Skip to content

Timescale prediction of complex multi-barrier pathways using flux sampling molecular dynamics and 1D kinetic integration: Application to cellulose dehydration

Second corpus PDF for the same J. Chem. Phys. article (DOI 10.1063/1.5126391) as 2020valdenaire-j-chem-phys-timescale-prediction, registered under the alternate on-disk filename spelling Valdennaire. Scientific content matches the primary wiki page; this note exists so manifest and hash provenance stay explicit.

Summary

The publication develops flux sampling along a one-dimensional reaction coordinate and one-dimensional kinetic integration to extrapolate rare-event kinetics for ReaxFF reactive MD of crystalline cellulose dehydration and decomposition at 1500–1900 K, comparing volatile product distributions to replica-exchange benchmarks and discussing Arrhenius parameters and breakdown when the order parameter is inadequate at lower temperature. This wiki duplicate-PDF slug is for provenance only: methods, findings, and limitations are curated on 2020valdenaire-j-chem-phys-timescale-prediction.

Methods

1 — MD and rare-event protocol. As on the VOR page: LAMMPS; 2013 C–C Reaxff; I-β cellulose supercell with PBC; atom count and lattice vectors in the VOR article; NPT ~2.5 GPa; 0.1 fs time step; flux sampling with 10 ps MD segments; temperatures 1500–1900 K in 100 K steps. N/A — re-deriving the full method narrative here: use [[2020valdenaire-j-chem-phys-timescale-prediction]] and its PDF for pagination and thermostat/barostat labels.

2 — Force-field training. N/A — not a separate refit; same published Reaxff as the article.

3 — Static QM. N/A.

4 — Review. N/A

Findings

Outcomes. The flux-sampling protocol reaches longer effective times than brute-force MD for the pathway class under study. Main products (H₂O, CO, CO₂) line up with replica-exchange data where decomposition completes. An Arrhenius form gives Eₐ ≈ 93 kcal mol⁻¹ and k₀ ≈ 9 × 10¹⁹ s⁻¹; at low T the kinetics can diverge from a single line if the order parameter is poor—as developed on the primary page. N/A for new science via this file path alone.

Comparisons, sensitivity, limitations, corpus honesty. Identical in substance to 2020valdenaire-j-chem-phys-timescale-prediction; this entry documents byte-level duplicate storage and spelling in pdf_path.

Limitations

Duplicate PDFs differ only by filename in the corpus; replacing one file without the other can cause hash and manifest drift—keep one path canonical for operational sync. Use the VOR slug for the canonical protocol narrative and ## Limitations of the work itself.

Relevance to group

Adri C. T. van Duin co-authorship; same contribution record as the primary slug.

Citations and evidence anchors

Reader notes (navigation)