Skip to content

Numerical simulations of yield-based sooting tendencies of aromatic fuels using ReaxFF molecular dynamics

Evidence and attribution

Authority of statements

Prose sections below (Summary, Methods, Findings, etc.) are curated summaries of the publication identified by doi, title, and pdf_path in the front matter above. They are not new primary claims by this wiki.

For definitive numerical values, reaction schemes, and interpretations, use the peer-reviewed article (and optional records under normalized/papers/ when present)—not this page alone.

Summary

The work presents a ReaxFF molecular dynamics workflow to compute the experimental Yield Sooting Index (YSI) for fuels, using a multi-stage procedure designed to mirror how YSI is obtained experimentally and in continuum models. Toluene and phenol are used as proof-of-concept aromatics with relatively well-characterized chemistry. Simulations capture key growth pathways expected from kinetics (toluene retaining and growing aromatic rings; phenol involving carbon-loss pathways with CO release). A quantitative YSI construction from ReaxFF output is compared to measurements with reasonable agreement, arguing that the approach can rank sooting tendency when detailed kinetic mechanisms are unknown. The introduction contrasts kinetic-model YSI workflows (accurate but mechanism-dependent) with group-additivity estimators (fast but weak when carbon types are underrepresented), positioning ReaxFF as a middle path for unknown fuel chemistry.

Methods

This slug uses the text-layer Fuel file papers/Kwon_Fuel_2019_online.pdf. Methods and citable numerical settings are in Fuel Sec. 2–4; the brief below paraphrases the VOR (DOI in front matter)**.

1 — MD application (ReaxFF, multi-stage NVT)

ReaxFF with the CHO-2016 field ([39], Fuel Sec. 2). System (stage 1): 3D PBC cubic ~38.6 Å with ~2300+ atoms (CH₄/O₂, ρ=0.28 kg/dm³, equivalence ratio 3.5, Fuel Sec. 2). NVT stages (temperatures): 1500 K (25 ps); 3000 K (Berendsen, 100 fs damping) until max radical count; then doped pools ~2308 or ~2224 atoms (ρ ~0.39 kg/dm³) with 42 benzyl or phenoxy; final NVT at 2200 K, 2400 K, 2600 K (five replicas). Production durations (Sec. 4.1): e.g. ~6 ns @ 2200 K, ~3.5 ns @ 2400 K, ~2 ns @ 2600 K (plateau criterion in Sec. 4.1). Time step (fs): N/A in the PDF text extracted here; use the VOR Methods (or SI if any) for reproduction-grade integration settings. MD program name: N/A in our string search of the local PDF (no explicit “LAMMPS” in that extract). Barostat / NPT (mean hydrostatic stress): N/A for the NVT-staged YSI RMD (Sec. 2). Electric field / replica / metadynamics: N/A.

2 — Force-field training

N/A—the paper uses CHO-2016 ([39]; improvement on CHO-2008 per Sec. 2 narration) and does not re-fit a new field in this article.

3 — Static QM / DFT

N/A—atomistic chemistry is ReaxFF-only; prior kinetic literature is cited for phenol / aromatic mechanism context (Sec. 1 & 4)**.

Findings

Mechanisms

Toluene pathways emphasize aromatic growth while retaining rings; phenol shows aromatic growth with CO-forming carbon-loss channels—aligned with prior kinetic understanding quoted in the paper.

Quantitative outcome

ReaxFF YSI values match experimental YSI reasonably for the proof-of-concept set—the article positions this as a first quantitative YSI framework from ReaxFF MD.

Future directions (from abstract)

Extension toward fuels with unknown or poorly known detailed kinetics (e.g. complex transportation or bio-derived streams) is the stated motivation.

Sensitivity and cross-checks

Relative sooting ranking depends on how trajectory cuts map to the YSI scalar; the paper compares ReaxFF YSI to tabulated experimental YSI for toluene and phenol (agreement described as reasonable in the abstract). Temperature and stage duration are levers in high-temperature RMD; numeric values belong in the Methods tables of the PDF, not invented here.

Corpus note

Use this online PDF slug for reproducibility; proof PDF sibling 2019kwon-venue-paper may differ in pagination (version-of-record integrity).

Limitations

Demonstration is limited to two aromatic compounds; extension to broader fuel spaces, blend effects, and quantitative error bars requires further validation. YSI mapping from atomistic trajectories introduces modeling choices that may not transfer unchanged to all fuel classes.

Relevance to group

Co-authored from Penn State mechanical engineering; extends group capabilities in ReaxFF-based combustion and soot precursor chemistry relevant to fuels and emissions modeling.

Citations and evidence anchors

Primary locator: papers/Kwon_Fuel_2019_online.pdf — abstract and methods for YSI protocol and toluene/phenol results. DOI: https://doi.org/10.1016/j.fuel.2019.116545