Skip to content

Atomic insight into tribochemical wear mechanism of silicon at the Si/SiO2 interface in aqueous environment: Molecular dynamics simulations using ReaxFF reactive force field

Summary

ReaxFF MD of tribochemical wear at a Si / amorphous SiO\(_2\) contact in water, motivated by MEMS aqueous environments and Si CMP. The simulations identify two Si removal pathways: (i) rupture of stressed Si–O–Si bridges on the Si side assisted by H attachment to bridging oxygens, and (ii) breaking Si–Si bonds in Si–Si–O–Si chains at the interface. Higher normal pressure on the silica counter-body increases interfacial Si–O–Si bridge formation and can increase Si removal. Water both oxidizes Si and, via steric/solvation effects, can limit intimate contact.

Methods

This slug is a duplicate PDF (papers/Wen_App_Surf_Sci_2016.pdf) for the same Applied Surface Science article as [[2016wen-applied-surf-atomic-insight]] (DOI 10.1016/j.apsusc.2016.08.082). Methods match that page: LAMMPS ReaxFF, 4.26 × 4.26 × 8.0 nm\(^3\) PBC in-plane slab-on-slab stack with 300 H\(_2\)O, ensemble: NVT; thermostat: Nose–Hoover at 300 K; timestep: 0.25 fs; 250 ps compression equilibration; 1 ns shear at 10 m/s (0.1 Å/ps); and 2.0 / 4.0 / 6.0 GPa normal-load series as reported in §2 of the issue PDF.

2 — Force-field training. N/A — merged published ReaxFF libraries (Si/Ge/H + water) with validation in SI.

3 — Static QM. N/A — not used for the wear trajectories.

4 — Replica / enhanced sampling. N/A — not used.

Findings

Outcomes / mechanisms. The VOR page documents two Si removal channels: stress-biased Si–O–Si rupture on the Si side with H assistance, and Si–Si failure within Si–Si–O–Si chains tied to interfacial bridges (§3–4 of Appl. Surf. Sci. 390).

Comparisons. Removal counts vs applied normal stress show no Si loss at 2.0 GPa, 7 Si at 4.0 GPa, and 14 Si at 6.0 GPa under the 0.3 bond-order molecular recognition criterion quoted in [[2016wen-applied-surf-atomic-insight]].

Sensitivity / levers. Pressure tunes bridge-bond counts and friction traces (Figure 8), while interfacial water simultaneously oxidizes Si and can separate surfaces, giving the dual-role interpretation in §4.

Limitations / outlook. Same MD vs AFM velocity caveats as the primary slug; duplicate PDFs should not be treated as independent experiments.

Corpus honesty. Prefer [[2016wen-applied-surf-atomic-insight]] for long-form prose; this page records alternate SHA-256 provenance for the same DOI.

Limitations

  • Simulation scales and speeds may not capture all real CMP slurry chemistry (particles, additives) beyond Si/SiO\(_2\)/water motifs.
  • Wear rates require careful mapping from atomistic events to continuum removal rates.
  • MEMS contacts can involve mixed oxides, organic contaminants, and electrochemical potentials absent from the dry/wet mechanical models summarized here.
  • Load cycles and frequency in experiment can induce fatigue and third-body particle formation not captured in single-pass MD shear protocols.
  • Slurry pH and ionic strength in CMP can alter oxide termination beyond the pure-water motifs emphasized in the abstract-level summary.
  • Applied Surface Science Simulation sections list system sizes and timesteps needed to reproduce reported wear pathways.

Relevance to group

van Duin-group coauthored ReaxFF tribochemistry paper bridging silicon MEMS/CMP applications with reactive silica–water chemistry.

Citations and evidence anchors