipitation of Calcium Silicate Hydrate Nanostructures under Two-Dimensional Confinement by TiO2: Implications for Advanced Concretes
Reactive MD (ReaxFF) simulations follow precipitation of calcium silicate hydrate (C–S–H) gel inside TiO₂ nanochannels of varying spacing. TiO₂ surfaces with oxygen dangling bonds adsorb Ca and Si differently, driving nanoscale segregation into Ca-rich and Si-rich regions; surface hydroxylation promotes silicate polymerization, with up to ~15% Q₄ silicon environments in the most polymerized confined C–S–H.
Summary¶
Nanoengineered concrete benefits from understanding C–S–H–nanomaterial interactions. The study shows faster polymerization and higher final polymerization for C–S–H formed under TiO₂ nanoconfinement, nanosegregated chemistry at the hydroxylated interface, and more compact gel morphology inferred from water mobility patterns—linking confinement and surface chemistry to nanostructure.
Methods¶
1 — MD application. LAMMPS with the Kim et al. ReaxFF SiO + CHON merger described in Section 2; 0.25 fs time step. TiO₂ rutile (110) slit pores at D19, D38, D59 (~19.35, 37.93, 59.27 Å); constant Si and Ca content across spacings; PACKMOL initial pack; NVT 300 K relaxation, then accelerated aging at 1500 K for 1 ns (prior cited protocol) for C–S–H condensation; NVT at 300 K for MSD of H; Qₙ tracking. N/A — NPT (constant-volume NVT in protocol as stated). N/A — metadynamics / replica exchange. N/A — external electric field. Thermostat details in Section 2–3; N/A here if a label is not repeated in the wiki—confirm the PDF. PBC in-plane for the confined slab geometry as in the article. N/A — NPT pressure control: NVT-based protocol.
2 — Force-field training. N/A — the article uses Kim et al. / merged Reaxff; no in-manuscript full refit is claimed.
3 — Static QM. N/A — the paper is MD-centric with literature-based FF.
4 — Review or non-simulation. N/A — research article with reactive MD.
Findings¶
Outcomes and mechanisms. Confinement in D19/D38/D59 spacings speeds C–S–H polymerization relative to the comparison cases in the paper, with up to ~15% Q₄ in the most polymerized confined region. Hydroxylated TiO₂ drives Ca vs Si adsorption differences and nanosegregation; Q₃/Q₄ concentrate in Si-rich areas. Water MSD supports a more compact nanostructure in confined films.
Comparisons and sensitivity. Channel spacing and interface chemistry (hydroxylation) are the main levers. N/A — new experimental data in the excerpt for this short summary; the full article has literature comparison.
Authored limitations. Extrapolation to long cure times and high-pH electrolytes is limited—see ## Limitations on the page.
Corpus honesty. Full supercell counts and time windows are in the PDF / Section 2.
Limitations¶
ReaxFF uncertainty for high-pH cement chemistry and long curing times means laboratory hydration experiments remain necessary for engineering property prediction beyond the nanoscale condensation stages summarized here.
Relevance to group¶
Standard ReaxFF application space for oxide–silicate interfacial reactivity in construction materials.
Citations and evidence anchors¶
- DOI:
10.1021/acsanm.9b02203