ReaxFF Molecular Dynamics Simulations of Hydroxylation Kinetics for Amorphous and Nano-Silica Structure, and Its Relations with Atomic Strain Energy
Summary¶
Reactive molecular dynamics with ReaxFF is used to follow water–silica hydrolysis on nanostructured and amorphous SiO₂, with a ReaxFF parameter refinement starting from the Fogarty et al. (2010) SiO₂–water description so that hydroxylation barriers for strained versus unstrained Si–O motifs better match DFT reference data. The work ties silanol formation to local strain and maps multiple hydroxylation pathways, including routes involving H₃O⁺, with implications for where water-derived chemistry localizes on silica under mechanical stress.
Methods¶
2 — Force-field training (ReaxFF SiO-2015). Parent: Fogarty et al. SiO/water interaction set. QM reference: B3LYP/6-311++G(d,p) barrier scans for strained vs unstrained Si–O–Si–O ring dimers feeding additional training points; Si–O bond/off-diagonal terms, Si–O–Si, O–Si–O, O–Si–Si, and O–H–O hydrogen-bond parameters are reoptimized to match the DFT curves (Figure 3). Post-fit barriers: ~31 kcal/mol (unstrained dimer) and ~20 kcal/mol (strained dimer) without ZPE corrections (ZPE < 1 kcal/mol per the article).
1 — MD application (ReaxFF). Engine / code: LAMMPS molecular dynamics with the SiO-2015 ReaxFF parameter file described above. Nanowire hydration test: 108 atoms (36 Si + 72 O) nonperiodic nanowire in a 30 × 30 × 30 Å\(^3\) cell with 200 H\(_2\)O (~0.22 kg/dm\(^3\) water density), 800 ps NVT segments at each 100 K step from 300 K to 1500 K using Berendsen thermostat (100 fs damping) and 0.1 fs timestep (§2.2). Additional setups (double-slit amorphous silica, strained surface dimers) extend the same SiO-2015 model to compare pathways vs temperature (§2.3–2.4). Barostat: N/A — NVT water-vapor boxes. Pressure control: N/A — not NPT; cells are held at the vapor density stated for each tutorial box. Electric field: N/A — not used. Replica / enhanced sampling: N/A — not used.
3 — Static QM. Covered under the training block (B3LYP dimer scans).
4 — Reviews. N/A — primary research article.
Findings¶
Barrier agreement. After SiO-2015 reoptimization, ReaxFF reproduces the DFT hydroxylation barrier ordering for strained vs unstrained Si–O dimers with the ~20 / ~31 kcal/mol targets quoted post-fit (Figure 3).
Strain localization. Nanowire trajectories show silanol genesis preferentially at high-strain edge sites (Figures 4–5), linking atomic strain energy maps to reaction propensity.
Pathways and temperature. The article documents multiple H\(_3\)O\(^+\)-mediated sequences (Figure 6 at 400 K) plus direct water dissociation channels that become more frequent at >1000 K (Figure 8 narrative), and summarizes how the H-bond network reorganizes across the 300–1500 K sampling ladder.
Amorphous silica. Double-slit a-SiO\(_2\) runs mirror the nanowire lesson: silanols concentrate where strain is elevated, with similar hydronium involvement (§2.4–2.5).
Tribology link. The Discussion ties strain-biased hydroxylation to where water-derived films may anchor on stressed silica contacts—read that section for caveats on transferability to engineering interfaces.
Limitations¶
Quantitative barrier values, additional pathway figures, and any SI-only benchmarks should be taken from the PDF at pdf_path (the indexed extract here is first pages only).
Relevance to group¶
Penn State / van Duin-group ReaxFF development and silica–water chemistry relevant to tribology and oxide interfaces.
Citations and evidence anchors¶
- DOI:
10.1021/acs.jpcc.5b09784(J. Phys. Chem. C).