Pyrolysis simulations of Fugu coal by large-scale ReaxFF molecular dynamics
Evidence and attribution¶
Authority of statements
This page summarizes the conference abstract PDF in pdf_path (2017 International Conference on Coal Science & Technology / Australia–China Symposium on Energy). It is not a full journal article; numerical claims track the abstract text.
Summary¶
Pyrolysis is the first chemical step in most coal conversion routes and strongly influences downstream processes. The abstract describes isothermal ReaxFF molecular dynamics on a multi-component Fugu sub-bituminous coal model with 23,898 atoms to study pyrolysis properties. Simulations run 2.0 ns from 1200–2200 K and 250 ps at 2400 K and 2600 K. Products are lumped by carbon number into char, heavy tar, light tar, and gas. The authors report that simulated product profiles vs temperature follow experimental trends, with total gas yield in fair agreement with literature values despite differing temperature windows. Time evolution analysis suggests high-temperature, short-duration runs can roughly reproduce distributions that would require much longer time at lower temperatures. VARxMD (a C++ visualization/analysis tool for reactive MD) parses detailed reaction pathways. Keywords listed: coal pyrolysis, multi-component structure model, product evolution, reaction mechanism, ReaxFF MD.
Methods¶
1 — MD application (atomistic dynamics). Engine / code: GPU-enabled ReaxFF MD in GMD-Reax (cited in the conference PDF). System: multi-component Fugu sub-bituminous coal model C\(_{11995}\)H\(_{10363}\)N\(_{159}\)O\(_{1366}\)S\(_{15}\) totaling 23,898 atoms in a cubic cell (indexed extract). Sampling protocol: isothermal ReaxFF MD for 2.0 ns each at 1200, 1400, 1600, 1800, 2000, and 2200 K, plus shorter 250 ps runs at 2400 K and 2600 K to probe the elevated-temperature acceleration strategy. Ensemble: NVT with periodic boundary conditions in all directions of the cubic box. Timestep: 0.25 fs in the indexed §2.2 text (the same paragraph’s “0.25 ps” unit is treated as a typo in the source—confirm against the PDF if ambiguous). Thermostat: Berendsen with 0.1 ps damping constant. Cutoffs: bond-order threshold 0.3 and nonbonded cutoff 10 Å as stated in §2.2. Post-processing: VARxMD for reaction mining and product classification. Barostat / pressure: N/A — NVT at unspecified volume; no NPT or stress control reported. Electric field: N/A — not used. Replica / enhanced sampling: N/A — not used (standard NVT reactive MD).
2 — Force-field training. N/A — the indexed pages describe application of ReaxFF to a constructed coal model, not a new QM refit in this file.
3 — Static QM / DFT-only. N/A — DFT is not the production engine for the pyrolysis trajectories summarized here.
Findings¶
Outcomes / mechanisms. Temperature-dependent evolution of char / heavy tar / light tar / gas lumps (carbon-number rules as in the authors’ prior work) shows radical-driven chemistry consistent with the ReaxFF coal literature cited in the introduction.
Comparisons. Normalized tar + gas trends vs fluidized-bed pyrolysis experiments (figure referenced in the extract) are described as agreeing on trend; total gas yield from simulation is reported in fair agreement with literature experiments despite different temperature windows.
Sensitivity / design levers. The authors compare long lower-T runs vs short higher-T runs to argue that high-T simulations can approximately reproduce product partitioning that would take much longer at lower temperature.
Limitations / outlook (as authored). Model construction details are deferred (“will be reported elsewhere” in the extract); quantitative agreement is framed cautiously given temperature-window mismatch.
Corpus / PDF honesty. Grounded in the conference PDF and normalized/extracts/2017fa-venue-acam-3_p1-2.txt; there is no DOI in front matter—treat as non-journal grey literature.
Limitations¶
Conference abstract only—no peer-reviewed DOI, and extraction may omit figures/tables from a longer manuscript. ReaxFF parameters for this coal model require verification against the group’s archived parameter tables. Reactor-scale extrapolation is not claimed.
Relevance to group¶
Illustrates VARxMD + large-system ReaxFF coal pyrolysis workflows represented in the corpus Chinese Coal Abstracts folder—useful as a pointer to methodology talks even without a journal DOI.
Citations and evidence anchors¶
- No DOI;
papers/ReaxFF_others/Chinese_Coal_Abstracts/Pyrolysis simulations of Fugu Coal by Largr-scale ReaxFF molecular dynamics.pdf; extractnormalized/extracts/2017fa-venue-acam-3_p1-2.txt.