Skip to content

Surface Reactivity and Leaching of a Sodium Silicate Glass under an Aqueous Environment: A ReaxFF Molecular Dynamics Study

Evidence and attribution

Authority of statements

Prose summarizes the J. Phys. Chem. C article identified by doi. The corpus PDF is a proof; verify pagination against the version of record.

Summary

Large-scale ReaxFF MD explores a sodium silicate glass–water interface, focusing on modifier-ion leaching, water dissociation, silanol formation, and proton transport among non-bridging oxygen (NBO) sites. The study emphasizes non-equilibrium evolution on nanosecond MD horizons and tracks Na⁺ release from surface-bound Na–OH-like configurations into solution. Alkali borosilicate and soda-lime glasses in geochemistry and engineering both show complex leaching layers; atomistic models aim to connect ion-exchange fluxes to local network hydrolysis without assuming sharp sharp-interface equilibrium (introduction themes).

Methods

  • Interactions model: ReaxFF reactive MD enabling dissociative water chemistry and Si–O–Na network rearrangement at the glass surface.
  • System class: Sodium silicate glass slab interfaced with liquid water (composition and dimensions per article §2–3).

  • Observables: Time evolution of Na⁺ near NBO sites, H⁺/OH⁻ formation, Si–OH counts, proton hopping between neighboring NBO sites, and Na mobility / leaching trajectories.

  • Trajectory diagnostics: authors monitor Na⁺ coordination shells, NBO protonation events, and inward H⁺ migration depth as function of time to show coupled ion release and acid ingress (methods narrative; abstract).

MD protocol (proof §Methods). System size & composition: sodium silicate glass slab plus water, ~15 000 atoms in the interfacial supercell (proof). PBC in three dimensions. Production leaching run quoted in the proof: 1 ns NVT MD at 300 K (triplicate water placements) after earlier equilibration in the same section. Engine / code: ReaxFF through the ReaxFF module in the Amsterdam Density Functional (ADF) package with the Na/Si/O/H parameter set referenced in the proof Methods. Integrator / timestep: Verlet updates with 0.25 fs timestep. Thermostat / barostat (build stages): Berendsen with 0.1 ps temperature coupling and 5.0 ps pressure coupling in the equilibration narrative; read the full NPT/NVT staging in the PDF before reproducing. Temperature: 300 K for the 1 ns interfacial NVT segment. Barostat in that production block: N/A — NVT is constant volume; NPT appears only in earlier relaxation steps. Duration: 1 ns of interfacial NVT sampling as stated. Pressure: N/A — for the NVT production run. Electric field: N/A — not used. Enhanced sampling: N/A — not used.

Findings

  • Water attack progressively removes Na⁺ coordinated to surface NBOs; molecular water dissociates to H⁺ and OH⁻ in the presence of modifier cations (abstract).
  • NBO protonation yields silanol; protons propagate into the glass via discrete hops between adjacent NBO sites (abstract).
  • Silanol population keeps growing over the simulated window—no steady state within the nanosecond trajectories (abstract).

  • Sodium first resides near the surface as Na–OH-related motifs with finite surface residence time, then releases to bulk water (abstract).

  • Non-steady-state leach: the continued silanol growth is interpreted as evidence that nanosecond windows capture early-stage gel-like alteration rather than a terminal stoichiometric dissolution steady state (discussion framing; abstract).

Comparisons, sensitivity, corpus honesty. The abstract frames reactive MD as complementary to QM cost limits and non-reactive glass potentials; quantitative agreement with experiment is developed in the Results. Sensitivity to pH, ion strength, and longer time horizons is not fully captured on nanosecond windows—see ## Limitations. This corpus file is an ACS proof PDF; cite the version-of-record for pagination and any editorial corrections.

Limitations

Proof PDF in corpus—verify pagination and any ACS editorial fixes against the version of record. Nanosecond trajectories may not reach thermodynamic equilibrium for leaching; system size and pH effects are idealized relative to experiment.

Ion strength and background electrolytes in natural waters are omitted; expect quantitative leach fluxes to shift when extending to seawater or brines.

Relevance to group

Glass–water ReaxFF from Hahn and van Duin—benchmark for modifier-ion leaching and hydroxylated silicate surfaces.

Citations and evidence anchors

  • DOI: 10.1021/acs.jpcc.9b02940